Hi,
Complete newbie here trying to resurrect a CPC 6128.
It seems to boot to a white screen with a black border. Digging through previous posts here, some seem to think that this means the Z80 is intact?
Suspecting the RAM, I did the "RAM bank swap trick" by removing the IC118 and replacing it with some jumper wires in the appropriate positions (triple-checked!). This made no difference - still the same white screen with black border.
At this point I'm thinking I could:
1. Replace the Z80 for around £10
2. Replace the 40010 for £35
3. I would also be tempted to replace the RAM banks, but with my soldering skills I'm concerned that I could cause more harm than good. (So that might be a last resort...)
Anyway, before I try any of those options I thought I'd see what you experts think. Help? :)
Thanks
M
CRTC working
ROM and Z80 sometimes working.
Press DEL. Do You hear sound?
40010 Pin 32 conect to Pin1 expansion port.(sound) Do You hear sound 300Hz? if no GA not work.
74LS145 pin 15 conect to Pin1 expansion port.(sound) Do You hear sound 50Hz? if no interrupts not work.
Quote from: McArti0 on 23:09, 31 March 24CRTC working
ROM and Z80 sometimes working.
Press DEL. Do You hear sound?
40010 Pin 32 conect to Pin1 expansion port.(sound) Do You hear sound 300Hz? if no GA not work.
74LS145 pin 15 conect to Pin1 expansion port.(sound) Do You hear sound 50Hz? if no interrupts not work.
I knew I'd forget to detail everything. No sound on pressing DEL.
Thanks for the diagnostics suggestions.
I got the 300Hz tone nice and clear, so the 40010 seems to be good.
With the 50Hz tone, it did make a sound but I needed to keep moving the connecting wire over the expansion port pin for the sound to continue. And it sounded more like static. Not sure if that's a pass or a fail!
Check this:
https://www.cpcwiki.eu/forum/general-discussion/cpc464-no-video/msg237132/#msg237132
Quote from: mmldrm on 08:57, 01 April 24Not sure if that's a pass or a fail!
Fail. 50Hz must be strong.
Quote from: McArti0 on 09:08, 01 April 24Check this:
https://www.cpcwiki.eu/forum/general-discussion/cpc464-no-video/msg237132/#msg237132
Thanks. Although I must admit that table doesn't mean much to me. :( Are those some other locations that I can use with the expansion pin to test for a signal/sound? I'll happily give those a try if that is correct.
Quote from: McArti0 on 09:08, 01 April 24Fail. 50Hz must be strong.
Damn. I wonder why moving the wire caused it to make the sound.
Interrupts not working - would that be a faulty 40010 chip?
(Thanks for your help!)
The black border and white/grey screen show, that the CPC started to boot and initialised the CRTC but as soon as the CPU tries to access RAM it stops. Therefore you can't hear a beep as that requires that everything works.
However due to that screen we can also expect that ROM and Z80 are almost certainly okay. Also CRTC and GateArray are most likely working.
But this error is a typical result from a RAM fault. Probably one of the RAM ICs or one of the ICs that isolate the RAM from the bus (the mixers and latches) is faulty.
If you have a Lower ROM board, Dandanator or M4 you can try to run Amstrad diagnostics. If all RAM ICs seem to be dead, it's probably a mixer IC and if it's single ICS then the respective RAM IC or the latch is the culprit.
you could also try to remove the PAL IC and insert a socket where you bridge a few pins to force the 6128 to use the secondary RAM bank as it's primary bank. If it's just a malfunctioning RAM IC (with no permanent short) this can give you a working 6128 - with just 64KB of RAM of course. but then you know for sure that it's the RAM.
https://www.youtube.com/watch?v=LLVHJO-aj3Q
of course, this will only work, if it's only an IC of the primary RAM bank that is failing and nothing else.
Quote from: mmldrm on 10:45, 01 April 24Although I must admit that table doesn't mean much to me
Do You have a multimeter?
No. Pin32 and 300Hz sound showing that GA 40010 work.
Z80 not run properly code in RAM.
ps. check sound on pin7 and pin9 all 74LS153. Another but obvious sound on all outputs. muliplexers are good
Quote from: eto on 11:22, 01 April 24you could also try to remove the PAL IC and insert a socket where you bridge a few pins to force the 6128 to use the secondary RAM bank as it's primary bank. If it's just a malfunctioning RAM IC (with no permanent short) this can give you a working 6128 - with just 64KB of RAM of course. but then you know for sure that it's the RAM.
https://www.youtube.com/watch?v=LLVHJO-aj3Q
of course, this will only work, if it's only an IC of the primary RAM bank that is failing and nothing else.
Thanks. I tried that with a few jumper wires I had, but got the same white screen with black border. Was hoping that would narrow it down to a single faulty RAM bank, but sadly not.
Quote from: McArti0 on 11:45, 01 April 24Do You have a multimeter?
I do, but to be honest I've used it once about 3-4 years ago and never since. Would have to find YouTube instructions first, to set it up correctly.
But, in short, yes I do. :)
Edit: this is the multimeter I have, in case it helps to suggest tests.
Quote from: McArti0 on 11:52, 01 April 24ps. check sound on pin7 and pin9 all 74LS153. Another but obvious sound on all outputs. muliplexers are good
This was interesting. Should the output noise be noticeably different across the pins? All were loud, if varied, except pin 7 on IC105 (the bottom one of the four).
(I'll try to get the multimeter set up today so I can see actual numbers as outputs. Forgot I had it, tbh.)
20V bottom-left is yours and now https://www.cpcwiki.eu/forum/general-discussion/cpc464-no-video/msg237132/#msg237132 (https://www.cpcwiki.eu/forum/general-discussion/cpc464-no-video/msg237132/#msg237132)
probes hole mid and right.
Quote from: mmldrm on 13:07, 01 April 24Should the output noise be noticeably different across the pins? All were loud, if varied, except pin 7 on IC105
All differents like bad piano :D
What about this Pin7 of IC105? died?
Quote from: McArti0 on 15:11, 01 April 2420V bottom-left is yours and now https://www.cpcwiki.eu/forum/general-discussion/cpc464-no-video/msg237132/#msg237132 (https://www.cpcwiki.eu/forum/general-discussion/cpc464-no-video/msg237132/#msg237132)
probes hole mid and right.
Sorry, I see the table of voltages on the GA/Z80 chips but I'm not sure what you are suggesting to try. :(
Not sure what you mean by 20V (literally 20 volts?) or the probes line. (My lack of technical skills causing trouble here I think!)
Happy to test anything with the multimeter, but imagine you are instructing a guy who sees it as two pointy sticks. (I don't want to fry a chip by pushing current across the wrong pins.)
imagine
Thanks. Where is best to ground on the motherboard?
watch on image
Quote from: McArti0 on 19:13, 01 April 24watch on image
Oh man... 🤦�♂️ Can't believe I missed that. I wasn't even being lazy. Did lots of searching for 'best ground on cpc motherboard, etc.
Dziękuję. I will try those tonight/tomorrow.
Quote from: McArti0 on 15:15, 01 April 24What about this Pin7 of IC105? died?
So, I tried Pin7 of IC105 with the multimeter and measured 1v. So the pin is not dead.
Also here are the results from the 40010:
Pin Me Expected
CCLK 4 1.76v ~1.2v
RAS 34 1.01v ~0.9v
RAMRD 29 4.93v ~3.5v
nCPU 14 2.57v ~2.5v
ROMEN 27 2.84v ~2.8v
PHI 19 2.06v ~2.1v
READY 22 1.33v ~1v
244en 23 4.95v ~4.5v
CCLK, RAMRD, and maybe READY all look too high.
Happy to test anything else, now that I used the multimeter and didn't blow anything up (so far).
This means that the RAM is not read very often.
Now You must have 4 hands.
ROMDIS on expansion port. signal to ground And check RAMRD voltage.
Ground is on screw at flopy drive too.
Quote from: McArti0 on 06:49, 02 April 24This means that the RAM is not read very often.
Now You must have 4 hands.
ROMDIS on expansion port. signal to ground And check RAMRD voltage.
Ground is on screw at flopy drive too.
Just to confirm:
1. Connect ROMDIS to ground first
2. Measure RAMRD voltage
I'll get one of my kids to help me. 🫣😅
Sorry. ROMDIS to ground is nonsens :picard:
Quote from: McArti0 on 11:52, 02 April 24Sorry. ROMDIS to ground is nonsens :picard:
Haha, no problem. ;) Are there more diagnostic tests I can do? Are the RAM chips a likely cause?
Ok ROMDIS to 5V pin28 of ROM. check if anything on the screen changes color.
and monte carlo REFRESH Z80 pin30 to BUSREQ pin37. check if anything on the screen changes color.
Quote from: McArti0 on 21:16, 02 April 24Ok ROMDIS to 5V pin28 of ROM. check if anything on the screen changes color.
and monte carlo REFRESH Z80 pin30 to BUSREQ pin37. check if anything on the screen changes color.
Thanks once again, for sticking with this.
Connecting Z80 pin30 to BUSREQ pin37 didn't change the screen. (Wasn't sure if monte carlo was an autocorrect thing)
I wasn't sure which ROM chip, so I tried pin28 of 40015, 40025, and also 40010... connected to ROMDIS pin43 on the expansion port. Also didn't change the screen.
Actually, I just checked which pin is REFRESH on Z80. Is it pin28 and not pin30?
I will try Z80 pin28 to BUSREQ pin37 in the morning.
Quote from: mmldrm on 12:03, 02 April 24Are the RAM chips a likely cause?
yes. it's just not clear which one(s).
Pin30 pin37 expansion port. Z80 pin 28,25 Sorry.
Quote from: McArti0 on 06:16, 03 April 24Pin30 pin37 expansion port. Z80 pin 28,25 Sorry.
Connecting Z80 pin30 -> expansion port pin37 : no change to screen
I wasn't sure what Z80 pin 28,25 meant, sorry.
Looking at Z80 docs, pin25 is BUSRQ and pin28 is REFSH. Connect those? I didn't want to make a guess and do damage. ;D
All what i write to You I was tested on my real 6128. Still live. :D
Quick short touch.
Quote from: McArti0 on 06:16, 03 April 24Pin30 pin37 expansion port. Z80 pin 28,25 Sorry.
what do you want to achieve by that?
Quote from: mmldrm on 13:54, 03 April 24Connecting Z80 pin30 -> expansion port pin37 : no change to screen
Your RAM is currently malfunctioning. The only thing you can achieve by connecting other pins would be that you have a totally black screen.
If you have a ROM board or Dandanator or M4 you can execute the Amstrad diagnostics ROM. If a subset of RAM IC shows in red it's most likely those ICs or maybe the LS373 latch. If all RAMs appear red, I would check the MUX chips.
Did you check if any of the RAM ICs gets hotter than the rest?
artifacts resulting from accidental writes to frames that I see on my screen
Quote from: eto on 15:43, 03 April 24Quote from: mmldrm on 13:54, 03 April 24Connecting Z80 pin30 -> expansion port pin37 : no change to screen
Your RAM is currently malfunctioning. The only thing you can achieve by connecting other pins would be that you have a totally black screen.
If you have a ROM board or Dandanator or M4 you can execute the Amstrad diagnostics ROM. If a subset of RAM IC shows in red it's most likely those ICs or maybe the LS373 latch. If all RAMs appear red, I would check the MUX chips.
Did you check if any of the RAM ICs gets hotter than the rest?
Unfortunately no ROM board, Dandanator, or M4 here. A friend owns this 6128 and I volunteered to investigate, with my very limited knowledge. (I'm more a software guy. I think I have already proven here that I am not much of a hardware guy! 😅)
Maybe I could buy a Dandanator on eBay for €50 and hopefully sell it again without losing too much money, but it is a risk for something I would use once (and not even my 6128).
I will check each RAM IC for temperature. Could I also do continuity or diode checks with a multimeter?
Thanks,
M
Quote from: McArti0 on 15:23, 03 April 24All what i write to You I was tested on my real 6128. Still live. :D
Quick short touch.
Haha, good to know! I also don't want to accidentally do something different to what you did. 🫣
So I connect Z80 pin25 to Z80 pin28? A quick touch? And I should see a change on screen?
That will lead to nothing.
You are trying to diagnose something, that you have no clue about on how it works, with a tool that is insufficient for the job (a multimeter cannot answer the question on which RAM chip is dead that way) and also you are not really familar on how to use a multimeter.
Just asking on how you want to get any diagnosis out of that approach. Do yourself a favor and find someone who can diagnose it for you.
There is a schematic available, you need to be able to read it and at least an oscilloscope to understand what signals are lookging good and which ones are not.
For now we can only check RAM bank one (assuming not any of the other chips are dead like the MUX chips or the Latch).
So check chips IC127-IC134. This is Bank1. Check Pin P14 of each chip. It should change its value. So on a voltmeter it would look like you have a voltage in between +5V and 0V. But it cannot be either +5V and not 0V. +5V would mean a line is allways high, and that is a stuck bit and a line that is allways 0V is stuck at low level. Both indicate a faulty RAM chip. If all Chips are the same (all low for instance) it is more an indication of a faulty logic chip.
Yes.
I'm trying to find a procedure that will allow us to recognize whether RAM or ROM is damaged, without using an oscilloscope or dandanator.
SerErris. RAMRD is near 5V Ram is never read by Z80.
RAMRD at 5V does not need to mean anything.
CPU is first reading from ROM and only after quite some initializations it is actually reading something from RAM. The GA setup with colors for instance already have happend at that stage, so you would at least see a blue screen, not white on black border.
If the ROM is the culprit here, then that would lead to this scenario as well.
You should check all the ROM datalines if they are stuck.
Quote from: mmldrm on 23:15, 01 April 24Quote from: McArti0 on 15:15, 01 April 24What about this Pin7 of IC105? died?
So, I tried Pin7 of IC105 with the multimeter and measured 1v. So the pin is not dead.
Also here are the results from the 40010:
Pin Me Expected
CCLK 4 1.76v ~1.2v
RAS 34 1.01v ~0.9v
RAMRD 29 4.93v ~3.5v
nCPU 14 2.57v ~2.5v
ROMEN 27 2.84v ~2.8v
PHI 19 2.06v ~2.1v
READY 22 1.33v ~1v
244en 23 4.95v ~4.5v
CCLK, RAMRD, and maybe READY all look too high.
Happy to test anything else, now that I used the multimeter and didn't blow anything up (so far).
The voltage readings are depending on your multimeter and how quick it reacts on changes (e.g. how much it averages). It is not the perfect tool for analyzing digital signals.
We are talking about 4Mhz frequency for the PHI signal for instance and all the others are also pretty much aligned to that frequency.
For real understanding on what is going on you would need at least a cheap oscilloscope with enough bandwidth:
https://www.amazon.de/dp/B0C1SLST75
The above one is the bare minimum capabale Oscilloscope I found on Amazon. The cheap ones, are not fast enough (500khz or even lower) to actually see anything.
You can also use a logic analyzer as they are just much cheaper, however not the same as an oscilloscope.
This is just one of the examples.
Sorry for the german Amazon links.
https://www.amazon.de/dp/B07K6HXDH1
Back to the Signals:
Your 244en is also always high. So that means that there is no IORequest coming into the GateArray. The 244en signal is created by two bits from sequencer (which is running as we already worked out) and the IORQ bit. It goes down when the CPU sends an IORQ to be able to get the output to the GA.
That it never goes down means, the CPU is never sending an IORequest.
A interpretation of that is: ROM is dead.
(obviously there are other reasons possible, like GA defect, but ROM is most likely here).
Please go ahead and measure the data pins on the ROM (IC103). If they are stuck, the ROM is dead.
Voltage on Data line of Rom is not any info, because always R-rfresh is on this line. :(
Quote from: SerErris on 22:11, 03 April 24That will lead to nothing.
You are trying to diagnose something, that you have no clue about on how it works, with a tool that is insufficient for the job (a multimeter cannot answer the question on which RAM chip is dead that way) and also you are not really familar on how to use a multimeter.
Just asking on how you want to get any diagnosis out of that approach. Do yourself a favor and find someone who can diagnose it for you.
First of all, thank you for your help and suggestions in later replies. I know it can be frustrating to be helping someone with little understanding of electrical engineering and CPC hardware. (I did maybe one university course 30 years ago, and have no access to an oscilloscope.)
This story started with a friend telling me he got his 6128 out from the attic after many years and it had this boot problem. I offered to take a look, hoping it was something simple enough for me to fix... but it looks like it isn't so simple.
I thought maybe with some expert knowledge on the forums that I/we might find the cause and that the fix might be simple (swap out a socketed IC, for example) so I registered and posted on here. Maybe it will reach a dead end without diagnostic tools, but I wanted to try my best and avoid this 6128 going into the bin.
You are 100% right that the best way to fix computer is to find a local expert. But time is money and I think an expert would cost more than the 6128 is worth, and this was just a fun challenge, trying to learn and save a little bit of 8-bit history.
Anyway, thank you and McArti0 for taking the time to read and reply to this post. I will trust you guys if you tell me there is nothing more that can be done (without equipment and knowledge), then I/we tried our best.
Thanks again
M
P.S. I will give those RAM/ROM measurements a try after work today, thanks.
Quote from: SerErris on 10:31, 04 April 24CPU is first reading from ROM and only after quite some initializations it is actually reading something from RAM. The GA setup with colors for instance already have happend at that stage, so you would at least see a blue screen, not white on black border.
Black border, grey/white screen is typical when RAM is malfunctioning. If ROM is dead, you will have a completely black screen. I have repaired both situations several times and it was always RAM when it's the grey/white screen.
And when looking at the firmware this also makes sense. That's my understanding from reading the source:
As long as the ROM works, the firmware can do (not many but) a bit of initialisation purely in ROM before RAM is absolutely crucial to work properly. These first steps in ROM include setting the GateArray to Mode 1 and setting the CRTC registers to the right values for a 50 or 60Hz display. The first few thinbgs in firmware are mostly sequential, where it uses JP or JR statements. No RAM is needed here.
But after the initial "easy" things, the firmware then starts to CALL subroutines. And a CALL needs the stackpointer in working RAM. So latest when the first subroutine tries to return, the boot process will fail as the return address cannot be read from RAM anymore. It simply can't get to the point where the palette is initialised.
When z80 has no data connected, does it read command 00 or FF?
This is how I recently simulated the problem on a 464: one of the Dout pins of a single RAM IC is connected to GND - and voilà - the black border / grey screen. Remove the short - and it's back to normal again.
Sounds absolutely reasonable.
So please to the check with the multimeter to check every data pin of bank1 (you can ignore bank2 for now) and check if any databit is stuck.
Also if you have a EPROM programmer, you can replace the ROM with a diagnostic ROM.
As the rest seems to function you will get an indication of which RAM is actually faulty.
Quote from: SerErris on 10:45, 04 April 24Your 244en is also always high. So that means that there is no IORequest coming into the GateArray. The 244en signal is created by two bits from sequencer (which is running as we already worked out) and the IORQ bit. It goes down when the CPU sends an IORQ to be able to get the output to the GA.
That it never goes down means, the CPU is never sending an IORequest.
A interpretation of that is: ROM is dead.
(obviously there are other reasons possible, like GA defect, but ROM is most likely here).
Please go ahead and measure the data pins on the ROM (IC103). If they are stuck, the ROM is dead.
I had a busy couple of days with work, but finally back to this. (Not sure if that is good news or not!)
So, the IC103/40025 ROM looks okay. I measured the data pins and all were >0.
Pin Value
11 4.14
12 4.14
13 4.13
15 4.19
16 0.63
17 4.21
18 4.22
19 0.74I will now go check the data pins on bank 1. (Thanks again for the help.)
Quote from: SerErris on 10:45, 04 April 24Your 244en is also always high. So that means that there is no IORequest coming into the GateArray. The 244en signal is created by two bits from sequencer (which is running as we already worked out) and the IORQ bit. It goes down when the CPU sends an IORQ to be able to get the output to the GA.
That it never goes down means, the CPU is never sending an IORequest.
The CPU is sending IO requests to the GateArray only when it tries to change something. E.g. screen mode or palette or enabling/disabling ROM. But in this case here, that happens only during the very first nanoseconds of the boot process - then it hangs and no further commands will be sent to the Gate Array.
It's absolutely impossible to measure these initial few commands with a multimeter.
If the boot process is successful this is probably different. I haven't tried it yet but I would assume that the firmware is constantly switching on/off the BASIC and firmware ROM to access the RAM in the same area. Then you will see activity on 244en. But if the system halts, there will just be no further interaction with the GateArray and thus the 244en signal will always be high.
Quote from: SerErris on 10:00, 05 April 24Sounds absolutely reasonable.
So please to the check with the multimeter to check every data pin of bank1 (you can ignore bank2 for now) and check if any databit is stuck.
Also if you have a EPROM programmer, you can replace the ROM with a diagnostic ROM.
As the rest seems to function you will get an indication of which RAM is actually faulty.
I just want to make sure - should I focus on testing pin2 (data in) and pin14 (data out) rather than the other pins, like address, etc.
(Unfortunately I have no access to an EPROM programmer.)
Quote from: SerErris on 10:00, 05 April 24Sounds absolutely reasonable.
So please to the check with the multimeter to check every data pin of bank1 (you can ignore bank2 for now) and check if any databit is stuck.
Also if you have a EPROM programmer, you can replace the ROM with a diagnostic ROM.
As the rest seems to function you will get an indication of which RAM is actually faulty.
So I checked pin2 and pin14 of each RAM chip (Bank 1, IC127-IC134).
Each pin2 read between 0.58v and 4.15v.
Each pin14 read 2.9v-3.0v, except IC127 which read 1.29v.
A side note - one time I switched on to this screen, which was a little different. Sadly, it hasn't happened again.
Quote from: eto on 22:06, 04 April 24This is how I recently simulated the problem on a 464: one of the Dout pins of a single RAM IC is connected to GND - and voilà - the black border / grey screen. Remove the short - and it's back to normal again.
I don't quite know how to apply this information to my malfunctioning 6128, other than it being more evidence that it is most likely a RAM problem, but thanks!
If I don't get positive confirmation of the issue, at least the RAM chips are cheap to replace. So I could try that if all else fails.
Thanks again!
Quote from: mmldrm on 17:48, 05 April 24Each pin14 read 2.9v-3.0v, except IC127 which read 1.29v.
I suspect that this shows that data is being read from all eight chips, with a mix of 1s and 0s mostly averaging out at ~3v, and that the 1.29v just happens to be data with more 0s than 1s? i.e. the RAM chips look ok?
Or, if it makes IC127 look suspicious, I am probably happy to try replacing IC127 to see if that makes a difference. I just assumed that if 7 of the 8 RAM chips are working then it would get further than the black and white screen. But I guess it depends which area in memory has a problem, or because the RAM chips work together.
Each individual 4164 IC is 1bit, and hence there are 8 to make up the 8 bit data bus. It only takes 1 IC to fail and you have corrupt data.
Quote from: mmldrm on 17:48, 05 April 24A side note - one time I switched on to this screen, which was a little different. Sadly, it hasn't happened again.
This is the correct image from cold memories right after turning on. Are you sure it wasn't an image from a wire PAL replacement?
Quote from: Rabs on 21:56, 05 April 24Each individual 4164 IC is 1bit, and hence there are 8 to make up the 8 bit data bus. It only takes 1 IC to fail and you have corrupt data.
Thanks, I did think that was the case (having read many old discussions here) but that confirms it.
More and more, I might just buy the 8 RAM chips and cross my fingers that there are no other issues. At least those are inexpensive (around £10 I think?).
Quote from: McArti0 on 22:09, 05 April 24This is the correct image from cold memories right after turning on. Are you sure it wasn't an image from a wire PAL replacement?
Sorry, what is a wire PAL replacement? If it is a mod, I can check the board for it.
But this Amstrad is from a friend who owned it since he was a kid in the 80s and not really used it since the 80s either. (Either never, or almost never.)
Very interesting that it is the correct image though! :o
Quote from: mmldrm on 22:41, 05 April 24Sorry, what is a wire PAL replacement? If it is a mod, I can check the board for it.
Quote from: mmldrm on 22:02, 31 March 24Suspecting the RAM, I did the "RAM bank swap trick" by removing the IC118 and replacing it with some jumper wires in the appropriate positions (triple-checked!).
Quote from: McArti0 on 06:55, 06 April 24Quote from: mmldrm on 22:41, 05 April 24Sorry, what is a wire PAL replacement? If it is a mod, I can check the board for it.
Quote from: mmldrm on 22:02, 31 March 24Suspecting the RAM, I did the "RAM bank swap trick" by removing the IC118 and replacing it with some jumper wires in the appropriate positions (triple-checked!).
Thanks. When I see PAL, I think of video signals.
Definitely not that. The IC118 is back in place.
Three full forum pages of faffing around with pointless tests. Just swap the RAM and get on with enjoying your CPC.
Bryce.
Quote from: Bryce on 11:33, 06 April 24Three full forum pages of faffing around with pointless tests. Just swap the RAM and get on with enjoying your CPC.
I have now ordered RAM, but was hoping to have positive confirmation that RAM is the problem before doing so.
For example, removing the HAL chip (40031) completely makes no difference. So maybe I'm replacing RAM when the 40031 is the problem. That's the kind of thing that has been putting me off buying replacement parts.
40031
pin7 and pin17 have the same voltage. ~3v
pin8, pin12 have the same voltage.
pin9, pin19 have the same voltage.
ps.
Quote from: mmldrm on 17:48, 05 April 24A side note - one time I switched on to this screen, which was a little different. Sadly, it hasn't happened again.
WHEN you seen this image! with 40031 or without?
Quote from: McArti0 on 19:38, 06 April 2440031
pin7 and pin17 have the same voltage. ~3v
pin8, pin12 have the same voltage.
pin9, pin19 have the same voltage.
pin7 (2.18v) - pin17 (3.63v)
pin8 (1.67v) - pin12 (3.73v)
pin9 (0.66v) - pin19 (0.75v)
But should it be pin7 to pin16? (And pin17 to pin20?)
Actually, pin16 was the same as pin17... 3.63v. And only pin9-pin19 values are close.
ps.
Quote from: McArti0 on 19:38, 06 April 24Quote from: mmldrm on 17:48, 05 April 24A side note - one time I switched on to this screen, which was a little different. Sadly, it hasn't happened again.
WHEN you seen this image! with 40031 or without?
With 40031. Still only ever seen it once. :(
I ordered 8x RAM chips (and sockets), so I can replace the main bank. But I get the impression it might not just be RAM that is faulty here...
ramove 40031 and add 3 wire jumper pin 7-17, pin8-12, pin9-19
check by Ώ on multimeter if pin 7 is join to RAM pin15 ic127-ic134 (mutimeter show ~0 )
check pin 8 is join to IC109 pin10
check pin 9 is join to IC109 pin 6
https://www.cpcwiki.eu/imgs/4/4a/CPC6128_Schematic.png
Quote from: McArti0 on 21:15, 06 April 24ramove 40031 and add 3 wire jumper pin 7-17, pin8-12, pin9-19
check by Ώ on multimeter if pin 7 is join to RAM pin15 ic127-ic134 (mutimeter show ~0 )
check pin 8 is join to IC109 pin10
check pin 9 is join to IC109 pin 6
https://www.cpcwiki.eu/imgs/4/4a/CPC6128_Schematic.png
With 40031 replaced by 3 jumpers:
pin7 to RAM pin15 - all 8 RAM chips show resistance of 47-48
Ώpin8 to IC109 pin10 - 0.5
Ώpin9 to IC109 pin - 0.6
Ώ
Ok. any changes on screen?
next changes not wire 7-17 but 7-16
Quote from: McArti0 on 06:06, 07 April 24Ok. any changes on screen?
next changes not wire 7-17 but 7-16
Both 7-17 and 7-16 get the same black border, white screen.
Should I have a pin going to pin20? Like 7-16 and 17-20? The YouTube video for this shows a 4th pin.
BTW with speaker volume high, there has always been a barely audible pulse, roughly 1Hz. I don't know if that means anything. (Sorry, I only realised it was coming from the 6128 now. I thought it was coming from outside the house, it is so faint.)
normal.
20-17 or 20-16 for other bank reduce noise. its ok.
Check RAMDIS to ground and turn on. And check visible condition R124.
Quote from: McArti0 on 19:32, 07 April 24Check RAMDIS to ground and turn on. And check visible condition R124.
Hi,
Sorry, I was away for a few days. Do you mean these steps:
1. Connect RAMDIS to ground.
2. Turn on 6128.
3. Check voltage across R124
R124 normally connects ramdis to ground. We want to check if this is really happening.
Then check if the voltage on IC112 pin8 is the same on pin9.
Quote from: McArti0 on 16:34, 13 April 24R124 normally connects ramdis to ground. We want to check if this is really happening.
Then check if the voltage on IC112 pin8 is the same on pin9.
RAMDIS is grounded (0v)
IC112 pin8 - 4.34v
IC112 pin9 - 4.93v
Try RAMDIS grounded and turn ON CPC.
Quote from: McArti0 on 18:29, 13 April 24Try RAMDIS grounded and turn ON CPC.
RAMDIS grounded while turning on, still white screen. :(
Ok. And now when CPC working +5V connect to ROMDIS. IC103 Pin22 to Pin28
Quote from: McArti0 on 18:59, 13 April 24Ok. And now when CPC working +5V connect to ROMDIS. IC103 Pin22 to Pin28
Connect IC103 Pin22 to Pin28, +5v to ROMDIS, and switch on?
NO First ON next Connect
Quote from: McArti0 on 19:06, 13 April 24NO First ON next Connect
Thanks.
I powered on, connected +5v to ROMDIS, then connected IC103 Pin22 to Pin28. No change on screen at all.
(I must admit, I am very close to buying a cheap oscilloscope if we are close to running out of ideas/options here.)
Quote from: mmldrm on 20:36, 13 April 24I must admit, I am very close to buying a cheap oscilloscope if we are close to running out of ideas/options here.)
have you replaced the RAM ICs already? Or did you try the Amstrad diagnostics ROM?
we cant see anything because we have all colors white. :'( BUT ... ;D
When you start CPU. Next Disable ROM by const +5V on ROMDIS
Voltages on working good bank RAMs are:
Pin14 IC127 ~2V
Pin14 IC128 ~0V
Pin14 IC129 ~0V
Pin14 IC130 ~2V
Pin14 IC131 ~2V
Pin14 IC132 ~2V
Pin14 IC133 ~0V
Pin14 IC134 ~0V
This is address 39h pushing on stack after doing FF (RST 38) reading from bad gate.
bingo!
Quote from: McArti0 on 21:48, 13 April 24we cant see anything because we have all colors white. :'( BUT ... ;D
When you start CPU. Next Disable ROM by const +5V on ROMDIS
Voltages on working good bank RAMs are:
Pin14 IC127 ~2V
Pin14 IC128 ~0V
Pin14 IC129 ~0V
Pin14 IC130 ~2V
Pin14 IC131 ~2V
Pin14 IC132 ~2V
Pin14 IC133 ~0V
Pin14 IC134 ~0V
This is address 39h pushing on stack after doing FF (RST 38) reading from bad gate.
bingo!
Thanks. So... I used expansion port VCC (pin27) -> ROMDIS (pin43) for . It was near, and I had no hands helping me.
Pin14 on all 8 ICs (127-134) are wild. None of them are stable values.
Check voltage on All pins 2.
Also all unstable values.
Interestingly, while my second pair of hands (daughter) was holding VCC to ROMDIS, we got a bunch of yellow screen artefacts. Maybe she lost connection or connected VCC to RSET or RAMDIS, but I thought I would share.
It can't be a weak connection.
mayby beter point for RomDis is here....
Quote from: mmldrm on 17:27, 14 April 24Also all unstable values.
Interestingly, while my second pair of hands (daughter) was holding VCC to ROMDIS, we got a bunch of yellow screen artefacts. Maybe she lost connection or connected VCC to RSET or RAMDIS, but I thought I would share.
More likely she was coming in contact with some part of the clock source (before TTL conversion) which would cause multiple resets and similar effects.
Bryce.
Quote from: McArti0 on 19:24, 14 April 24It can't be a weak connection.
mayby beter point for RomDis is here....
I have attached an image of what I did with ROMDIS while checking the DRAM pins. Just to confirm it was correct.
Quote from: McArti0 on 19:24, 14 April 24It can't be a weak connection.
mayby beter point for RomDis is here....
In your latest image, what should I do between R112 and diode D201?
Should I take +ve from pin40 (IC201) and connect it to ROMDIS on the expansion port?
Place between r112 and diode is the same ROMDIS but i think better to hold probe. Pin40 of 765 is better to hold 5v
Quote from: McArti0 on 06:23, 15 April 24Place between r112 and diode is the same ROMDIS but i think better to hold probe. Pin40 of 765 is better to hold 5v
Thanks. We did this and there was nothing on-screen. I think this is correct because without ROM the board does nothing?
I measured pin2 and pin14 of IC127-IC134. All are still fluctuating between 0-5v, same as last time. :(
Quote from: mmldrm on 20:01, 15 April 24All are still fluctuating
but how? does ic127 have a different voltage from time to time?
What have you now in PAL 40031 socket?
Remember that we do not expect any changes on the screen because we do not have any colors set. All colors are white.
Before measuring, connect +5V to RAMDIS for 1 second, but the PAL socket should not be 40031, but 4 wires.
1 Turn ON
2 connect +5V to ROMDIS const
3 connect +5V to RAMDIS 1sek
4 measuring Pin14 and pin 2
Quote from: McArti0 on 20:23, 15 April 24Remember that we do not expect any changes on the screen because we do not have any colors set. All colors are white.
Before measuring, connect +5V to RAMDIS for 1 second, but the PAL socket should not be 40031, but 4 wires.
Ahh ok.
I had +5V to ROMDIS for the whole time I was measuring the RAM pin2/pin14 voltages.
I also had the 40031 IC in the socket.
I will do the tests again, with the 4 wires replacing the 40031 IC.
Quote from: McArti0 on 20:16, 15 April 24but how? does ic127 have a different voltage from time to time?
Yes, IC127-134 pin2/pin14 voltages were always moving between ~0.2V and ~4.8V. But I will retest with 40031 replaced with the 4 wires.
Should I keep +5V between r112 and diode while I am measuring the RAM voltages? Or remove it after 1 second, then test RAM?
rOm Dis long lasting connected
rAm Dis 1sek.
measuring voltages ... pin14 all memories and write where is ~0V
Quote from: McArti0 on 21:48, 13 April 24Voltages on working good bank RAMs are:
Pin14 IC127 ~2V
Pin14 IC128 ~0V
Pin14 IC129 ~0V
Pin14 IC130 ~2V
Pin14 IC131 ~2V
Pin14 IC132 ~2V
Pin14 IC133 ~0V
Pin14 IC134 ~0V
This is address 39h pushing on stack after doing FF (RST 38) reading from bad gate.
bingo!
Ok, I finally got a helper. :)
Pin14 IC127 ~2.5
Pin14 IC128 ~0.04
Pin14 IC129 ~0.04
Pin14 IC130 ~2.5
Pin14 IC131 ~2.5
Pin14 IC132 ~2.5
Pin14 IC133 ~0.04
Pin14 IC134 ~0.1 (I think - maybe 0.04)
Pin2 IC127 ~4.2
Pin2 IC128 ~4.2
Pin2 IC129 ~4.2
Pin2 IC130 ~4.2
Pin2 IC131 ~0.6
Pin2 IC132 ~4.2
Pin2 IC133 ~4.2
Pin2 IC134 ~0.7
So they are very clearly hi/lo values, if a little off (2.0v vs. 2.5v). Is pin2 ok at 4.2v hi?
Is this looking like RAM is ok but maybe RAM management isn't?
(Also, I finally bought the oscilloscope this morning. It won't arrive until around 24/4. Maybe we will have a solution before it arrives? 😂 )
Also, I know this has probably been a terrible process, but I am really enjoying learning about the 6128 internals. Thank you.
Yes RAM has been written!!! by 0011 1001 = 39h next this is instruction ADD HL,SP no write to RAM.
But why on Pin2 6Fh?
Quote from: McArti0 on 21:30, 17 April 24Yes RAM has been written!!! by 0011 1001 = 39h next this is instruction ADD HL,SP no write to RAM.
But why on Pin2 6Fh?
Fantastic! This gives me some hope. (Even though I already bought the replacement RAM. Oops.)
Tomorrow I can do the same test again, and make sure the pin2 result is definitely 11110110 / 0x6F. Maybe I made a mistake when writing the results.
Quote from: McArti0 on 21:30, 17 April 24Yes RAM has been written!!! by 0011 1001 = 39h next this is instruction ADD HL,SP no write to RAM.
But why on Pin2 6Fh?
...because I screwed up!
I just retried and got the same results on both pins.
Pin2 TAKE TWO
Pin2 IC127 hi
Pin2 IC128 lo
Pin2 IC129 lo
Pin2 IC130 hi
Pin2 IC131 hi
Pin2 IC132 hi
Pin2 IC133 lo
Pin2 IC134 lo
Pin14 TAKE TWO
Pin14 IC127 hi
Pin14 IC128 lo
Pin14 IC129 lo
Pin14 IC130 hi
Pin14 IC131 hi
Pin14 IC132 hi
Pin14 IC133 lo
Pin14 IC134 loSo 0011 1001 = 39h both times!
Ok. Now you need to check why the ram is not being read.
Give us sharp Photo your mainboard.
I removed the IC118 wires so you could see everything.
ok. Now test 74LS244.
When connect red wire whole CPC halted, screen too.
You have to be sure that you have drawn 0V on Pin1 LS244,
if not, you connect again.
Then check if D0-D7 on the left and right are the same.
Work for 6 hands.
Quote from: McArti0 on 21:14, 18 April 24ok. Now test 74LS244.
How do I test it? Do I start with the same ROMDIS and RAMDIS steps, then measure voltages?
no, I added it above. Sorry :blank:
Quote from: McArti0 on 21:14, 18 April 24ok. Now test 74LS244.
When connect red wire whole CPC halted, screen too.
You have to be sure that you have drawn 0V on Pin1 LS244,
if not, you connect again.
Then check if D0-D7 on the left and right are the same.
Work for 6 hands.
Should I do the same ROMDIS and RAMDIS steps first?
And 40031/IC118 is still replaced by 4 wires?
NO
YES
Thanks. I had to have the red wire connected before switching on, or it didn't do anything. I hope that is correct.
Results are interesting I think. I will share the full measurements across the IC, and then my comparisons. Maybe I got them wrong, so it is better you see everything.
IC114 Pin Measurements
----------------------
pin voltage
1 0
d0 2 1.4
3 0.09 d7
d1 4 4.15
5 0.09 d6
d2 6 0
7 0.09 d5
d3 8 4.15
9 5 d4
10 0
d4 11 4
12 5 d3
d5 13 0
14 0 d2
d6 15 0
16 5 d1
d7 17 0
18 5 d0
19 0
20 5
Comparisons
-----------
left right
D0 1.4 5
D1 4 5
D2 0 0
D3 4 5
D4 4 5
D5 0 0
D6 0 0
D7 0 0
Everything matches, except D0 has a strange 1.4v value.
I double checked these to make sure. But I can do the same test again, if it is good to have two sets of results.
Quote from: mmldrm on 10:50, 20 April 24Thanks. I had to have the red wire connected before switching on, or it didn't do anything. I hope that is correct..
This is an indication that after a certain point there are no writes to GA 40010.
Quote from: McArti0 on 12:49, 20 April 24This is an indication that after a certain point there are no writes to GA 40010.
Do I need to check traces and signals now?
You have done an amazing job guiding me with a multimeter, but I expect the oscilloscope to arrive in 5 days. I hope that helps to find the fault (or faults).
much better will be to buy Dandanator with test of components.
now we have 1.4v on D0. test it again and see if READY is 0
Quote from: McArti0 on 17:31, 20 April 24much better will be to buy Dandanator with test of components.
now we have 1.4v on D0. test it again and see if READY is 0
I seem to be very good at buying the wrong things (RAM, oscilloscope) at the wrong times. 😅
Ran the test again. With 1.4v on D0, READY (pin22 on 40010) is 0.
We won't stop where we should.
Look Fig. 8 here... https://bread80.com/2021/06/03/understanding-the-amstrad-cpc-video-ram-and-gate-array-subsystem/
N244EN have 0 two times. We find the first meaningless one. :picard:
Ok first We must short IORQ and READY , and turn on CPC.
pin 20 Z80 should be 0
First instruction is:
LD BC,#7F89
OUT (c),c
Our Z80 is stopped at write to GA
next we add RED connect and check if Pin1 244 is 0. If not, RED to reconnect.
GA is stopped.
now 244 is open and Left and Right side should be 89h 1000 1001.
After I short IORQ and READY, then turn on the CPC, pin20 of Z80 is 0 (well, 0.14v so I guess 'low' is correct).
Should I continue to short IORQ and READY? Or stop?
I removed IORQ->READY and added the RED connection. But voltage was 5v.
I expect you will say I must keep IORQ->READY so the Z80 remains stopped, while I do RED connect and measurements. But I wanted to confirm.
More simply, should I continue to short IORQ->READY while I also do IC115->R143 (red wire)? So both wires at the same time. (4 hands!)
(Also, I now have an oscilloscope, if that is useful during these tests.)
Yes permanetly stop Z80 by continue iorq-ready and next permanetly stop GA by continue Red.
Thanks. I thought that must be correct.
Is there a way of doing this with less than 3 people? 😅 I did see that I could use a little solder to fix the wires to the pins, but I'm not doing this because I don't own this 6128.
If not, it's ok, I have two teenagers that can help.
MacGyver could do it himself with one hand.
You can do It on stiff wires. z80 wait line (Ready) and iorq pin 20
Quote from: McArti0 on 07:46, 21 April 24Ok first We must short IORQ and READY , and turn on CPC.
pin 20 Z80 should be 0
First instruction is:
LD BC,#7F89
OUT (c),c
Our Z80 is stopped at write to GA
next we add RED connect and check if Pin1 244 is 0. If not, RED to reconnect.
GA is stopped.
now 244 is open and Left and Right side should be 89h 1000 1001.
I'm back! Sadly no MacGyver here, but a few kind family members are better than nothing.
Left and right sides were:
Left Right
D0 1.5 5
D1 0 0
D2 0 0
D3 0 0
D4 5 5
D5 0 0
D6 5 1.5
D7 0 0Maybe that is right and left? Left=pin2-17. Right=pin3-18.
Anyway, 10001010 I think. 8Ah? 🤔 Not quite right.
I will try again when my helpers are home, to get two sets of data .
Have you often confirmed that pin1 244 is 0?
I am not sure I do understand what you mean by stopping GA, GA will only stop if you stop the clock. Otherwise it will continue to output catch clock an will send RAS CAS to memory. And that will influence what you see on the right side of LS244.
You may hunt a ghost here esp. with the only tool available being a voltmeter.
Also if you see 1.5V on the left side, you have either not stopped the CPU or something is influencing the D7 line.
And again a Voltmeter is not good to troubleshoot digital logic.
I'm starting to think that this is some sort of never ending April fools joke. The initial failure you showed in the very first post is a classic symptom of a failed RAM and by far the most likely culprit. But instead of just swapping them and getting on with life, you've now spent over a month doing ends tests of questionable usefulness, which don't really tell you anything for certain and some of which were definitely totally pointless.
So I really have to ask. Why don't you just swap the RAMs?
Bryce.
Quote from: Bryce on 13:35, 30 April 24I'm starting to think that this is some sort of never ending April fools joke. The initial failure you showed in the very first post is a classic symptom of a failed RAM and by far the most likely culprit. But instead of just swapping them and getting on with life, you've now spent over a month doing ends tests of questionable usefulness, which don't really tell you anything for certain and some of which were definitely totally pointless.
So I really have to ask. Why don't you just swap the RAMs?
I genuinely apologise if this comes across as trolling or something. It isn't. I'm just trying to follow instructions from whoever is willing to take the time to help me.
As it isn't my 6128, nor my budget, I've been trying to avoid purchasing anything unnecessarily. For example, I'd love to go buy a Dandanator, but I can't justify it for this one-off investigation.
I did buy 64Kb of RAM (and sockets), but I think McArti0 managed to diagnose that the fault lies elsewhere (IC114/IC115 maybe?).
I could replace the RAM. I just hoped that we could get a definite diagnosis of where the fault lies first, rather than risk my soldering skills introducing more doubt.
I did buy a oscilloscope, so if that in any way helps to diagnose at this point then I'm more than willing to justify the purchase by using it.
Quote from: SerErris on 12:46, 30 April 24I am not sure I do understand what you mean by stopping GA, GA will only stop if you stop the clock. Otherwise it will continue to output catch clock an will send RAS CAS to memory. And that will influence what you see on the right side of LS244.
You may hunt a ghost here esp. with the only tool available being a voltmeter.
Also if you see 1.5V on the left side, you have either not stopped the CPU or something is influencing the D7 line.
And again a Voltmeter is not good to troubleshoot digital logic.
I do have an oscilloscope now (the one that you suggested), if that gives more options to try.
(I might also buy those mini grabber clips, rather than trust the assistance of my kind but inexperienced family members.)
Quote from: McArti0 on 07:12, 30 April 24Have you often confirmed that pin1 244 is 0?
I just bought some hook probes (now that I know that the are suitable) and evicted my family members from the team.
These should arrive by Friday, and will allow more reliable results.
Quote from: Bryce on 13:35, 30 April 24I'm starting to think that this is some sort of never ending April fools joke..
Then prove to him that his memory is faulty. So far I have found out that the 3rd byte of the Rom has the value 61 instead of 89. If at all he did it well with 3 people. For now, it turns out that z80 writes to frames. And that the ram has the correct contents. Maybe you could stop being mean and start helping.
Quote from: SerErris on 12:46, 30 April 24I am not sure I do understand what you mean by stopping GA, GA will only stop if you stop the clock.
Take a look at the illustration in post no. 100. GA stops itself due to the Lo signal on pin 1 of 244. This freezes the clock generator. I'll never believe you could be as blind as Bryce It's not like you.
A very cheap option would be to desolder the rom and put in a socket and a diagnostic eeprom.
That will tell you which ram chip is the problem here.
For all the RAM chips of bank1 check each output pin with the oscilloscope. This may show a stuck bit. However it might be also a problem with addressing. For instance a ram chip might have a singled dead or shorted address line. That would deliver wrong results upon reading as well.
But now with an oscilloscope you can start looking att the 244 enable line
What you are looking for is a square wave of some kinds when CPU wants to write to gate array. At best you solder a button from reset line to gnd to be able to reset the machine while having the oscillator running.
You can also check the other signals from gate array like crtc clock , ras and cas signals. They all need to form nice square waves with 5 volts. Any of those signals going to anything other than 0v or 5V is an indicator for a problem. Also any of those lines permanently stuck to 5V or 0v is as well a problem.
Quote from: McArti0 on 14:04, 30 April 24Quote from: SerErris on 12:46, 30 April 24I am not sure I do understand what you mean by stopping GA, GA will only stop if you stop the clock.
Take a look at the illustration in post no. 100. GA stops itself due to the Lo signal on pin 1 of 244. This freezes the clock generator. I'll never believe you could be as blind as Bryce It's not like you.
I am not sure what excuse you have to offend people here.
Neither is Bryce blind, nor is me. You might be a genius that we have been missing here all the time, but from what you are writing, it is not the case.
The signal on pin one of 244 is normally generate by the gate array, which means that pin23 of the gate array is an output. It will not react to whatever you do to the line. The worst case if you ground that line is, that you will create a direct short and damage the output buffer in the GA.
The gate array does not stop if you pull pin 23 low. It will actually not result in any change of the ga operation whatever you do to pin 23 of the gate array or pin1 of the 244.
You may want to read some of the available documentation.
Quote from: SerErris on 16:50, 30 April 24Neither is Bryce blind, nor is me. You might be a genius that we have been missing here all the time, but from what you are writing, it is not the case.
The signal on pin one of 244 is normally generate by the gate array, which means that pin23 of the gate array is an output. It will not react to whatever you do to the line.
Genius McArti for BLINDs.
ForBlind.png
Quote from: McArti0 on 13:58, 30 April 24Quote from: Bryce on 13:35, 30 April 24I'm starting to think that this is some sort of never ending April fools joke..
Then prove to him that his memory is faulty. So far I have found out that the 3rd byte of the Rom has the value 61 instead of 89. If at all he did it well with 3 people. For now, it turns out that z80 writes to frames. And that the ram has the correct contents. Maybe you could stop being mean and start helping.
I don't have proof, I have experience. I've repaired 100's of CPC's over many, many years and despite owning an armada of expensive test equipment, if I get a start screen that looks like that I swap the RAM after just a few basic voltage checks. 99% of the time it fixes the problem. Time is more expensive than a set of 4164's.
To put it another way: If it looks like a duck and quacks like a duck, it's most likely a duck.
I am being helpful. I'm passing my experience on, and my experience says swap the RAMs.
Bryce.
And the client wants proof without premonitions, because it's NOT HIS CPC.
He doesn't have the tools, so I help him the way he is capable of doing.
And don't insult my work.
Quote from: Bryce on 19:20, 30 April 24To put it another way: If it looks like a duck and quacks like a duck, it's most likely a duck.
If it looks like a duck and quacks like a duck, sometimes it's not a duck.
Quote from: McArti0 on 13:58, 30 April 24. Maybe you could stop being mean and start helping.
He did. He said very early in this process that it's most likely the RAM. Even if it's not the RAM there are only 6 more ICs that can be the culprit. Even if he would have swapped all relevant ICs it would have been maybe €20 and max 2h of work in total.
Quote from: eto on 20:18, 30 April 24Even if he would have swapped all relevant ICs it would have been maybe €20 and max 2h of work in total.
Ask mmldrm how many times he has desoldered an integrated circuit in other equipment in his life. ::)
@McArti0 , I really don't care whether you're right or wrong. Be careful before insulting people next time.
Quote from: Gryzor on 07:54, 01 May 24Be careful before insulting people next time.
Quote from: Bryce on 13:35, 30 April 24I'm starting to think that this is some sort of never ending April fools joke.
Way to keep it up.
Quote from: eto on 20:18, 30 April 24Even if he would have swapped all relevant ICs it would have been maybe €20 and max 2h of work in total.
If you have sources for the ICs (UK would be best for me) then I would love to know them.
The 64Kb of RAM that I bought was around £10, and so far I only saw a 40031 for £18 on eBay.
It would be good to know the best websites for CPC parts.
Quote from: mmldrm on 08:58, 01 May 24If you have sources for the ICs (UK would be best for me) then I would love to know them.
The 64Kb of RAM that I bought was around £10, and so far I only saw a 40031 for £18 on eBay.
It would be good to know the best websites for CPC parts.
I did not factor in the HAL as it is almost certainly not your problem. Especially since you already bridged it (after message #6 in this thread) and got no change. So even if it would be damaged, something else would be too. But just to be sure: bridge the HAL and replace the RAM, then you can be 100% sure that it's not the cause and once your computer works (don't forget to change the bridge so you can test the second RAM bank too), put in the HAL again and see if it works.
In case this HAL is the problem, you don't need the original. You can replace it by an ATF16V8. Not sure what it costs in the UK but here I can get it for about 2€ per piece in my favorite shop - or 5 pieces for 4-6€ from China. Programming is easy if you have an Eprom programmer or access to one. Even if not, I would be surprised if there is not someone on the forum who could program it for you.
There are no websites for CPC parts unfortunately. But that is no surprise as almost all ICs are standard ICs that can be sourced easily on Ebay or AliExpress for cheap. E.g. all the other 6 ICs that can be the culprit in your case are usually below 1€ per piece.
@McArti0 It's obvious that you have thought about what you suggest - however it's not clear what exactly you want to achieve with your ideas - and if they are correct or not. It would be good if you could not only write what test to perform, but also what purpose this test has. Without that context it's impossible to distinguish if your test is useless - or genius.
E.g. what do you achieve with the connection to CE of IC115?
@eto 1) iorq to Ready/wait (better by diode)
2) start cpc
Z80 executes LD BC,#7F89. OUT(C),C
OUT Produces iorq which does an eternal wait.
3) GA reads the number 89h a million times a second.
4) When the first clock inverter is blocking by HI or Lo. The clock stops working.
Attention. The old mainboard has the first inverter on pin 5 of the 74hcu04 but the new one on pin 13.
5) When connected pin 1. 244 to pin 5 (13) of the first clock inverter, the clock will stand and GA will be stopped in its tracks. If we draw Lo on pin 1 244, we have stopped CPC from 89h on the z80 side and gate 244 is open. (Of course, the diode helps connect only the LO state to the clock)
6) You can easily check whether the number 89h is also on the GA side on the date lines.
Quote from: McArti0 on 13:26, 01 May 241) iorq to Ready/wait (better by diode)
what will that do?
@etoDuring the wait line test, the TW z80 encounters a low state from IORQ in each clock. Therefore, the write execution of the first OUT instruction never completes.
How to test 74LS373 without desoldering.
The pictures show the procedure for stopping CPC on a single read from the internal RAM memory. at the end we have the state of open 74LS373. RAMRD and READY are low.
ORG #0638
LD HL,#B100
LD DE,#B101
LD BC,#07F9
LD (HL),A
.L0642
LDIR <----- here we are stoping CPC
By setting the data pins of the internal bus, you can check whether these states pass to the external bus side. RAM pin 14 set to Lo (grounded) should show the same on pin 2.
If you are surprised by the polarization of diodes, know that I am surprised too.
(WRONG IMAGES REMOVED UPON REQUEST)
Where is catode?
my multimeter works according to alternative physics. It's strange that the NPN transistor shows correctly.
Back to real physics, the previous pictures have incorrectly drawn diodes.
Quote from: McArti0 on 23:56, 02 May 24How to test 74LS373 without desoldering.
The pictures show the procedure for stopping CPC on a single read from the internal RAM memory. at the end we have the state of open 74LS373. RAMRD and READY are low.
ORG #0638
LD HL,#B100
LD DE,#B101
LD BC,#07F9
LD (HL),A
.L0642
LDIR <----- here we are stoping CPC
By setting the data pins of the internal bus, you can check whether these states pass to the external bus side. RAM pin 14 set to Lo (grounded) should show the same on pin 2.
If you are surprised by the polarization of diodes, know that I am surprised too.
Just out of interest. Have you actually tried to do this setup in reality and confirmed it works? Or is this all based on theory?
The reason I ask, is that connecting anything at all directly to a crystal pin will usually dis-balance the capacitance enough so that
the clock will either not start at all or will be unstable.
Bryce.
Quote from: Bryce on 10:25, 03 May 24Just out of interest. Have you actually tried to do this setup in reality and confirmed it works? Or is this all based on theory?
I did exactly that yesterday. Everything I write in this thread is tested (on real CPC16128)
Quote from: Bryce on 10:25, 03 May 24The reason I ask, is that connecting anything at all directly to a crystal pin will usually dis-balance the capacitance enough so that
the clock will either not start at all or will be unstable.
And that's how it should work. GA is supposed to permanently turn off the clock and thereby freeze itself in a specific state.
Quote from: McArti0 on 10:32, 03 May 24Quote from: Bryce on 10:25, 03 May 24Just out of interest. Have you actually tried to do this setup in reality and confirmed it works? Or is this all based on theory?
I did exactly that yesterday. Everything I write in this thread is tested (on real CPC16128)
Quote from: Bryce on 10:25, 03 May 24The reason I ask, is that connecting anything at all directly to a crystal pin will usually dis-balance the capacitance enough so that
the clock will either not start at all or will be unstable.
And that's how it should work. GA is supposed to permanently turn off the clock and thereby freeze itself in a specific state.
Well that's at least good to know. However, there's a reason why those capacitors and the 74 are extremely close to the crystal with very short traces. Just the additional capacitance and inductance of the extra wire is enough to disturb the clock. If the jumper wire is lying across the PCB and (as is the case here), the user doesn't have a scope to confirm the clock signal, then they have no way of knowing if their test is running as expected or not and the results can't be trusted.
Bryce.
Quote from: Bryce on 10:41, 03 May 24If the jumper wire is lying across the PCB and (as is the case here), the user doesn't have a scope to confirm the clock signal, then they have no way of knowing if their test is running as expected or not and the results can't be trusted.
I do have an oscilloscope now! And pin grabbers/clips should arrive today.
I know my kit was woefully inadequate initially (and still isn't ideal now) but hourly these will provide more reliable test results.
Thanks again for putting up with this situation.
@Bryce.
Notice that the clock inverter gets a heavy zero at the input - it has no chance of ticking even once. The freeze is sure and strong.On the address lines there is &B100 or &B101. I use rectifier diodes so they have a 60ns propagation time, they may not stop at the first LDIR address.
Quote from: Bryce on 10:41, 03 May 24no way of knowing if their test is running as expected or not and the results can't be trusted.
Test is Pin 24 z80 WAIT is permanent Lo after Step 3
Test 74LS373 by spider hardware ;D
Quote from: McArti0 on 11:01, 03 May 24@Bryce.
Notice that the clock inverter gets a heavy zero at the input - it has no chance of ticking even once. The freeze is sure and strong.On the address lines there is &B100 or &B101. I use rectifier diodes so they have a 60ns propagation time, they may not stop at the first LDIR address.
Hi McArti0,
you may have misunderstood what I meant (or I explained it badly?).
Yes, the clock will definitely stop as planned with your setup, that's certain. My query was whether it runs properly before that point
and it seems you've just answered that question yourself with the screenshot. If you take a look at the diagram, the clock glitched
a few cycles earlier. So the clock was unstable and it can't be said with certainty that the CPU acted correctly from that point onwards
and the entire system was in an unknown state there after.
Bryce.
Quote from: Bryce on 11:54, 04 May 24it can't be said with certainty that the CPU acted correctly from that point onwards
and the entire system was in an unknown state there after.
This diagram is not mine. It is borrowed from Bread80. 8)
https://bread80.com/2021/06/03/understanding-the-amstrad-cpc-video-ram-and-gate-array-subsystem/
Fig.5
I understand what you are talking about, but for now it is not important to us because even if I have disturbed the program code, for now we want GA to stop with READY Lo and RAMRD Hi or Lo. And that's exactly what's happening. even the address is correct. You can statically check whether 373 is open and output responds to input and 373 save state. The next test will be to catch 373 in latch. i.e. the first RAM reading, READY Lo, RAMRD Lo without steps, using an external OR gate made on the 74LS02. the reading should be 00h.
Quote from: Bryce on 11:54, 04 May 24the clock glitched
a few cycles earlier. So the clock was unstable
I tried this. My tests show that the clock is only wrong when touching the contacts with a bare hand and rubbing the wires against each other when the clock is active. there is no interference from the wires as the antenna MW radio.
never connect the pins this way.
I pushed the IORQ pin apart and now it doesn't connect to the socket without it.
Any update? Did the RAM ever arrive and get installed?
Bryce.
We are wating for 74LS02 and Breadboard for test 373. We have diods to test 244. We are waiting for some time.
mmldrm is reading z80 manual Fig 4-8.
Quote from: Bryce on 12:06, 14 May 24Did the RAM ever arrive and get installed?
If mmldrm fixes this CPC, he will have to give it back. Would you give away your first CPC in your life? ;D Besides, here mmldrm learns digital electronics.
By the time it's fixed (when he gets around to replacing the RAM), he'll have spent so much money and time, that he could have bought his own one! :D
Bryce.
@Bryce He started by buying an oscilloscope, I'm afraid to ask what kind, but I think it was RIGOL, so the CPC costs went first. ;D
Rigol is a very good mid-range brand. I'd highly recommend them for hobby electronics. I use some of their devices too.
(but I already had my CPC when I bought them :) ).
Bryce.
A new memory test occurred to me.
RAMDIS HI, ROMDIS HI (of course after plugin wired version PAL)
and grounded D5,D4,D1.
Z80 always takes the CDh,CDh,CDh,CDh... i.e. CALL #CDCD (push to stack CDCD+3=CDD0)
Whole of memory fills by CDD0
1 1 0 0 1 1 0 1
1 1 0 1 0 0 0 0
_________________
1 1 0½ ½½0½
Its test other bits than FF (RST 38h)
I'll check if it works :D
@mmldrmTry one think. Cross Pin21 Z80 RD to other RD like RD expansion port.
Finally the 74LS02 has arrived!
I will try the above tests too, if I understand them. :o
My oscilloscope is a Zoyi ZT-702S: https://zoyitools.com/product/zt-702s/
I found it for £43, and figured it's something I can use for years after this. This is true for the breadboard and components I bought for the tests - I like playing with electronics and would like to do more hardware projects in the future.
If these tests from
@McArti0 can reduce the number of chips that I desolder and replace, I am less likely to destroy the motherboard in the process. Unorthodox, but I'm happy to proceed with caution on this.
Thanks for staying with me this far on the journey. I know there is a variety of opinions on the path we have taken. :)
Ok. Make OR Gate on breadbord. (From 2 NOR gate in 74Ls02). ;)
Quote from: McArti0 on 13:26, 16 May 24Ok. Make OR Gate on breadbord. (From 2 NOR gate in 74Ls02). ;)
I could do this when I was 18. 😅 I'll go read up on it and get back to you. 🫡
Quote from: mmldrm on 12:23, 16 May 24Finally the 74LS02 has arrived!
I will try the above tests too, if I understand them. :o
My oscilloscope is a Zoyi ZT-702S: https://zoyitools.com/product/zt-702s/
I found it for £43, and figured it's something I can use for years after this. This is true for the breadboard and components I bought for the tests - I like playing with electronics and would like to do more hardware projects in the future.
If these tests from @McArti0 can reduce the number of chips that I desolder and replace, I am less likely to destroy the motherboard in the process. Unorthodox, but I'm happy to proceed with caution on this.
Thanks for staying with me this far on the journey. I know there is a variety of opinions on the path we have taken. :)
The scope is ok for basic "signal present" checking, but only up to about 3MHz for proper diagnosis. That's a bit on the low side even for retro computer repair.
Bryce.
Quote from: Bryce on 14:18, 16 May 24Bryce.
Ouch. That would be a bit disappointing. It's the one suggested on page 2 of this thread.
The spec shows 10MHz. Am I misunderstanding it? (Quite possible.)
48Msps :-X
Although zt-703 and ... I dreamed about Rigol
https://bread80.com/2021/06/03/understanding-the-amstrad-cpc-video-ram-and-gate-array-subsystem/
Fig.7
See when n244EN and nCPU ADDR are LOW for the first time together.
Thanks. Ok I got the 74LS02 set up on the breadboard. (Ignore 74HC02 - TinkerCAD had no 74LS02.)
Where is nCPU ADDR on the Z80? Not sure where to connect that.
Let me know if
1. IC118 should be replaced (i.e. not wires)?
2. should Z80 pin20 and pin24 still be connected?
Quote from: mmldrm on 14:29, 16 May 24Quote from: Bryce on 14:18, 16 May 24Bryce.
Ouch. That would be a bit disappointing. It's the one suggested on page 2 of this thread.
The spec shows 10MHz. Am I misunderstanding it? (Quite possible.)
That is what you can get for that price and it is on the low end of usable, however still massively better than a multimeter for digital cirquits. You can actually see what the lines are doing (up, down etc). And exact timing (e.g. fractions of a µs) is not important here.
Also the next better suited would be a proper oscilloscope like the Rigol 1054z (or any other entry level brand and model). But you know, it is even used for 200-300€ .. So I did not recommend it here, but that would be actually my normal recommendation to get into an Oscilloscope.
Nowadays you can get the new Rigol DHO800 series, with the DHO804 the cheapest option. That goes for ~400 Euros. This is a huge upstep with 12bit resolution vs 8bit of the old Rigol and 100Mhz vs. 50Mhz.
Quote from: mmldrm on 09:21, 18 May 24Thanks. Ok I got the 74LS02 set up on the breadboard. (Ignore 74HC02 - TinkerCAD had no 74LS02.)
Where is nCPU ADDR on the Z80? Not sure where to connect that.
Let me know if
1. IC118 should be replaced (i.e. not wires)?
2. should Z80 pin20 and pin24 still be connected?
Have You DM74 or SN74?
/CPU (nCPU ADDR) Pin14 GA 40010 or Pin2 LS153 or Pin15 AY3 or Pin11 PAL 40030/1
1) wires is better.
2) NO. Now the Force is with us.
When CPC stops on address lines A15-A0 you will have address #7F89
Z80 Pins 5,4,3,2,1,40,39... is 30
Next on Pins
0,1,1,1, 1,
1,1,1, 1,0,0,0, 1,0,0,1
RD Hi, WR Lo, IORQ Lo, nCPU Lo, Pin1 244 Lo.
You can tests D0-D7 Pin2-Pin14 RAM chips
Quote from: mmldrm on 14:29, 16 May 24Quote from: Bryce on 14:18, 16 May 24Bryce.
Ouch. That would be a bit disappointing. It's the one suggested on page 2 of this thread.
The spec shows 10MHz. Am I misunderstanding it? (Quite possible.)
The quick explanation is that the quoted 10MHz is for a sinewave signal. To measure and display a squarewave the bandwidth of the scope needs to be around 5 times the frequency of the signal. I won't go into further details here, because I wrote a more detailed article about this a while ago, which you can read here: https://www.cpcwiki.eu/forum/amstrad-cpc-hardware/understanding-retro-electronics/msg144677/#msg144677
However, as
@SerErris quite rightly mentioned above, a 50MHz Scope will cost at least €200, even for a simple budget device.
Bryce.
Quote from: SerErris on 12:05, 18 May 24That is what you can get for that price and it is on the low end of usable, however still massively better than a multimeter for digital cirquits. You can actually see what the lines are doing (up, down etc). And exact timing (e.g. fractions of a µs) is not important here.
Also the next better suited would be a proper oscilloscope like the Rigol 1054z (or any other entry level brand and model). But you know, it is even used for 200-300€ .. So I did not recommend it here, but that would be actually my normal recommendation to get into an Oscilloscope.
Nowadays you can get the new Rigol DHO800 series, with the DHO804 the cheapest option. That goes for ~400 Euros. This is a huge upstep with 12bit resolution vs 8bit of the old Rigol and 100Mhz vs. 50Mhz.
Hi. I do appreciate the suggestion of the ZT-702S/ZT-703S. I'm just clueless regarding what level of oscilloscope is sufficient for this kind of work.
So I was a little disheartened to be told that the ZT-702S isn't really up to the job of helping to diagnose the issue here, after buying it for that purpose.
I've requested a return on the ZT-702S. Hopefully that will work out and I can spend the money on a Dandanator instead.
I can also almost hear the shouts of "replace the damn RAM chips" and to be honest it does regularly cross my mind to do this, but I would love for a positive confirmation via McArti0'd tests before desoldering anything.
I hope the above doesn't sound too ungrateful! Thanks again for taking the time to reply here.
Quote from: McArti0 on 12:36, 18 May 24Have You DM74 or SN74?
SN74 (SN74LS02N).
Quote from: mmldrm on 13:29, 18 May 24McArti0'd tests before desoldering anything
Let's GO! ;D
Quote from: mmldrm on 13:29, 18 May 24Quote from: SerErris on 12:05, 18 May 24That is what you can get for that price and it is on the low end of usable, however still massively better than a multimeter for digital cirquits. You can actually see what the lines are doing (up, down etc). And exact timing (e.g. fractions of a µs) is not important here.
Also the next better suited would be a proper oscilloscope like the Rigol 1054z (or any other entry level brand and model). But you know, it is even used for 200-300€ .. So I did not recommend it here, but that would be actually my normal recommendation to get into an Oscilloscope.
Nowadays you can get the new Rigol DHO800 series, with the DHO804 the cheapest option. That goes for ~400 Euros. This is a huge upstep with 12bit resolution vs 8bit of the old Rigol and 100Mhz vs. 50Mhz.
Hi. I do appreciate the suggestion of the ZT-702S/ZT-703S. I'm just clueless regarding what level of oscilloscope is sufficient for this kind of work.
So I was a little disheartened to be told that the ZT-702S isn't really up to the job of helping to diagnose the issue here, after buying it for that purpose.
I've requested a return on the ZT-702S. Hopefully that will work out and I can spend the money on a Dandanator instead.
I can also almost hear the shouts of "replace the damn RAM chips" and to be honest it does regularly cross my mind to do this, but I would love for a positive confirmation via McArti0'd tests before desoldering anything.
I hope the above doesn't sound too ungrateful! Thanks again for taking the time to reply here.
No issue. I won't shout anything ;D . It's your machine and it's your decision how you want to fix it. If you intend always using your Scope where a computer is nearby, there are some very good USB Scopes around, that will give you a lot more bandwidth and all the features for under €200.
You can get a Hantek PC 6074BC (4 Channel 70MHz USB Scope) for €180.
Bryce.
Quote from: McArti0 on 12:36, 18 May 24/CPU (nCPU ADDR) Pin14 GA 40010 or Pin2 LS153 or Pin15 AY3 or Pin11 PAL 40030/1
Thanks, I used pin15 of AY3.
Quote from: McArti0 on 12:36, 18 May 24When CPC stops on address lines A15-A0 you will have address #7F89
Z80 Pins 5,4,3,2,1,40,39... is 30
Next on Pins
0,1,1,1, 1,
1,1,1, 1,0,0,0, 1,0,0,1
I keep getting high on the address lines. A0-A14, all 1. Strangely, A15 was 0. I tried this three times and got the same results.
Not every startup seemed to halt. I saw the white screen and black border occasionally, but presumably this kind of system halting is not exact.
Quote from: McArti0 on 12:36, 18 May 24RD Hi, WR Lo, IORQ Lo, nCPU Lo, Pin1 244 Lo.
I confirmed nCPU Lo and Pin1 244 Lo. Are RD/WR/IORQ all Z80 pins? Or expansion port? Or... are those the same thing anyway? I guess I can confirm all of them tomorrow.
I will also check the wiring again, but I think it is all correct. Please let me know if my breadboard diagram above is incorrect.
Quote from: Bryce on 22:48, 18 May 24No issue. I won't shout anything ;D . It's your machine and it's your decision how you want to fix it. If you intend always using your Scope where a computer is nearby, there are some very good USB Scopes around, that will give you a lot more bandwidth and all the features for under €200.
You can get a Hantek PC 6074BC (4 Channel 70MHz USB Scope) for €180.
Thanks for the info, but even €180 is hugely more than I can justify. Maybe if I survive this process and fix a few Amstrad, Spectrum, or Commodore computers then I will treat myself to a Hantek USB scope. ;D
Quote from: mmldrm on 22:57, 19 May 24Are RD/WR/IORQ all Z80 pins?
You are no longer allowed to ask such questions.
https://www.cpcwiki.eu/imgs/4/4a/CPC6128_Schematic.png
Signal name is constant.
Quote from: mmldrm on 22:57, 19 May 24Please let me know if my breadboard diagram above is incorrect.
Have too many joins. +5V should join one wire near Pin14, one hole above. Groung one wire near Pin7,one hole under.
Please, show me photo of all setup.
Quote from: McArti0 on 23:24, 19 May 24Quote from: mmldrm on 22:57, 19 May 24Are RD/WR/IORQ all Z80 pins?
You are no longer allowed to ask such questions.
https://www.cpcwiki.eu/imgs/4/4a/CPC6128_Schematic.png
Signal name is constant.
😆 Good answer. I think I am still learning that the state is constant across the board. No co-processors, etc. I sometimes found pins with the same name but on different ICs, and not sure which is the 'right' one. Now I know that there is no wrong answer. ;D
Quote from: McArti0 on 23:24, 19 May 24Have too many joins. +5V should join one wire near Pin14, one hole above. Groung one wire near Pin7,one hole under.
I found my mistake. I connected pins 2->3 and 3->4, making them all the same input. It should have been 2->4 and 3->4. :picard: Will try again with it corrected.
Quote from: McArti0 on 23:24, 19 May 24Please, show me photo of all setup.
It seems difficult to show everything in enough detail, but I will try (in a separate reply).
Please ignore 2x green hooks on Z80. Those are from our previous test and not connected to anything.
All clips are connected when I power on, including the pin1 LS244->R143.
Occasionally on startup I get the white screen and black border, like the Z80 didn't halt. Other times the screen remains 'dead', so I continued.
I checked Z80 IORQ/RD/WR, LS244, and R143 for hi/lo, and they were all correct when reading A0-A15. If not correct values then I powered off, waited 5s, and powered back on to try again.
I took GND/+VCC from IC108 because the pins were easiest to connect hooks to.
A0-A15 all hi. :( Still possible that I am doing something wrong, but it feels like I checked everything twice...
Check whether when you short both 74LS02 inputs, pin 5,6 to ground, the CPC freezes.
You can add a second gate to the output by connecting PIN13 to Pin1 and pin11,12 to the group of pin2,3,4. 74LS02
If this does not work stably, we will add support with a diode pin 20 to pin 24 as before.
Quote from: mmldrm on 20:44, 20 May 24found my mistake. I connected pins 2->3 and 3->4, making them all the same input. It should have been 2->4 and 3->4. Will try again with it corrected.
It really doesn't matter at all.
Quote from: McArti0 on 11:40, 21 May 24Quote from: mmldrm on 20:44, 20 May 24found my mistake. I connected pins 2->3 and 3->4, making them all the same input. It should have been 2->4 and 3->4. Will try again with it corrected.
It really doesn't matter at all.
Only now do I realise that I was visualising current going from left-to-right across the 74LS02 diagram and not from right-to-left. :picard:
Quote from: McArti0 on 11:36, 21 May 24Check whether when you short both 74LS02 inputs, pin 5,6 to ground, the CPC freezes.
You can add a second gate to the output by connecting PIN13 to Pin1 and pin11,12 to the group of pin2,3,4. 74LS02
If this does not work stably, we will add support with a diode pin 20 to pin 24 as before.
Thanks I will try with the second gate and maybe the diode. Probably tomorrow - too late tonight. :(
Also, I got a refund for the oscilloscope. So if I manage to break you with my lack of knowledge/skills/tools, I can order that Dandanator Mini. ;D
I just found something interesting on the LS02. Two lo inputs (0v and 0.3v) are giving me hi output. Pins 5 and 6 input, pin 4 output.
I checked the tolerances for my SN74LS02N here (https://www.ti.com/lit/ds/symlink/sn74ls02.pdf?ts=1716446507160&ref_url=https%3A%2F%2Fhu.mouser.com%2F). Max lo is 0.8v, min hi is 2v.
Not sure if this is causing the problem, but it sure surprised me.
(I forgot to add, the second NOR gate was on pins 11,12,13 when I measured this.)
Very good! You're starting. understand what's going on here. But what you are writing about is the correct operation of what we want to build. When Pin1 244 and nCPU are low, only then we want the diode and pin1 LS02 to be low.
Maybe you should add a capacitor between pin 14 and pin 7, i.e. on the LS02 power supply .
It worked!! :o
Quote from: McArti0 on 12:36, 18 May 24When CPC stops on address lines A15-A0 you will have address #7F89
Z80 Pins 5,4,3,2,1,40,39... is 30
Next on Pins
0,1,1,1, 1,
1,1,1, 1,0,0,0, 1,0,0,1
RD Hi, WR Lo, IORQ Lo, nCPU Lo, Pin1 244 Lo.
You can tests D0-D7 Pin2-Pin14 RAM chips
Here are the test results.
Apologies for use of 0, 1, and 3. The '3' readings were unexpected and I wasn't sure how to represent them in the table.
Notes:
0 = lo (0-0.2V)
1 = hi (4-5V)
3 = 3.67V except pin14 which was always 3.23V
---------------------pins---------------------
IC. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
127 0 3 1 0 3 3 3 1 3 1 3 3 3 1 1 0
128 0 3 1 0 3 3 3 1 3 1 3 3 3 1 1 0
129€ 0 3 1 0 3 3 3 1 3 1 3 3 3 1 1 0
130 0 3 1 0 3 3 3 1 3 1 3 3 3 1 1 0
131 0 3 1 0 3 3 3 1 3 1 3 3 3 1 1 0
132 0 3 1 0 3 3 3 1 3 1 3 3 3 1 1 0
133 0 3 1 0 3 3 3 1 3 1 3 3 3 1 1 0
134 0 3 1 0 3 3 3 1 3 1 3 3 3 1 1 0I was working from IC134 up to IC127. At the point marked (€) the screen suddenly brightened. I retested other pins and the 3.2V-3.6V pins became 4V when retested. I didn't have time to test all of them before the voltage dropped again. :(
After noting some of these new values, just as I touched ground again with multimeter probe (and nothing on +ve probe), screen dulled and back to 3.2V-3.6V. Probably a coincidence, but maybe not.
I wondered if this shows a power problem. I checked some random pins on other ICs and found other 3.67V values.
Hmm, now I checked the data sheet for for the 4164 chips and minimum hi value seems to 2.4V, so maybe this isn't the "smoking gun" that I originally thought it was...
Anyway, lots of consistent values across the RAM chips.
Quote from: McArti0 on 12:36, 18 May 24Z80 Pins 5,4,3,2,1,40,39... is 30
Next on Pins
0,1,1,1, 1,
1,1,1, 1,0,0,0, 1,0,0,1
:-X :laugh:
Ps. The address lines in memory change 4 million times per second. So there's not much to measure on them.
Quote from: McArti0 on 07:13, 24 May 24Quote from: McArti0 on 12:36, 18 May 24Z80 Pins 5,4,3,2,1,40,39... is 30
Next on Pins
0,1,1,1, 1,
1,1,1, 1,0,0,0, 1,0,0,1
:-X :laugh:
I'm 99% sure these were the values I saw, because I only checked the memory ICs after passing the CPU A0-A16 address values 'test'. But I'll do it again today to confirm.
Quote from: McArti0 on 07:13, 24 May 24Ps. The address lines in memory change 4 million times per second. So there's not much to measure on them.
Oops, I thought
everything was halted during this test. :doh:
Have you seen the picture frame? Because that means it wasn't stopped.
No picture frame. Just a blank screen.
It looked like static, but with horizontal lines. Then I connected LS244-R143 and get a fixed bright screen. (White? Difficult to tell on the monitor.)
Tried again today. No static screen or white screen, just blank/dark screen the whole time.
1. Switched on (with all connections as described previously, except LS244-R143 is disconnected)
2. Connect LS244-R143
3. Check Z80 pins 20/21/22 are lo/hi/lo
4. Check Z80 address pins
5. Measure RAM values
So Z80 address (A0-A15) are correct.
10010001 11111110
DRAM values are the same for all 8 ICs.
Today I measured DRAM pins 1-16 in correct order. Yesterday I accidentally measured 1-8, then 16-9, and had to rearrange the table. Maybe I made a mistake there, because now Pin13 is lo, not hi.
---------------------pins---------------------
IC 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
127 0 3 1 0 3 3 3 1 3 1 3 3 0 1 1 0
128 0 3 1 0 3 3 3 1 3 1 3 3 0 1 1 0
129 0 3 1 0 3 3 3 1 3 1 3 3 0 1 1 0
130 0 3 1 0 3 3 3 1 3 1 3 3 0 1 1 0
131 0 3 1 0 3 3 3 1 3 1 3 3 0 1 1 0
132 0 3 1 0 3 3 3 1 3 1 3 3 0 1 1 0
133 0 3 1 0 3 3 3 1 3 1 3 3 0 1 1 0
134 0 3 1 0 3 3 3 1 3 1 3 3 0 1 1 0
This is OUT #7F89 , .... #FF
D2 lines shows that z80 does not break through with value #89
Check pin1 LS373 should be Hi.
Check R160,R161 should be Hi
Ps. You can all connect before turn on cpc
Pin1 LS373 is Hi
R160 is Lo
R161 is Hi
I do still have wires replacing IC118, if that makes a difference.
Check MREQ should be Hi
R160 is Lo , very bad. Check pin16 40010 should be Hi and wire 5v to pin16 in socket for PAL 40030 (i belive is not present)
MREQ = hi
pin16 of 40010 = hi
Correct, there is no IC below the 40010.
Should I solder a wire from +5v to the pin16 point on the board?
I'm understanding... just about... ;)
https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg238062/#msg238062
Pin7 to pin16 or pin17. Pin20 5v to empty 16 or 17.
Quote from: McArti0 on 10:59, 24 May 24https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg238062/#msg238062
Pin7 to pin16 or pin17. Pin20 5v to empty 16 or 17.
With this setup, both R160 and R161 are hi.
(Makes me want to double check the previous readings...)
But only D2 and D14 . You should see #89 1000 1001
Ok I got 1000 1001, on pin2 and pin14, both RAM banks (IC127-134, and IC119-126).
Just confirming - this is if I read from bottom to top on the board. i.e. IC134->IC127, and IC126->IC119. (I guess that is correct - lowest bit is on lowest IC number.)
https://www.cpcwiki.eu/imgs/4/4a/CPC6128_Schematic.png
;D
Now when you check pin14 , you quick grounded by diode (catode to ground) pin2. Multimeter should show low on pin14.
Quote from: McArti0 on 12:53, 24 May 24Now when you check pin14 , you quick grounded by diode (catode to ground) pin2. Multimeter should show low on pin14.
Every time I try, I still get 10001001 on pin2 and pin14. Is that unexpected?
I even tried pin2 D0-D7, pin14 D0-D7, and then checked pin2 D0-D7 again... thinking maybe checking/grounding pin14 could affect values. But still 10001001.
Maybe Friday at 5pm is a bad time for me to understand why multimeter should be lo on pin14. ???
244 is like eight pipes.
When Pin1 244 is Lo. pipes are open.
When you grounded RAMs pin2, this ground flow by open 244 to Pin14 RAM.
If You are checking Pin14 by multimeter and grounding Pin2 by second hand then you are seeing this ground on multimeter.
Quote from: McArti0 on 19:59, 24 May 24244 is like eight pipes.
When Pin1 244 is Lo. pipes are open.
When you grounded RAMs pin2, this ground flow by open 244 to Pin14 RAM.
If You are checking Pin14 by multimeter and grounding Pin2 by second hand then you are seeing this ground on multimeter.
Thanks, now I understand how checking pin2 first could ground pin14.
Although I did only ever test one pin at a time. Black multimeter probe on ground, red probe on pinX. No second hand, unless you mean the red and black multimeter probes.
Silly question... what should I do next?
Quote from: mmldrm on 12:40, 26 May 24Black multimeter probe on ground,
By crocodile, not in hand.
Now.
If You are checking Pin14 by multimeter and grounding Pin2 by second hand then you are seeing this ground on multimeter.
Quote from: McArti0 on 15:23, 26 May 24Now.
If You are checking Pin14 by multimeter and grounding Pin2 by second hand then you are seeing this ground on multimeter.
Ok, I grounded pin2 and checked pin14 on IC127-IC134.
All pin14s were lo (~0.05V) except IC127 which remained hi (~4.95V). Have we found a bad RAM chip?
@mmldrmTo sum up:
Pin1 244: Lo
CAS0,CAS1 (R160,R161) Hi
D0 Pin2 IC127 Lo (Grounded) and Pin14 Hi ? ? ? ?
This would mean that 244 is damaged.
Check it again the same thing in another place: Pin2 244 grounded and Pin18 244 multimeter
Quote from: McArti0 on 19:42, 27 May 24mmldrm (https://www.cpcwiki.eu/forum/index.php?action=profile;u=5900)
To sum up:
Pin1 244: Lo
CAS0,CAS1 (R160,R161) Hi
D0 Pin2 IC127 Lo (Grounded) and Pin14 Hi ? ? ? ?
This would mean that 244 is damaged.
Or IC127 is driving Pin2 so much (shorted to +5V) that it is overwriting Pin18 of 244, which will sink the current.
If IC127 is the culprit, you should see a permanent high signal if you let the computer run normally.
The problem is, that you cannot really decide which on of both it is (I would still bet on the IC127) as both can drive the line.
If you measure your situation above, then you should see the same high signal on IC127, IC119 (both Pin14) and IC115 Pin18 as well as IC114 PIN3 and GA Pin35. All those ICs are directly connected and should have the same voltage. If they do not have the same voltage, you are measuring something wrong.
you could test if the RAM is the culprit, if you bring pin 1 of IC115 to high. Then the output of LS244 is high impedance (actually no output) and whatever the RAM chips are doing is then what you will see on Pin14 of IC127.
But still it cannot be that you only see high on a single of all those chips. It is all chips high or any other voltage, but it is not IC127 high and IC119 low.
Quote from: SerErris on 08:59, 28 May 24Or IC127 is driving Pin2 so much (shorted to +5V) that it is overwriting Pin18 of 244, which will sink the current.
But in another dynamic scenario where GA works, pin14 is 2.5v so it is often zero there.
Look here...
https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg238645/#msg238645
Ok we have test hardware. Test.
Pin5 LS02 to internal D0- Pin14 Ram ic 127.
Pin6 LS02 to pin1 244.
Turn ON cpc.
If cpc stoped then check D0 lo, cas0 hi ,cas1 hi, iorq lo, pin 1 244 lo, address pin 5-1,40-30 z80.
Quote from: McArti0 on 19:42, 27 May 24@mmldrm
To sum up:
Pin1 244: Lo
CAS0,CAS1 (R160,R161) Hi
D0 Pin2 IC127 Lo (Grounded) and Pin14 Hi ? ? ? ?
This would mean that 244 is damaged.
Check it again the same thing in another place: Pin2 244 grounded and Pin18 244 multimeter
When Pin2 LS244 is grounded, Pin18 LS244 is lo
Quote from: SerErris on 08:59, 28 May 24If you measure your situation above, then you should see the same high signal on IC127, IC119 (both Pin14) and IC115 Pin18 as well as IC114 PIN3 and GA Pin35. All those ICs are directly connected and should have the same voltage. If they do not have the same voltage, you are measuring something wrong.
Thanks. I have checked all of these, with pin2 of IC127 grounded, and can confirm that they are all high.
IC127 pin14
IC119 pin14
IC115 pin18
IC114 pin3
GA pin35
Quote from: McArti0 on 10:55, 28 May 24Ok we have test hardware. Test.
Pin5 LS02 to internal D0- Pin14 Ram ic 127.
Pin6 LS02 to pin1 244.
Turn ON cpc.
If cpc stoped then check D0 lo, cas0 hi ,cas1 hi, iorq lo, pin 1 244 lo, address pin 5-1,40-30 z80.
With LS02 pin5+pin6 connected to IC127/LS244, the CPC does not stop. I tried 3-4 times.
LS02 pin5 was already connected to pin1@LS244
LS02 pin6 was already connected to pin15@AY3 (/CPU)
I have attached the diagram of how the LS02 is set up right now. I just want to make sure everything is correct, before I try the new connections you suggested.
Quote from: mmldrm on 13:10, 28 May 24When Pin2 LS244 is grounded, Pin18 LS244 is lo
There is no problem. This is your measurement error.
Pin18 244 is the some wire as pin14 ram ic 127.
Pin2 244 is the some wire as pin2 ram ic 127.
Quote from: mmldrm on 12:32, 24 May 24Ok I got 1000 1001, on pin2 and pin14, both RAM banks (IC127-134, and IC119-126).
Just confirming - this is if I read from bottom to top on the board. i.e. IC134->IC127, and IC126->IC119. (I guess that is correct - lowest bit is on lowest IC number.)
How do You do that?
How set 0 color to purple.
All ground on z80 side. (Pin2 RAM)
A) Ready/Wait. -----|>|------ IORQ and keep (Connect by diode, catode at IORQ)
B) start cpc or reset.
C) ground D0 and keep , next ground D3 and keep, quick short ground D7 and remove all ground
D) ground D0 and keep, Hi on D6 (best way by serial 4x diodes ) and keep, quick short ground D7 and remove all ground and Hi set.
Background screen should be purple.
But for this to work, the 244 must be operational.
Quote from: McArti0 on 20:45, 28 May 24Quote from: mmldrm on 12:32, 24 May 24Ok I got 1000 1001, on pin2 and pin14, both RAM banks (IC127-134, and IC119-126).
Just confirming - this is if I read from bottom to top on the board. i.e. IC134->IC127, and IC126->IC119. (I guess that is correct - lowest bit is on lowest IC number.)
How do You do that?
I finally went back to retest this. Steps were :QuoteWhen CPC stops on address lines A15-A0 you will have address #7F89
Z80 Pins 5,4,3,2,1,40,39... is 30
Next on Pins
0,1,1,1, 1,
1,1,1, 1,0,0,0, 1,0,0,1
RD Hi, WR Lo, IORQ Lo, nCPU Lo, Pin1 244 Lo.
You can tests D0-D7 Pin2-Pin14 RAM chips
But I must have messed up reporting the results. Maybe I recorded D7-D0. So I tried again today.These are the results on pin2 and pin14 (and tested twice):IC127 D0 1
IC128 D1 0
IC129 D2 0
IC130 D3 1
IC131 D4 0
IC132 D5 0
IC133 D6 0
IC134 D7 1
Quote from: McArti0 on 08:46, 29 May 24How set 0 color to purple.
All ground on z80 side. (Pin2 RAM)
A) Ready/Wait. -----|>|------ IORQ and keep (Connect by diode, catode at IORQ)
B) start cpc or reset.
C) ground D0 and keep , next ground D3 and keep, quick short ground D7 and remove all ground
D) ground D0 and keep, Hi on D6 (best way by serial 4x diodes ) and keep, quick short ground D7 and remove all ground and Hi set.
Background screen should be purple.
But for this to work, the 244 must be operational.
Do you mean I need to ground pin2 on all RAM chips (IC127-IC134)?
If I need 8 pin clips to ground 8 RAM chips, then I will need to take apart some of the other things we set up, like the LS02. (I only have 10 pin clips. Very cheap to get more, but takes time.)
This sounds like a great test but also a very difficult one to get right. 😅
Quote from: mmldrm on 09:27, 02 June 24When CPC stops on address lines A15-A0 you will have address #7F89
Once you write that you are testing the data, the second time you write that CPC does not stop at nCPU and Lo on pin1 244. We want to test the Data when the CPC stops.
Quote from: mmldrm on 09:35, 02 June 24Do you mean I need to ground pin2 on all RAM chips (IC127-IC
134)?
NO,NO,NO.
https://www.cpcwiki.eu/imgs/4/4a/CPC6128_Schematic.png
Quote from: McArti0 on 08:46, 29 May 24How set 0 color to purple.
All ground on z80 side. (Pin2 RAM)
A) Ready/Wait. -----|>|------ IORQ and keep (Connect by diode, catode at IORQ)
B) start cpc or reset.
C) ground D0 and keep , next ground D3 and keep, quick short ground D7 and remove all ground
D) ground D0 and keep, Hi on D6 (best way by serial 4x diodes ) and keep, quick short ground D7 and remove all ground and Hi set.
Background screen should be purple.
But for this to work, the 244 must be operational.
You are aware, that grounding a data pin will mean that the IC driving that data line will then have a direct short to ground with absolutely no current limiting if it's trying to pull that line high (trying to send a "1") and will most likely destroy the output? It may survive if there's a pull-up resistor or if the short is only for a few seconds, but otherwise you've fried a chip.
Bryce.
Quote from: Bryce on 14:32, 02 June 24You are aware, that grounding a data pin will mean that the IC driving that data line will then have a direct short to ground with absolutely no current limiting if it's trying to pull that line high (trying to send a "1") and will most likely destroy the output?
Bryce.
I totally agree. That's why I wrote "best way by serial 4x diodes" 5v - 4x0.6v = ~2.5v.
I checked that the 5v to 0v short circuit current for cmos is 20mA.
Quote from: McArti0 on 16:29, 02 June 24Quote from: Bryce on 14:32, 02 June 24You are aware, that grounding a data pin will mean that the IC driving that data line will then have a direct short to ground with absolutely no current limiting if it's trying to pull that line high (trying to send a "1") and will most likely destroy the output?
Bryce.
I totally agree. That's why I wrote "best way by serial 4x diodes" 5v - 4x0.6v = ~2.5v.
I checked that the 5v to 0v short circuit current for cmos is 20mA.
Ok, but just for my understanding... When he has done all of this, what exactly will this have shown? Or how will it bring him any closer to knowing what the route cause of the issue is? I have to say, I'm struggling at this point to understand the goal of many of these experiments. Testing the 244 is a simple matter of comparing the inputs and outputs when the /E pin goes low and this can be done with any cheap multimeter without any jumpers (in fact back when I didn't have the possibility to own fancy equipment, I used to do this with an LED and a resistor). Also, the chances of a 244 failing are 1000's of times less likely than a RAM failing.
Bryce.
@BryceTest whether memory works. And memory writes and reads. Test if 244 works. Test whether you can set the color in GA. RAM works but may be unstable, then mmldrm will buy dandanator mini, which we have agreed on for a long time by PM. But we must be sure that GA shows colors. We should be done with the 373 test, but mmldrm doesn't have control over the 74LS02 gates yet. His increase in experience is also important here.
Now the interesting thing is that z80 writes memory, GA reads memory and z80 does not read through 373.
Quote from: McArti0 on 18:14, 02 June 24@Bryce
Test whether memory works. And memory writes and reads. Test if 244 works. Test whether you can set the color in GA. RAM works but may be unstable, then mmldrm will buy dandanator mini, which we have agreed on for a long time by PM. But we must be sure that GA shows colors. We should be done with the 373 test, but mmldrm doesn't have control over the 74LS02 gates yet. His increase in experience is also important here.
Now the interesting thing is that z80 writes memory, GA reads memory and z80 does not read through 373.
So to put it in perspective... If his car breaks down and an experienced mechanic says "It's your fuel pump, they're always failing", he should still start by testing every single other component on the vehicle before he goes and buys a cheap pump? Even if some of those tests could easily result in him damaging something much more expensive on the car?
As I tend to document everything, I've just gone through the 100+ CPC repairs I've done over the last few years. I have never once needed to swap either the 244, nor any other TTL chip, other than a multiplexer once or twice and the keyboard driver.
Bryce.
Quote from: Bryce on 20:23, 02 June 24If his car breaks down
Its NOT his car.
Quote from: Bryce on 20:23, 02 June 24Even if some of those tests could easily result in him damaging something much more expensive on the car?
YES! because z80 is in socket. RAM not.
If I confirm that the ROM contains the correct procedure for using RAM and it does not work because of RAM and not 373, it will solder 256 RAM pins.
Think of it as gathering mmldrm experience and my fun in coming up with tests without an oscilloscope.
CPC stepper.
Stepper can go to next M1, next IORQ, next WR.
At least one crocodile must always hold an diode.
First hold M1 crocodile. turn on CPC. CPU halt at addr 0000
next hold IORQ crocodile to other diode
remove M1 crocodile. CPU halt at WR IORQ after addr 0004
hold M1 crocodile .
remove IORQ crocodile. CPU halt at addr 0005
hold IORQ crocodile .
remove M1 crocodile. CPU halt at WR IORQ after addr 0595
e.t.c.
Quote from: McArti0 on 21:30, 02 June 24Quote from: Bryce on 20:23, 02 June 24If his car breaks down
Its NOT his car.
Quote from: Bryce on 20:23, 02 June 24Even if some of those tests could easily result in him damaging something much more expensive on the car?
YES! because z80 is in socket. RAM not.
If I confirm that the ROM contains the correct procedure for using RAM and it does not work because of RAM and not 373, it will solder 256 RAM pins.
Think of it as gathering mmldrm experience and my fun in coming up with tests without an oscilloscope.
It's not your CPC either and your "fun" is at the expense and risk of someone else's property.
As for learning. mmlmrd has now learnt in detail how NOT to diagnose and repair a CPC.
At the moment, the status of the CPC is less clear than when he started. On page one, it was blatantly clear that the RAM had failed. Good advice would have been to tell him this. During any of these hair-brained experiments he may have touched a wrong pin by mistake and fried an output without knowing. He now has a CPC that may have additional failures and damaged parts and it will be much more complicated to confirm this.
He would have soldered all 256 pins and had a working CPC weeks ago. It would have been less work, less expense and he would have a working CPC now. Instead he still has a non-working CPC, now with possible additional failures. If that's your idea of good repair practise, you also have a lot to learn.
Here's some stuff he, you and others should learn:
1 - Most electronics will have a common weak point. If many other owners had the same issue and the cause was almost always the same, then follow that advice. Any additional poking around is risking damage to parts that are much more difficult to find and/or expensive to fix.
2 - Don't spend time or money testing a part that is cheaper than the equipment needed to test it.
3 - Learn what parts rarely fail and assume they work until you have ruled out the weakest parts.
4 - NEVER inject signals into a circuit at a point that has outputs connected to it. Signal injection is useful for repairing power supplies, audio equipment and test gear. It's not common or good practice to manipulate bits on a running computer.
5 - Electronic repair isn't just about voltages, currents and components. It's about recognising what most likely failed.
Bryce.
Quote from: McArti0 on 21:30, 02 June 24Quote from: Bryce on 20:23, 02 June 24If his car breaks down
Its NOT his car.
Quote from: Bryce on 20:23, 02 June 24Even if some of those tests could easily result in him damaging something much more expensive on the car?
Think of it as gathering mmldrm experience and my fun in coming up with tests without an oscilloscope.
Oh ok, last time I went to the mechanic we had lots of fun tearing the car down instead of using some foam to keep the fuel line in place, and I learned so much in the process!
Quote from: Bryce on 07:59, 03 June 24He now has a CPC that may have additional failures and damaged parts and it will be much more complicated to confirm this.
Bryce.
OMG "realy" ? No risk no fun.
Quote from: Bryce on 07:59, 03 June 244 - NEVER inject signals into a circuit at a point that has outputs connected to it.. It's not common or good practice to manipulate bits on a running computer.
Like all extension RAM for 464. And examples of pin "blocking" in this topic.
Quote from: Bryce on 07:59, 03 June 245 - Electronic repair isn't just about voltages, currents and components. It's about recognising what most likely failed.
Bryce.
Ugly approach to the matter. Blind replacement without diagnosis.
I am very glad that this mainboard is not mechanically damaged by desoldering the frames and there are no traces that have been detached.
Quote from: McArti0 on 09:29, 03 June 24Quote from: Bryce on 07:59, 03 June 24He now has a CPC that may have additional failures and damaged parts and it will be much more complicated to confirm this.
Bryce.
OMG "realy" ? No risk no fun.
Quote from: Bryce on 07:59, 03 June 244 - NEVER inject signals into a circuit at a point that has outputs connected to it.. It's not common or good practice to manipulate bits on a running computer.
Like all extension RAM for 464. And examples of pin "blocking" in this topic.
Quote from: Bryce on 07:59, 03 June 245 - Electronic repair isn't just about voltages, currents and components. It's about recognising what most likely failed.
Bryce.
Ugly approach to the matter. Blind replacement without diagnosis.
I am very glad that this mainboard is not mechanically damaged by desoldering the frames and there are no traces that have been detached.
You totally fail to take into consideration the risk/benefit ratio, as well as the cost (monetary and time) involved.
From personal experience, when it comes to fixing something, I always made it worse when I tried to bite more than I could chew. And I'm willing to bet I'm not the only one.
Quote from: Gryzor on 09:44, 03 June 24You totally fail to take into consideration the risk/benefit ratio, as well as the cost (monetary and time) involved
Please, no demagogy.
Quote from: Gryzor on 09:44, 03 June 24From personal experience, when it comes to fixing something, I always made it worse when I tried to bite more than I could chew. And I'm willing to bet I'm not the only one.
And it's completely the opposite for me. It always ends with something revealing.
Quote from: McArti0 on 10:03, 03 June 24Quote from: Gryzor on 09:44, 03 June 24You totally fail to take into consideration the risk/benefit ratio, as well as the cost (monetary and time) involved
Please, no demagogy.
Quote from: Gryzor on 09:44, 03 June 24From personal experience, when it comes to fixing something, I always made it worse when I tried to bite more than I could chew. And I'm willing to bet I'm not the only one.
And it's completely the opposite for me. It always ends with something revealing.
The bus contention used by external RAM expansions is controlled, not a random wire with no current limiting. And still it's very bad practise.
Demagogy would suggest that you are somehow elite? :D
Having "fun" at the expense of others and suggesting (without warnings) risky experiments to someone without the experience to understand the risks involved is not ok.
mmldrm didn't come asking for fun. He came looking for advice on the best way to repair the CPC and up to now you have brought him no closer to his goal.
Bryce.
Quote from: McArti0 on 10:03, 03 June 24Quote from: Gryzor on 09:44, 03 June 24You totally fail to take into consideration the risk/benefit ratio, as well as the cost (monetary and time) involved
Please, no demagogy.
Ok, I actually laughed on this one.
Quote from: Bryce on 10:10, 03 June 24mmldrm didn't come asking for fun. He came looking for advice on the best way to repair the CPC and up to now you have brought him no closer to his goal.
Unfortunately, he doesn't sit on cpc all the time, he doesn't have much time and he doesn't learn very quickly.
Setting color 0 without the risk of forcing the 4 pins states is even longer.
Quote from: Gryzor on 10:17, 03 June 24Quote from: McArti0 on 10:03, 03 June 24Quote from: Gryzor on 09:44, 03 June 24You totally fail to take into consideration the risk/benefit ratio, as well as the cost (monetary and time) involved
Please, no demagogy.
Ok, I actually laughed on this one.
https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg237939/#msg237939
Are you laughing at this post too?
Quote from: McArti0 on 10:43, 03 June 24Quote from: Gryzor on 10:17, 03 June 24Quote from: McArti0 on 10:03, 03 June 24Quote from: Gryzor on 09:44, 03 June 24You totally fail to take into consideration the risk/benefit ratio, as well as the cost (monetary and time) involved
Please, no demagogy.
Ok, I actually laughed on this one.
https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg237939/#msg237939
Are you laughing at this post too?
No, only laughing at you and your inability to keep up with a discussion.
Quote from: Gryzor on 10:53, 03 June 24No, only laughing at you and your inability to keep up with a discussion.
Ah ok. Greetings for your wife.
Yeah, insulting family of others - that's a first around here. Much as I'll miss the fun, he's banned for a week.
Quote from: McArti0 on 10:39, 03 June 24Quote from: Bryce on 10:10, 03 June 24mmldrm didn't come asking for fun. He came looking for advice on the best way to repair the CPC and up to now you have brought him no closer to his goal.
Unfortunately, he doesn't sit on cpc all the time, he doesn't have much time and he doesn't learn very quickly.
Well at least you took into consideration that he doesn't have much time and helpfully suggested the quickest way to get the CPC working again... :picard:
Also nice that you consider him a slow learner. I'm sure he'll be pleased to hear what you think of him. :picard2:
Bryce.
Quote from: McArti0 on 10:43, 03 June 24Are you laughing at this post too?
How is my post related to that?
It was just to demonstrate that a short on a data-pin will have exactly that effect. And that was my own hardware and my own risk. I would not recommend anyone to do that.
To be honest, the learning out of this thread for someone new to digital electronics and CPC in general is pretty much zero.
Probing things without knowledge on why and what and the background of what you actually do is pretty much confusing and not helping any bit to better understand what is going on.
Setting a Color in the GA with wires is the strangest thing I heard and I am not sure if I would be able to execute this, even knowing how all those thing work. It would be just out of my manipulation capabilities to get this timed correctly and working.
I would still say:
If anyone has no knowhow on any of those topics, you should not attempt to fix it. If you have good soldering skills (which we also do not know), you can attempt to replace the RAM.
However if you do not have ever desoldered a multi pin chip and do not have proper tools to do that, I would still recommend to give it to someone, that has the skills required.
Desoldering the RAM chips has the inherent danger of damaging the mainboard and creating even more damage to the CPC, that is actually beneath the RAM issue probably fine. (e.g. no critical parts damaged (like GateArray)).
Quote from: SerErris on 10:19, 05 June 24To be honest, the learning out of this thread for someone new to digital electronics and CPC in general is pretty much zero.
Probing things without knowledge on why and what and the background of what you actually do is pretty much confusing and not helping any bit to better understand what is going on.
Setting a Color in the GA with wires is the strangest thing I heard and I am not sure if I would be able to execute this, even knowing how all those thing work. It would be just out of my manipulation capabilities to get this timed correctly and working.
I would still say:
If anyone has no knowhow on any of those topics, you should not attempt to fix it. If you have good soldering skills (which we also do not know), you can attempt to replace the RAM.
However if you do not have ever desoldered a multi pin chip and do not have proper tools to do that, I would still recommend to give it to someone, that has the skills required.
Desoldering the RAM chips has the inherent danger of damaging the mainboard and creating even more damage to the CPC, that is actually beneath the RAM issue probably fine. (e.g. no critical parts damaged (like GateArray)).
If you don't need to save the chip you are removing, then this is probably the safest way to do it: https://www.cpcwiki.eu/index.php/IC_Repair
You don't need a desoldering gun to do it either.
Bryce.
Quote from: SerErris on 10:19, 05 June 24Setting a Color in the GA with wires is the strangest thing I heard and I am not sure if I would be able to execute this, even knowing how all those thing work. It would be just out of my manipulation capabilities to get this timed correctly and working.
I've never heard of it either. But I do it with one hand probably because I wanted to do it.
https://youtu.be/9e_AJJ8AfyQ
Turn on captions!!!;
; /--|>|-------- XXX------ M1
; READY/WAIT ---<
; \--|>|-------- XXX------ IORQ
; | crocodile
; diode
; |
; A15 -------> -----|>|-------GND
;
; always you can check Address and data when Z80 stopped.
;
; clip IORQ crocodile, clip M1 crocodile and reset (or swith on CPC).
;
ORG #0000
LD BC,#7F89 ;recatch M1 (1)
OUT (C),C ;recatch IORQ (2)
.0005
JP #0591 ;recatch M1 (3)
ORG #0591
DI
LD BC,#F782
.0595
OUT (C),C ;recatch IORQ (4)
.0597
LD BC,#F400 ;recatch M1 (5)
.059A
OUT (C),C ;store by short A15 ground and recatch IORQ (6)
.059C
LD BC,#F600 ;recatch M1 (7)
.059F
OUT (C),C ;second chance to store by short A15 ground and recatch IORQ ( 8 )
.05A1
LD BC,#EF7F ;recatch M1 (9)
.05A4
OUT (C),C ;ground one data line 0 to 4, and store by short A15 ground
; pen 0 set to 5 different colors
; set ROMDIS to +5V and remove crocodiles. If You see vertical bands RAMs work.
I'm back! You can decide if that is good or bad news. ;D
Before trying the latest test from
@McArti0, I finally took the CPC to a local computer repairs guy. (He has repaired C64s before but not CPCs. Not many options around here for repair shops.)
Yesterday he replaced the lower 64Kb of RAM and... still the same white screen and black border.
I will collect the CPC today but unfortunately no more updates for a while, because I'm away for the next couple of weeks.
So, I guess I need to decide what to do next. Dandanator? More experiments? Or the recycle bin.
(Thanks again for all time spent helping me here...)
..
Quote from: mmldrm on 10:45, 21 June 24Yesterday he replaced the lower 64Kb of RAM and... still the same white screen and black border.
(Thanks again for all time spent helping me here...)
The crystal divination ball for CPC duck diagnostics did not work. (https://www.cpcwiki.eu/forum/Smileys/SoLoSMiLeYS1/huh.gif)
Btw. If you don't have time, we won't anything.
Quote from: mmldrm on 10:45, 21 June 24Yesterday he replaced the lower 64Kb of RAM and... still the same white screen and black border.
Just to be sure: he replaced the right column of RAM ICs, right? (When I did my first repair I thought the left column is lower RAM.)
Were the RAM ICs fast enough and properly tested?
Even if he is experienced he still might have caused physical harm, which can have the exact same symptom as a defect RAM IC. Did you already check for broken traces / removed pads?
Quote from: mmldrm on 10:45, 21 June 24Dandanator? More experiments? Or the recycle bin.
What Dandanator would give you is the information which data line gives the wrong data. Based on that you can usually identify the defective RAM IC but it could also be that one of the other ICs related to the RAM circuit is defective. If you see all ICs in red then usually a different IC is the culprit. You might consider that while doing those massive tests you might have fried another IC accidentally. Or it was another IC from the start.
And please definitely not the recycle bin. This one will be saved with max. 2h of work and max 20€ for parts (worst case scenario) if followed the "usual" repair steps. Where are you located? Maybe someone is willing to do the repair for you if you cover the costs for parts and shipping? Or before you throw it away send it to someone who pays for shipping and will fix it.
Even if he did swap the wrong bank of RAM, you could now try the PAL jumper trick to see if the swapped bank works.
Bryce.
Quote from: McArti0 on 11:41, 21 June 24The crystal divination ball for CPC duck diagnostics did not work. (https://www.cpcwiki.eu/forum/Smileys/SoLoSMiLeYS1/huh.gif)
Instead of crystal ball it was rather experience based recommendations.
And the most problematic part here is, that after all those snake oil tests you recommended, it could also be that an accidental and unrecognized short has fried another IC. So maybe (just maybe) replacing the RAM first would have fixed it - but now it has been worsened. We will never know.
If you are willing to pay the shipping to Spain and back, I can fix it for you.
Quote from: eto on 11:09, 22 June 24Quote from: McArti0 on 11:41, 21 June 24The crystal divination ball for CPC duck diagnostics did not work. (https://www.cpcwiki.eu/forum/Smileys/SoLoSMiLeYS1/huh.gif)
So maybe (just maybe) (...) but now it has been worsened. We will never know.
Next crystal divinaton ball too :laugh:
Unfortunately, I won't make anyone think.
Quote from: abalore on 11:13, 22 June 24If you are willing to pay the shipping to Spain and back, I can fix it for you.
I will offer this option to my friend who owns the CPC, although I imagine the costs will be more than he wishes to pay (after buying the RAM and paying for the soldering).
Either way, thank you. 😊
I also noticed this on startup once or twice. After this, it boots to the usual white screen with black border. (The soldering guy thought it was not white but cyan after accidentally leaving on for many hours, but it still looks white to me.)
Not sure if these videos (particularly the blue lines) show progress, or regression. 😕
Anyway, I'm away for the next couple of weeks, so this is the last update for a while.
When I get back I'll do my best to check the RAM pins and maybe finally learn how to read the schematic diagram for traces and start that long process. The soldering guy said he checked all the RAM after installation, but it's still good to check it myself.
This is a regression. Connect +5v to RAMDIS and turn on CPC.
Ps. Don't listen to anyone here who scares you And fortune tellers
Did you get back the memory that was removed from CPC?
Ps2. Bright Cyan is color with all bit Hi like
OUT &7FFF,0 : OUT &7FFF, &40+&x11111
Ahhh yes nothing like joining a twenty year old community and calling some of the most respected and well known members 'fortune tellers'. You get 10 brownie points.
I lived behind the Iron Curtain where you died for the truth. Reliability of information was a highly valued good, regardless of who said it and when.
@mmldrmYou need to check whether all legs 40030/31 have a good connection to the socket. But now I would go back to the wire connection of the PAL IC.
Quote from: McArti0 on 13:23, 22 June 24I lived behind the Iron Curtain where you died for the truth. Reliability of information was a highly valued good, regardless of who said it and when.
If information is valued
regardless of who said, the place where you lived is just an argumentum ab auctoritate fallace.
For now, the information is that mmldrm was persuaded to desolder the memory by a random person and it is unknown whether with the GA removed.
The effects can be seen in the video. CRTC is no longer set correctly. Which was to be expected.
For better or for worse... I'm back. Apologies to anyone who groaned upon reading that. ;D
To summarise, I got Bank 0 of the RAM replaced but it hasn't changed anything. There was a regression to a bad CRTC signal, but thankfully that is happening less and less. Occasionally on startup the video signal is out of sync, but given a few seconds and a couple of restarts, it does get back to the familiar grey box and black border. This is true with the 40031 and also with the "wire trick".
Looking for things to try, I saw Noel's video* on PCB tracks and tried tracing a signal from IC108 data bus pins (26-33), through the RAM chips, and then traced through to the Z80 data pins.
It was a great demonstration on how to trace signals and how to read the 6128 circuit diagram. However, all of traces from IC108 data pin to Z80 pins were fine.
Am I correct in thinking that the upper (Bank 1) RAM cannot affect the boot process? Because I'm wondering if there is an issue with that RAM.
Happy to do more traces, or any other suggestions, but I wonder if I am better off getting a diagnostic ROM installed.
Also, I do still have the original RAM.
Thanks again for all time/thought/experiments spent on this.
*https://www.youtube.com/watch?v=2yvkz5VHN-0
OMG! You are alive. :D
@mmldrm"
Am I correct in thinking that the upper (Bank 1) RAM cannot affect the boot process? Because I'm wondering if there is an issue with that RAM"Anything is possible. Something might try to work then heat up and stop.
When cas1 is always Hi then ram bank 1 should not work.
ps. remove all RAM chips (don't confuse with old), wire PAL for second bank RAM try do this procedure... https://youtu.be/9e_AJJ8AfyQ
So, I'm back again. I've been dabbling in the background but wanted to retry several things before returning here...
Sorry
@McArti0, I still can't figure out how to follow your YouTube video. :( It is impressive that you can step through the instructions that way, but I am more likely to misunderstand something and get bad results or, worse, cause damage.
I've been going over the previous messages in this topic, now that I have a slightly better understanding of the CPC and general electronics, and have a few questions:
- Touching/wiggling capacitor at C309 causes audio crackling. Others don't. Probably irrelevant but thought I'd mention it. Is that normal?
- Leaving the 6128 running for a while, only the AY chip seems to get warmer than others. Not very warm, but warmer than the others. Is that a sign that the AY chip may have issues? Or would it have to be HOT to be a symptom of a bad chip?
- 40025 ROM data lines are not stuck anywhere near 0V or 5V, but most values are unchanging and don't fluctuate over time. Does ROM code halt when it displays the black/grey screen? I would have expected to see voltage fluctuations if 'random' binary data was being processed.
- 40010 voltages are nothing like what is expected in the table at https://www.cpcwiki.eu/forum/index.php?msg=237132 - again, is this simply because the boot process has not completed? Or maybe a sign of bad chip/chips?
- How best to check IC pin signals? Oscilloscope (50Hz) with dual channels, comparing signals at different points?
- Would it be better to spend £45 on a Dandanator or £60 on an oscilloscope (Zoyi ZT-703S)? Oscilloscope might be useful in the future. Dandanator will probably need to be resold, but if it is likely to give clear indications of chip failures then it seems worth trying first.
Apologies for some of those being general electronics questions (chip temperature, checking signals). I have also been watching many of Noel's videos (and others) trying to get a better understanding before writing this.
Thanks
M
1) This is the input of the audio amplifier. Normal.
2) A bit of fortune-telling. If something burned, the matter would be obvious.
3) You see the average value from one fast loop of the program. The values should be funny and constant.
4) These should be close, let's say with an accuracy of +/- 0.3V.
GA 40010 GA 40007
------------------------------------------------
CCLK ~1.2V pin4 pin40
RAS ~0.9V pin34 pin7
nCPU ~2.5V pin14 pin1
PHI ~2.1V pin19 pin39
READY ~1V pin22 pin2
5) It's best to check wisely :D
1MHz signals are useful, i.e. READY, nCPU, etc. For example, on one channel READY, on the other nCAS.
details after studying the page:
https://bread80.com/2021/06/03/understanding-the-amstrad-cpc-video-ram-and-gate-array-subsystem/
6) A better oscilloscope is DPOX180H. If Dandanator does not show you an image or hangs, we will continue without tools like an oscilloscope.
Quote from: McArti0 on 16:58, 08 December 246) A better oscilloscope is DPOX180H. If Dandanator does not show you an image or hangs, we will continue without tools like an oscilloscope.
I finally cracked and bought the DPOX180H. I was ready to get the Zoyi ZT-703S, but managed to find the DPOX180H for only £15 more. It's Christmas, and I didn't need anything else. Maybe I can get into retro electronics more in the future...
Thanks for the other answers. The scope should arrive just after Christmas. I will probably still do the above tests, but hopefully the scope will give more clues soon. :)
I hope everyone had a great Christmas and New year. I am finally the owner of a low-end handheld oscilloscope - the Fnirsi DPOX180H.
Today I was going to ask for suggestions to test with the oscilloscope, when I had another look at the board and STARING RIGHT AT ME was a broken capacitor on the lower DRAM IC130. :o
I think possibly the repair guy who socketed and replaced the lower DRAM may have broken it, or it somehow broke in transit (by me). Either way, it is irrelevant... it looks like I will need to replace this capacitor.
If I can do signal tests with the 'scope, any suggestions would (as always) be appreciated.
Quote from: mmldrm on 14:27, 25 August 24Am I correct in thinking that the upper (Bank 1) RAM cannot affect the boot process? Because I'm wondering if there is an issue with that RAM.
Yes it can affect the boot process. If an output pin of one of the RAM ICs is always high or low, it can stop the boot process exactly as observed here.
Quote from: mmldrm on 14:14, 08 December 24- 40025 ROM data lines are not stuck anywhere near 0V or 5V, but most values are unchanging and don't fluctuate over time. Does ROM code halt when it displays the black/grey screen? I would have expected to see voltage fluctuations if 'random' binary data was being processed.
The system will most likely not halt (unless it hits a HALT instruction of course) but the CPU will end in some kind of loop. That's even the case when the CPC properly boots and shows the BASIC prompt. As long as there is no active change of data in RAM/ROM (e.g. by starting a program) the voltage on each pin looks static.
Quote from: mmldrm on 13:44, 02 January 25I hope everyone had a great Christmas and New year. I am finally the owner of a low-end handheld oscilloscope - the Fnirsi DPOX180H.
Today I was going to ask for suggestions to test with the oscilloscope, when I had another look at the board and STARING RIGHT AT ME was a broken capacitor on the lower DRAM IC130. :o
I think possibly the repair guy who socketed and replaced the lower DRAM may have broken it, or it somehow broke in transit (by me). Either way, it is irrelevant... it looks like I will need to replace this capacitor.
If I can do signal tests with the 'scope, any suggestions would (as always) be appreciated.
The capacitor is unlikely the issue. Breaking a bit off this type of ceramic capacitor will just lower the capacity. If that's a 100n cap, then it will now be about 60n, but still fully functional (and that's not enough to bother the CPC).
They have a single set of metal plates with a solid dielectric in between and don't short out when they break.
In the early days of radio, this was even deliberately done to tune the circuit. It was called "trimming". Small variable capacitors are still called "Trimmers" today, possibly from this?
Bryce.
Quote from: Bryce on 20:41, 02 January 25The capacitor is unlikely the issue. Breaking a bit off this type of ceramic capacitor will just lower the capacity. If that's a 100n cap, then it will now be about 60n, but still fully functional (and that's not enough to bother the CPC).
They have a single set of metal plates with a solid dielectric in between and don't short out when they break.
In the early days of radio, this was even deliberately done to tune the circuit. It was called "trimming". Small variable capacitors are still called "Trimmers" today, possibly from this?
Bryce.
Wow, that's interesting! I did think that it was 'not impossible' that it might still function, but I did think it was very unlikely. Apparently not. I love the sound of the methods from the early days. :)
I'll try looking for stuck high/low pins on the upper RAM bank (and probably check the lower RAM bank again), now that
@eto has kindly confirmed that it can affect the boot process.
Feels like I would be best to start from scratch and redo the tests I did with the multimeter, now that I can see the signals and not just the average voltages.
Thanks again for the advice, folks.
Have you made a HAL adapter to swap the upper and lower banks?
Bryce.
Quote from: Bryce on 15:11, 03 January 25Have you made a HAL adapter to swap the upper and lower banks?
I haven't made one, due to having no 20 pin sockets, but I did previously use wires to manually create the adapter (from your post, via Noel's video, and McArti0's suggestion).
I could do that again. (I'm also tempted to get a 20 pin socket to make a permanent adapter).
You need to be sure that after inserting the IC or adapter all connections are secure and stable. Sometimes it is thought that it is, but it is not.
Check by oscylloscope signal RAS, CAS, CLK 4 MHz, READY.
On Probe setting 10x. And trigger in oscyllo single.
Quote from: McArti0 on 17:09, 03 January 25You need to be sure that after inserting the IC or adapter all connections are secure and stable. Sometimes it is thought that it is, but it is not.
Check by oscylloscope signal RAS, CAS, CLK 4 MHz, READY.
On Probe setting 10x. And trigger in oscyllo single.
If he's getting a border and screen area, the CPU and the CLK are working correctly.
Bryce.
I did all the tests, then realised that it should have been with the upper bank adapter in place. :picard:
Well, here are the results anyway. RAS, CAS, READY, CCLK (1MHz) from the 40010 and also Z80 CLK (4MHz).
I also checked the readings from one of the lower RAM bank ICs (IC134). In fact, the other RAM chips had identical results. Once I format and label the images, I'll share those too.
IC134 (lower RAM chip) readings
Quote from: mmldrm on 14:27, 04 January 25IC134 (lower RAM chip) readings
Pin15 CAS0 is null. How have You configure PAL 40030?
Quote from: McArti0 on 16:04, 04 January 25Quote from: mmldrm on 14:27, 04 January 25IC134 (lower RAM chip) readings
Pin15 CAS0 is null. How have You configure PAL 40030?
The HAL (40031 here) is the original chip. No override.
I'll check all the pin readings on the 40031, to have a better overview of that IC.
When CAS on GA is present then CAS0 or CAS1 is copy CAS.
Try CAS, CAS0 and CAS1 but on PINs/Legs HAL(PAL).
CAS (pin 16 on GA/40010) has a nice clear binary signal.
CAS0 and CAS1 on the HAL are both still null, like I shared previously.
I see there is also NCAS on the HAL but I could find no description for it. I measured NCAS and it looks the same as CAS on GA..?
Quote from: mmldrm on 13:22, 06 January 25I see there is also NCAS on the HAL but I could find no description for it. I measured NCAS and it looks the same as CAS on GA..?
NCAS on the HAL is CAS from the GateArray.
CAS0 should be identical to NCAS as long as RAM banking is not active. If the HAL is inserted and there is no CAS0 signal, then the HAL is not working.
Time to make the HAL adapter. If you don't have a socket you could also make one using pin headers - even a permanent/robust one with a perfboard.
Quote from: eto on 14:21, 06 January 25NCAS on the HAL is CAS from the GateArray.
CAS0 should be identical to NCAS as long as RAM banking is not active. If the HAL is inserted and there is no CAS0 signal, then the HAL is not working.
Time to make the HAL adapter. If you don't have a socket you could also make one using pin headers - even a permanent/robust one with a perfboard.
Thanks. I had previously shoved jumper wires into the HAL socket, but I have a 20pin socket arriving in the next couple of days.
Will get a HAL adapter soldered up and see if that improves things.
https://www.cpcwiki.eu/forum/amstrad-cpc-hardware/banked-video-ram/msg231994/#msg231994
Check all connections PAL/HAL (after insert) with others path point.
Quote from: McArti0 on 15:29, 06 January 25https://www.cpcwiki.eu/forum/amstrad-cpc-hardware/banked-video-ram/msg231994/#msg231994
Check all connections PAL/HAL (after insert) with others path point.
Thanks. I'll do that while I wait for the 20 pin socket to arrive.
In this image, "OK" means "measurements were consistent across test points". I don't know if the values themselves are correct.
Measurements marked "OK?" were consistent across test points but voltages looked bad to me, i.e. not high or low.
Finally, I couldn't see where to test pin 20 elsewhere on the board. As it is Vcc and was 5V, I figured it didn't matter too much.
What have you on screen after switch ON CPC?
Quote from: McArti0 on 19:34, 06 January 25What have you on screen after switch ON CPC?
Still the white screen with black border (the one that usually indicates a RAM issue).
Pin 16 is CAS1 and 1.2V does not seem right. That should always be high unless someone has activated banking.
Pin 17 is CAS0 and LO seems totally wrong. That means CAS0 would always be set. What do you get on Pin 7 (NCAS) or Pin 31 on the GateArray (CASAD)?
Pin 2 should be identical on IC114 and IC115. If not you have a crack in the lanes. Pin 35 on the GateArray is not connected to D0 as tha Gate Array is isolated from the data bus via a latch.
Do it again.
https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg238062/#msg238062
Quote from: eto on 20:28, 06 January 25Pin 16 is CAS1 and 1.2V does not seem right. That should always be high unless someone has activated banking.
Pin 17 is CAS0 and LO seems totally wrong. That means CAS0 would always be set. What do you get on Pin 7 (NCAS) or Pin 31 on the GateArray (CASAD)?
Pin 2 should be identical on IC114 and IC115. If not you have a crack in the lanes. Pin 35 on the GateArray is not connected to D0 as tha Gate Array is isolated from the data bus via a latch.
Pin16 comment: Why? All RAM needs to be refreshed.
Bryce.
Quote from: Bryce on 14:41, 07 January 25All RAM needs to be refreshed..
RAM is "RAS only refreshed". Don't need CAS.
Quote from: McArti0 on 14:49, 07 January 25Quote from: Bryce on 14:41, 07 January 25All RAM needs to be refreshed..
RAM is "RAS only refreshed". Don't need CAS.
Sorry, you're right. The coffee machine is broken today. I probably shouldn't be posting at all. :D
Bryce.
Quote from: eto on 20:28, 06 January 25Pin 16 is CAS1 and 1.2V does not seem right. That should always be high unless someone has activated banking.
Pin 17 is CAS0 and LO seems totally wrong. That means CAS0 would always be set. What do you get on Pin 7 (NCAS) or Pin 31 on the GateArray (CASAD)?
Pin 2 should be identical on IC114 and IC115. If not you have a crack in the lanes. Pin 35 on the GateArray is not connected to D0 as tha Gate Array is isolated from the data bus via a latch.
Both pins 16 and 17 are definitely both consistently LO. Just checked again to confirm.
NCAS (pin 7 @ HAL) is a nice clean HI/LO signal.
CASAD (pin 31 @ GA) is also a nice clean HI/LO signal.
Pin 2 on IC114 and IC115 are indeed identical.
The 20 pin socket should arrive in the next day or two - I'll make that HAL adapter and see what readings I get then.
Either PAL/HAL does not work or nCPU does not reach.
Quote from: McArti0 on 18:17, 07 January 25Either PAL/HAL does not work or nCPU does not reach.
Where does nCPU originate? I can't find any references to it. Only on the PAL pin-outs.
I thought I would try to trace the origin (still learning, here) and try to measure the value before it reaches the PAL. (Also learning to call it PAL and not HAL.)
Edit: I see it on the GA, in the circuit diagram. But I can't tell how that relates to the PAL - difficult to tell from the diagram.
n is the same as a line above the text.
_____
CPU = nCPU - Pin14 GA
Quote from: mmldrm on 20:34, 06 April 24Select Expand
pin7 (2.18v) - pin17 (3.63v)
pin8 (1.67v) - pin12 (3.73v)
pin9 (0.66v) - pin19 (0.75v)
Old measure pin17 PAL CAS0 work.
Quote from: mmldrm on 20:34, 07 January 25Quote from: McArti0 on 18:17, 07 January 25Either PAL/HAL does not work or nCPU does not reach.
Where does nCPU originate? I can't find any references to it. Only on the PAL pin-outs.
I thought I would try to trace the origin (still learning, here) and try to measure the value before it reaches the PAL. (Also learning to call it PAL and not HAL.)
Edit: I see it on the GA, in the circuit diagram. But I can't tell how that relates to the PAL - difficult to tell from the diagram.
All "n" signals originate from the GA. That's what the "n" means.
Bryce.
Thanks. I'm slowly learning the notation as we go. (I should look up a reference to minimise the simplest questions.)
nCPU is a clean signal, both on GA pin 11 and PAL pin 14.
So it's looking like (at a minimum) the PAL is toast?
Quote from: McArti0 on 23:09, 07 January 25Quote from: mmldrm on 20:34, 06 April 24Select Expand
pin7 (2.18v) - pin17 (3.63v)
pin8 (1.67v) - pin12 (3.73v)
pin9 (0.66v) - pin19 (0.75v)
Old measure pin17 PAL CAS0 work.
You are right. I just repeated the above measurements with my multimeter and they are very different. :(
Quote from: mmldrm on 09:12, 08 January 25Thanks. I'm slowly learning the notation as we go. (I should look up a reference to minimise the simplest questions.)
That unfortunately doesn't always help as it's not used everywhere properly.
E.g. the CPU signal is shown as nCPU on the GateArray but is then called /CPU on the PAL. And on the Wiki pages about the Gate Array signals nCAS is documented as /CAS. (https://www.cpcwiki.eu/index.php/Gate_Array_and_ASIC_Pin-Outs) In some other docs you will find the CAS0 being called nCAS0.
What usually helps me most is to look up the schematics, then follow the lanes. That also helps to better understand how everything works together. https://www.cpcwiki.eu/index.php/Schematics
Quote from: mmldrm on 11:08, 08 January 25You are right. I just repeated the above measurements with my multimeter and they are very different. :(
But still bad.
ps. how many now?
Quote from: eto on 11:15, 08 January 25Quote from: mmldrm on 09:12, 08 January 25Thanks. I'm slowly learning the notation as we go. (I should look up a reference to minimise the simplest questions.)
That unfortunately doesn't always help as it's not used everywhere properly.
E.g. the CPU signal is shown as nCPU on the GateArray but is then called /CPU on the PAL. And on the Wiki pages about the Gate Array signals nCAS is documented as /CAS. (https://www.cpcwiki.eu/index.php/Gate_Array_and_ASIC_Pin-Outs) In some other docs you will find the CAS0 being called nCAS0.
What usually helps me most is to look up the schematics, then follow the lanes. That also helps to better understand how everything works together. https://www.cpcwiki.eu/index.php/Schematics
That is actually all the same. nCPU ^CPU /CPU !CPU all means, that this is active low signal - nothing more nothing less.
And there is a difference between high and low active and you can use the same signal for different purposes. For instance if you have on chip that has for instance chip enable as active low, and the other one active high, you can connect both to the same signal and the will activate at oposite times.
This is actually something that is beeing used by the CPC.
nCPU is at the same time the Soundchip clock. That is because it is just a 1mhz clock signal, which is what is required by the sound chip. The beauty of it is, that the sound chip is automatically synced to the 4Mhz clock (exactly 1/4 the speed) and to CRTC and other components. Not that it is really required in this scenario :-)
Would you like me to check the values on the PAL now?
Quote from: McArti0 on 17:09, 03 January 25You need to be sure that after inserting the IC or adapter all connections are secure and stable. Sometimes it is thought that it is, but it is not.
Check by oscylloscope signal RAS, CAS, CLK 4 MHz, READY.
On Probe setting 10x. And trigger in oscyllo single.
So, with the PAL adapter in place the readings are much cleaner.
PAL adapter:
CAS1 pin 16 signal - same as CAS pin 16 on GA
CAS0 pin 17 HI
Lower RAM Bank 0:
RAS pin 4 same signal as RAS pin 34 on GA
CAS pin 15 HI
Upper RAM Bank 1:
RAS pin 4 signal- same as RAS pin 34 on GA
CAS pin 15 signal - same as CAS pin 16 on GA
So the CAS and RAS signals are getting to the upper RAM, and the 'PAL adapter' CAS0 is HI.
Screen is still white with black border. But I have not replaced the upper RAM.
Is it possible to use the adapter to force the lower RAM? Those chips were replaced and socketed.
Looks like I can do this by swapping the wires in pins 16 (CAS1) and 17 (CAS0) of the adapter. Seems obvious, but I will wait for the experts to confirm before going rogue!
Quote from: mmldrm on 11:10, 11 January 25Is it possible to use the adapter to force the lower RAM? Those chips were replaced and socketed.
yes. Just make sure nCAS goes to CAS0 and CAS1 is high.
But don't be surprised if nothing changes: if one of your upper RAM ICs has a short, nothing will change. The data pins are not isolated and if any of them is setting a bit high or low constantly then there is no change. Same if the latch has an issue that isolates the RAM from the bus: nothing will change.
Quote from: eto on 11:28, 11 January 25yes. Just make sure nCAS goes to CAS0 and CAS1 is high.
But don't be surprised if nothing changes: if one of your upper RAM ICs has a short, nothing will change. The data pins are not isolated and if any of them is setting a bit high or low constantly then there is no change. Same if the latch has an issue that isolates the RAM from the bus: nothing will change.
Thanks for the quick answer. To no-one's surprise, nothing changed. At least I knew in advance to not get my hopes up!
Am I going to have to replace the upper RAM before I can do further tests? (Given that it was £40 to do the first bank, I may have to be brave and replace the upper RAM myself...)
Quote from: mmldrm on 11:43, 11 January 25Am I going to have to replace the upper RAM before I can do further tests?
Probably yes.
I would do two things before:
1) Remove the socketed RAM from bank 0 and try again with the adapter (which connects CAS to CAS1) just to make sure none of the new RAM ICs is fried
2) check for any shorts introduced by the repair guy - you never know if it's done well or not
If that does not show a change, then I personally would probably try the following:
- Put in the RAM again in bank 0, PAL adapter CAS to CAS0 so it runs on bank 0
- On bank 1, cut the Dout pin of each RAM IC one after the other and in between check if the CPC then works
If it's really just the Dout then the CPC should start once you have cut the pins of the failing IC(s) and just need to replace that IC. You probably also can reconnect the disconnected pins of the other ICs (at least if you don't cut it too close to the IC package) which is much less work and less risk than replacing all ICs.
To be honest, I never did that myself as I am usually using the Amstrad Diagnostics. And it might also be that it is as likely that an address pin is short, which might have the same effect. Let's ask someone who really knows what he's talking about...
@Bryce would that idea make sense or is that a stupid idea?
Unfortunately step 1) still resulted in the same white screen with black border.
Quote from: eto on 12:26, 11 January 252) check for any shorts introduced by the repair guy - you never know if it's done well or not
Could you describe that process in more detail? Reading up on it, I understand the principle that there should be high resistance between neighbouring traces/pads, but not sure where is best to measure across in this scenario (checking the new RAM/sockets).
Quote from: mmldrm on 19:12, 16 January 25Unfortunately step 1) still resulted in the same white screen with black border.
Not really a surprise but it was important to test that, so you don't exchange the ICs of bank 1 - and then it's one of the ICs in bank 0.
Quote from: mmldrm on 19:12, 16 January 25Could you describe that process in more detail? Reading up on it, I understand the principle that there should be high resistance between neighbouring traces/pads,
1) download the schematics, and have a look at the part with the RAM
2) check that all pins that should be connected indeed are connected. See attached one example: Pin 10 of
all RAM ICs should be connected. With your multimeter you can put one probe to any IC and then check if all other ICs have a connection on pin 10. Finally also check if there is a connection to pin 7 of IC104 (resistance should show 47 Ohms). If any connection fails, there must be fault somewhere. Just an example what happened to me: When removing a RAM IC I accidentally removed a solder pad - and once I inserted the IC socket and didn't recognize that one leg was bent - and then it had no connection.
3) check that there is no connection between pins that should not be connected. As you already checked the correct connections in step 2 it should be sufficient to test
a) on each RAM IC: check all pins - start with one probe on pin 1, then check Pin 2-16 with the other probe, move on to pin 2, check pin 3-16 ... and so on
b) on one bank: check all D-Out pins. start on Bank 0, RAM IC 1, put probe on Pin 14 and check Pin 14 on all RAM IC 2-8 of bank 0 with the other probe, move on to IC 2 and check PIN 14 on RAM IC 3-8
If there is any connection where no connection should be, then a short was introduced. E.g. I once had a tiny piece of wire that was stuck under an IC socket and connected two pins with each other.
Quote from: eto on 20:03, 16 January 252) check that all pins that should be connected indeed are connected. See attached one example: Pin 10 of all RAM ICs should be connected. With your multimeter you can put one probe to any IC and then check if all other ICs have a connection on pin 10. Finally also check if there is a connection to pin 7 of IC104 (resistance should show 47 Ohms). If any connection fails, there must be fault somewhere. Just an example what happened to me: When removing a RAM IC I accidentally removed a solder pad - and once I inserted the IC socket and didn't recognize that one leg was bent - and then it had no connection.
Thanks once again for the very clear instructions. I've done continuity checks on all 8 data pins (5,6,7,9,10,11,12,13) on all RAM ICs, and the appropriate pin on one of the 74LS153. The resistance showed 50-52 Ohms in all cases. I also checked the resistance across one or two of the resistors, and it was also 50 Ohms.
In short, all the connections between RAM ICs, and the 74LS153s, are intact.
Quote from: eto on 20:03, 16 January 253) check that there is no connection between pins that should not be connected. As you already checked the correct connections in step 2 it should be sufficient to test
a) on each RAM IC: check all pins - start with one probe on pin 1, then check Pin 2-16 with the other probe, move on to pin 2, check pin 3-16 ... and so on
Well this was fun! ??? My eyes did cross a few times. But definitely worthwhile. In short, there are no shorts between pins on any of the RAM ICs.
Quote from: eto on 20:03, 16 January 25b) on one bank: check all D-Out pins. start on Bank 0, RAM IC 1, put probe on Pin 14 and check Pin 14 on all RAM IC 2-8 of bank 0 with the other probe, move on to IC 2 and check PIN 14 on RAM IC 3-8
The connections between pins on the same data line were all intact, and I checked all 8 data pins on each IC. Maybe worth mentioning that on Bank 0 I was checking sockets, as the RAM ICs are currently removed.
Happy to do the same tests with the RAM ICs in place, if that would rule out any disconnect between socket and IC. In fact, the guy that socketed the Bank 0 RAM did say that he thought the sockets were a bit crap. Some RAM ICs don't sit well in the sockets, so it might be worth me doing these tests again with the ICs in place. Even though I am pretty sure that I already checked the signals with a scope. Might be safer if I do that, just to rule things out.
Well, I learned a bit more from this and also got peace of mind that the RAM connections in both banks seems to be in good shape.
Still puzzled that the current situation seems to be:
- RAM seems to pass all manual tests with multimeter/oscilloscope
- PAL chip seems screwed
- PAL override (bank switching trick) doesn't solve the problem
This makes me wonder about the other ICs involved in RAM management (the four 74LS153s) or the data latch / buffers (74LS244 and 74LS373).
Then again, as you know, I have no idea what I'm doing. ;D I'd love to suggest a Dandanator to the owner of this board, so I can run the diagnostic ROM and he could use it to load games in the future, but I'm still not convinced that this board is working enough for the diagnostic ROM to give correct feedback.
Quote from: mmldrm on 23:34, 19 January 25, but I'm still not convinced that this board is working enough for the diagnostic ROM to give correct feedback.
If you get a white screen with black border it will be enough to run the diagnostics ROM. "white screen with black border" means that ROM can be executed.
Quote from: mmldrm on 23:34, 19 January 25This makes me wonder about the other ICs involved in RAM management (the four 74LS153s) or the data latch / buffers (74LS244 and 74LS373).
Yes, also possible. More likely though that it's one of the RAM ICs of bank 1.
Quote from: mmldrm on 23:34, 19 January 25PAL override (bank switching trick) doesn't solve the problem
E.g.
1) If just one IC of bank 1 is fried in a way that its data out is always high, then also bank 0 cannot work properly as the lanes of the data pins are shared. the HIGH from bank 1 will always set the respective bit to 1 and the CPU will not read correct data.
2) if one of the other ICs is malfunctioning, either the address is not corrcetly calculated or the data out will not be properly passed though to the CPU.
In all cases the observed effect can be the same: white screen + black border.
Quote from: mmldrm on 23:34, 19 January 25Quote from: eto on 20:03, 16 January 252) check that all pins that should be connected indeed are connected. See attached one example: Pin 10 of all RAM ICs should be connected. With your multimeter you can put one probe to any IC and then check if all other ICs have a connection on pin 10. Finally also check if there is a connection to pin 7 of IC104 (resistance should show 47 Ohms). If any connection fails, there must be fault somewhere. Just an example what happened to me: When removing a RAM IC I accidentally removed a solder pad - and once I inserted the IC socket and didn't recognize that one leg was bent - and then it had no connection.
Thanks once again for the very clear instructions. I've done continuity checks on all 8 data pins (5,6,7,9,10,11,12,13) on all RAM ICs, and the appropriate pin on one of the 74LS153. The resistance showed 50-52 Ohms in all cases. I also checked the resistance across one or two of the resistors, and it was also 50 Ohms.
In short, all the connections between RAM ICs, and the 74LS153s, are intact.
Quote from: eto on 20:03, 16 January 253) check that there is no connection between pins that should not be connected. As you already checked the correct connections in step 2 it should be sufficient to test
a) on each RAM IC: check all pins - start with one probe on pin 1, then check Pin 2-16 with the other probe, move on to pin 2, check pin 3-16 ... and so on
Well this was fun! ??? My eyes did cross a few times. But definitely worthwhile. In short, there are no shorts between pins on any of the RAM ICs.
Quote from: eto on 20:03, 16 January 25b) on one bank: check all D-Out pins. start on Bank 0, RAM IC 1, put probe on Pin 14 and check Pin 14 on all RAM IC 2-8 of bank 0 with the other probe, move on to IC 2 and check PIN 14 on RAM IC 3-8
The connections between pins on the same data line were all intact, and I checked all 8 data pins on each IC. Maybe worth mentioning that on Bank 0 I was checking sockets, as the RAM ICs are currently removed.
Happy to do the same tests with the RAM ICs in place, if that would rule out any disconnect between socket and IC. In fact, the guy that socketed the Bank 0 RAM did say that he thought the sockets were a bit crap. Some RAM ICs don't sit well in the sockets, so it might be worth me doing these tests again with the ICs in place. Even though I am pretty sure that I already checked the signals with a scope. Might be safer if I do that, just to rule things out.
Well, I learned a bit more from this and also got peace of mind that the RAM connections in both banks seems to be in good shape.
Still puzzled that the current situation seems to be:
- RAM seems to pass all manual tests with multimeter/oscilloscope
- PAL chip seems screwed
- PAL override (bank switching trick) doesn't solve the problem
This makes me wonder about the other ICs involved in RAM management (the four 74LS153s) or the data latch / buffers (74LS244 and 74LS373).
Then again, as you know, I have no idea what I'm doing. ;D I'd love to suggest a Dandanator to the owner of this board, so I can run the diagnostic ROM and he could use it to load games in the future, but I'm still not convinced that this board is working enough for the diagnostic ROM to give correct feedback.
The RAM tests you've done aren't conclusive. I suspect that there are one or more failed RAM chips.
Bryce.
So once again, only more consciously...
PAL wire
RAMDIS set Hi
ROMDIS set Hi
RESET one touch Lo
A short program is executed
38h read RAM FFh
write RAM 00h -Address-1
write RAM 39h -Address-2
out RAM D7-D0 0011-1001
Quote from: McArti0 on 19:02, 20 January 25So once again, only more consciously...
PAL wire
RAMDIS set Hi
ROMDIS set Hi
Where is best to set RAMDIS/ROMDIS to high? It is not ideal to fix pins to the expansion port with tape.
Can I set RAMDIS high on pin 4 of the PAL socket, even with the wires? I also see it on pin 10 of IC112 (74LS32)... would that work?
Can I set ROMDIS high on pin 22 of IC103 (40025 ROM)?
https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg238468/#msg238468
nothing new
@mmldrm Where in the world are you Located ?
Keep Safe. Ray
Quote from: Audronic on 10:15, 22 January 25@mmldrm
Where in the world are you Located ?
Scotland. Near Edinburgh.
Quote from: McArti0 on 18:05, 21 January 25https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg238468/#msg238468
nothing new
Thanks. I thought I had better connection points but couldn't remember where. Your memory is better than mine (and my CPC's).
(Is there another for RAMDIS?)
R124 but ROMDIS is enough as I checked mine.
Of course you won't see vertical bars.
Quote from: mmldrm on 11:12, 22 January 25Quote from: Audronic on 10:15, 22 January 25@mmldrm
Where in the world are you Located ?
Scotland. Near Edinburgh.
Ok Thanks
Perhaps somebody near you would have a Diag Cart or some suitable tester, Try putting a callout and see what happens.
as its been a VERY long journey so dar Keep Safe. Ray
Quote from: Audronic on 23:47, 22 January 25Ok Thanks
Perhaps somebody near you would have a Diag Cart or some suitable tester, Try putting a callout and see what happens.
as its been a VERY long journey so dar
It sure has! And I'm incredibly thankful for all the help, but yes it's been.. a journey..
Where would be the best place to put a call-out?
@mmldrm Here On the forum in this Saga.
Keep Safe
Ray
Quote from: mmldrm on 00:35, 23 January 25Quote from: Audronic on 23:47, 22 January 25Ok Thanks
Perhaps somebody near you would have a Diag Cart or some suitable tester, Try putting a callout and see what happens.
as its been a VERY long journey so dar
It sure has! And I'm incredibly thankful for all the help, but yes it's been.. a journey..
Where would be the best place to put a call-out?
Hi, on the South Coast, so a bit far but happy to help. Message me, maybe I can loan you a lower DiagROM.
Quote from: McArti0 on 12:02, 22 January 25R124 but ROMDIS is enough as I checked mine.
Of course you won't see vertical bars.
Just before I do this test, I checked the D0-D7 values when I first switch on. All data pins were high, except D4 and D7 which looked like this. I measured this both at Z80 and on pin 2 of the RAM chips themselves.
Thought I should share, in case it helps. e.g. maybe it is a sign that IC126 and IC123 are the faulty ones.
I will do the ROMDIS test tonight if I can get time to do it.
Quote from: mmldrm on 00:35, 23 January 25Quote from: Audronic on 23:47, 22 January 25Ok Thanks
Perhaps somebody near you would have a Diag Cart or some suitable tester, Try putting a callout and see what happens.
as its been a VERY long journey so dar
It sure has! And I'm incredibly thankful for all the help, but yes it's been.. a journey..
Where would be the best place to put a call-out?
You could have drastically reduced the journey time by just replacing the RAM's from the very start. The problem is obviously related to RAM, but from all of the IC's involved in reading/writing to/from RAM, the RAM IC's themselves are many, many 1000's of times more likely to fail than every other chip involved. So 99.9% of the time swapping the RAM's will resolve the issue without any need for time consuming testing.
Bryce.
Quote from: Bryce on 19:39, 23 January 25You could have drastically reduced the journey time by just replacing the RAM's from the very start. The problem is obviously related to RAM, but from all of the IC's involved in reading/writing to/from RAM, the RAM IC's themselves are many, many 1000's of times more likely to fail than every other chip involved. So 99.9% of the time swapping the RAM's will resolve the issue without any need for time consuming testing.
To be honest I feel worst for the forum members who have tired of seeing this topic. I've learned quite a bit about electronics, the layout of the CPC boards, and the general principles of 8-bit home computer design.
On the other hand, yes, if I had realised that replacing the lower RAM only gave a 50/50 chance of fixing 'the RAM problem' then I would probably have had both bank replaced. That misunderstanding is on me.
If the budget had been mine, I would probably have replaced the RAM earlier. I have now suggested buying a Dandanator and diagnosing using that. At least he can use it later to load games, or sell it on eBay for a minimal loss.
Actually, if it's another £50 to get someone to replace the upper RAM vs. £45 for the Dandanator (and probably
then £50 for the RAM installation) then maybe I will recommend replacing the upper RAM. I just didn't want him to spend more money and still have no definite diagnosis of the fault, if it continued to fail.
I also didn't realise that the chances of it being a non-RAM IC were quite so incredibly low.
Quote from: McArti0 on 12:02, 22 January 25R124 but ROMDIS is enough as I checked mine.
Of course you won't see vertical bars.
Please ignore the previous message. I halted the ROM with ROMDIS and the PAL swap trick, and checked again with the scope. After altering the horizontal scale on the scope, I now realise that I'm seeing a series of 1s and a series of 0s.
And yes, D7-D0 does indeed read 0011-1001.
This is on the upper RAM. The new socketed RAM ICs are removed.
I've attached what I am interpreting as a constant refresh of 1, and 0.
Quote from: Rabs on 07:44, 23 January 25Hi, on the South Coast, so a bit far but happy to help. Message me, maybe I can loan you a lower DiagROM.
Thanks! I really appreciate the offer. Unfortunately the ROM chips aren't socketed and you can probably tell I'm trying to minimised required soldering, as I would be (to some extent) learning on the job if I do it myself.
So I'm having to decide between Dandanator (or equivalent) to override the ROM, or blindly replacing/socketing the upper RAM.
Nevertheless, thanks for the kind offer. :)
Quote from: mmldrm on 21:27, 23 January 25And yes, D7-D0 does indeed read 0011-1001.
D7-D0 on pin14 RAM too?
Test at ROMDIS Hi pin14 RAM D5,D4,D3,D0.
Quote from: McArti0 on 22:05, 23 January 25Quote from: mmldrm on 21:27, 23 January 25And yes, D7-D0 does indeed read 0011-1001.
D7-D0 on pin14 RAM too?
Sorry, forgot to check RAM outputs. I have attached RAM I/O for D0-D7. I think the lo values look strange on all four (D1, D2, D6, D7) but hi values look ok?
(Happy to test while I wait for my friend's response on Dandanator and replacing upper RAM.)
D1 ,D2,D6,D7 shouldn't have Hi peaks, and freq is not properly.
Try now after ROMDIS Lo , remove CAS1 at PAL , Set CAS1 to Hi and measure out D7-D0
All Hi or Lo peaks should be disapear
Set H=500ns and click ZOOM
Quote from: McArti0 on 19:57, 24 January 25D1 ,D2,D6,D7 shouldn't have Hi peaks, and freq is not properly.
Try now after ROMDIS Lo , remove CAS1 at PAL , Set CAS1 to Hi and measure out D7-D0
All Hi or Lo peaks should be disapear
Set H=500ns and click ZOOM
I wouldn't be so worried about the frequency, because cheap scopes are really bad at accurately measuring the frequency with so little data, especially when it's not a sinewave. But the voltage levels are definitely not good. It's definitely time to swap out any original RAM's.
Bryce.
Quote from: McArti0 on 19:57, 24 January 25D1 ,D2,D6,D7 shouldn't have Hi peaks, and freq is not properly.
Try now after ROMDIS Lo , remove CAS1 at PAL , Set CAS1 to Hi and measure out D7-D0
All Hi or Lo peaks should be disapear
Set H=500ns and click ZOOM
"ROMDIS lo" means a normal startup process, yes? I set CAS1 to hi and put the 8x Bank 0 RAM ICs back in. (They were removed in the previous test, when the PAL trick was forcing Bank 1.)
I measured both banks, D0-D7. The signals look good to my eyes. But I am suspecting I misunderstood your setup instructions.
Quote from: Bryce on 16:53, 25 January 25I wouldn't be so worried about the frequency, because cheap scopes are really bad at accurately measuring the frequency with so little data, especially when it's not a sinewave. But the voltage levels are definitely not good. It's definitely time to swap out any original RAM's.
Thanks. I'll tell the owner that the other bank needs replaced, and let him decide whether I do it or we pay someone to do it.
ROMDIS Hi ofcourse , my error.
without RAM bank 0.
Quote from: mmldrm on 16:59, 25 January 25Thanks. I'll tell the owner that the other bank needs replaced, and let him decide whether I do it or we pay someone to do it.
RAM is probably good.
Quote from: Bryce on 16:53, 25 January 25But the voltage levels are definitely not good.
:laugh:
Good condition CPC 6128, value 0039h in all RAM. D4,D5,D7
Quote from: McArti0 on 22:02, 25 January 25Good condition CPC 6128, value 0039h in all RAM. D4,D5,D7
I will try to do the same tests here.
I do have two probes if it helps with tests.