Hi all,
Yesterday I decided to take my CPCs out for some playing time, but unfortunately, destiny had another plan for me...
First up was the 6128. I turned it on, was happy to see the boot screen, but when I went to try a game or two, first I noticed like some weird dots/dashes appearing, and then the screen garbled :picard:
A few more power-offs/power-ons had the same result, turn it on, screen comes up and after a few seconds, garbled screen. Last attempts just had a grey screen. What could be the problem? RAM?
Second up, I said ok, let's play a bit woth the 464. Boot screen comes on, and cursor start advancing rapidly on its own, like if the space bar was being held down. Pressing any other key stops it, and it doesn't start again (the cursor that is), however:
1) The space bar does not work (all other keys work) on the basic screen
2) Something is still going on with the space bar, as I tried a racing game where first screen is the track selection which is changed precisely with space, and it keeps constantly changing, again on its own, as if the space bar was being held down.
Not sure what could be the problem with this one.
Any help is much appreciated!
For your CPC6128, it sounds like it could be the ram, but it could be other things also.
First things to try is re-seat the Z80 and the Gate Array (40010 chip), to make sure they both have good contact.
For the CPC464, try disconnecting the keyboard, see if the cursor stops advancing, then your problem is the keyboard (clean the folio). Otherwise it's probably the AY8912 chip.
Quote from: Duke on 21:41, 13 May 18
For your CPC6128, it sounds like it could be the ram, but it could be other things also.
First things to try is re-seat the Z80 and the Gate Array (40010 chip), to make sure they both have good contact.
For the CPC464, try disconnecting the keyboard, see if the cursor stops advancing, then your problem is the keyboard (clean the folio). Otherwise it's probably the AY8912 chip.
Thanks. Have some free time now, will try it out and report back.
Quote from: Duke on 21:41, 13 May 18
For your CPC6128, it sounds like it could be the ram, but it could be other things also.
First things to try is re-seat the Z80 and the Gate Array (40010 chip), to make sure they both have good contact.
For the CPC464, try disconnecting the keyboard, see if the cursor stops advancing, then your problem is the keyboard (clean the folio). Otherwise it's probably the AY8912 chip.
Ok, reseated both chips and nothing. The behavior now is just like the last of my initial tests. Powers on straight to grey screen with black border :(
That is looking like a ram problem then. Are any of the ram chips getting hot?
Quote from: Dagger on 00:09, 14 May 18
That is looking like a ram problem then. Are any of the ram chips getting hot?
I checked, and at least to the touch they feel cold.
What felt warm was the small chip "Amstrad 40031"; and considerably hotter, one bigger chip to the lower left of the board labeled "HD46505SP" (and below that HD6845SP")
Quote from: Duke on 21:41, 13 May 18
For your CPC6128, it sounds like it could be the ram, but it could be other things also.
First things to try is re-seat the Z80 and the Gate Array (40010 chip), to make sure they both have good contact.
For the CPC464, try disconnecting the keyboard, see if the cursor stops advancing, then your problem is the keyboard (clean the folio). Otherwise it's probably the AY8912 chip.
Ok, the CPC464, unplugging stopped it, cleaning did nothing but it seems it's just mechanical, as raising the spacebar a bit did produce a "click" sound and it's not doing it now. Not sure if there is still an issue, as I can still "raise it"a little after pressing it a few times, but at first glance seems to be OK.
Quote from: walterg74 on 22:59, 13 May 18
Ok, reseated both chips and nothing. The behavior now is just like the last of my initial tests. Powers on straight to grey screen with black border :(
I would say that CRTC, Z80 and Gate Array seems ok.
One of the first things that's done in the lowerrom at start up is setting up the CRTC registers which produces the grey screen with black border.
So most likely it's your ram chips, if you let it run for some time, are you sure none of them gets hot?
Oups, if your 40031 chip gets very hot, that's the one controlling the expanded ram of the 6128. That could well be it. You could try to bypass it by removing the chip and setting some LK's on the mainboard (probably you can find which somewhere in the WIKI), so it only uses the base 64KB ram.
Quote from: Duke on 15:07, 14 May 18
I would say that CRTC, Z80 and Gate Array seems ok.
One of the first things that's done in the lowerrom at start up is setting up the CRTC registers which produces the grey screen with black border.
So most likely it's your ram chips, if you let it run for some time, are you sure none of them gets hot?
Oups, if your 40031 chip gets very hot, that's the one controlling the expanded ram of the 6128. That could well be it. You could try to bypass it by removing the chip and setting some LK's on the mainboard (probably you can find which somewhere in the WIKI), so it only uses the base 64KB ram.
Yeah, I mean ram is definitely cold. The 40031 is "warm" (not sure how hot is too hot in this case). It is socketed, so removing is easy. Not sure what I have to set that you mention about LKs?
Also, what about the other chip? Is it normal for that one to be hot? The schematics say that's the VDU controller?
Quote from: Duke on 15:07, 14 May 18
I would say that CRTC, Z80 and Gate Array seems ok.
One of the first things that's done in the lowerrom at start up is setting up the CRTC registers which produces the grey screen with black border.
So most likely it's your ram chips, if you let it run for some time, are you sure none of them gets hot?
Oups, if your 40031 chip gets very hot, that's the one controlling the expanded ram of the 6128. That could well be it. You could try to bypass it by removing the chip and setting some LK's on the mainboard (probably you can find which somewhere in the WIKI), so it only uses the base 64KB ram.
I think I found it. Is it this?
http://www.cpcwiki.eu/forum/amstrad-cpc-hardware/need-some-help-fixing-cpc6128-not-booting/msg75204/#msg75204 (http://www.cpcwiki.eu/forum/amstrad-cpc-hardware/need-some-help-fixing-cpc6128-not-booting/msg75204/#msg75204)
If so does that mean I take out the 40031 and since it's socketed I can just connect thise pins with a ciuoke of wires?
Quote from: walterg74 on 16:33, 14 May 18
Yeah, I mean ram is definitely cold. The 40031 is "warm" (not sure how hot is too hot in this case). It is socketed, so removing is easy. Not sure what I have to set that you mention about LKs?
Also, what about the other chip? Is it normal for that one to be hot? The schematics say that's the VDU controller?
I guess is a question of definition :)
My CRTC (the vdu) is about 45 deg. C atm. so I guess it can be quite warm. I don't have an open 6128 so don't know about the PAL chip, if its only "warm" I guess, it's nothing to worry about.
For the LK's here is the link: http://www.cpcwiki.eu/index.php/LK_Links Look under "CPC6128:LK5,LK6,LK8 - Disable 128K RAM Banking" and as for swapping the banks you can do as in the thread you linked to.
For hot RAM, you will almost burn your fingers touching them :)
Quote from: Duke on 19:46, 14 May 18
I guess is a question of definition :)
My CRTC (the vdu) is about 45 deg. C atm. so I guess it can be quite warm. I don't have an open 6128 so don't know about the PAL chip, if its only "warm" I guess, it's nothing to worry about.
For the LK's here is the link: http://www.cpcwiki.eu/index.php/LK_Links (http://www.cpcwiki.eu/index.php/LK_Links) Look under "CPC6128:LK5,LK6,LK8 - Disable 128K RAM Banking" and as for swapping the banks you can do as in the thread you linked to.
For hot RAM, you will almost burn your fingers touching them :)
Ok, will take a look at the link and see if I can figure it out. Regarding the RAM, definetly not the case the. (The symptom at least), since as mentioned, they're all just cold.
The PAL (40031) will get relatively hot while running. They also rarely fail.
For the bank swap you can do as you guessed: Remove the PAL IC and link the pins with jumper wires as described in the post from gerald. This then get the CPC to use the upper 64K instead of the lower RAM.
Bryce.
Quote from: Bryce on 07:51, 15 May 18
The PAL (40031) will get relatively hot while running. They also rarely fail.
For the bank swap you can do as you guessed: Remove the PAL IC and link the pins with jumper wires as described in the post from gerald. This then get the CPC to use the upper 64K instead of the lower RAM.
Bryce.
Haven't gotten to test this yet, will try to do it tomorrow, but just to get ahead of things, what part numbers can I look for to buy for RAM replacement..?
The original CPCs used 4164 RAM ICs, but there are also some equivalents such as the 3764 or the 4264 which can also be used. They need to be 64Kx1bit DRAM ICs, pin compatible with the 4164.
Bryce.
Edit: I should also add, that you shouldn't mix RAM chips, so ideally you need 8 of one of those ICs and not a mixture.
Quote from: Bryce on 07:49, 23 May 18
The original CPCs used 4164 RAM ICs, but there are also some equivalents such as the 3764 or the 4264 which can also be used. They need to be 64Kx1bit DRAM ICs, pin compatible with the 4164.
Bryce.
Edit: I should also add, that you shouldn't mix RAM chips, so ideally you need 8 of one of those ICs and not a mixture.
Hi Bryce,
While I haven't gotten around to doing this yet (will soon), I have my brother travelling around the states and think it's a good opportunity to buy some ram, both in case that's the problem and proactively/for soares if that's not the case.
Anywho, which would you recommend, just the 4164s? Could I get them somewhere like mouser/etc? (I would need to ship them to CA).
I would go for 4164's if you can get them. However, I very much doubt that any component suppliers like Mouser still stock these or any other similar RAMs. ebay is the only place to find them now.
Bryce.
Quote from: Bryce on 14:40, 04 July 18
I would go for 4164's if you can get them. However, I very much doubt that any component suppliers like Mouser still stock these or any other similar RAMs. ebay is the only place to find them now.
Bryce.
While I did not find them at mouser, I did find some at Jameco (In fact I think this is where I had purchased a set of 16 for an Apple IIe a couple of months ago)
They do say "refurbished", what do you think? Also what about speed? Anything on that requirement? These seem to be 150ns...
https://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?catalogId=10001&freeText=64kx1&langId=-1&storeId=10001&productId=41662&krypto=xRkYkiWnWr9kvzd3ruQdQmb8PyDuigMVYRvh56BH6%2BAHF8vwAjsYZ94lTRTivmnZywpl%2BNgcfUfxjcJsggq7O78kdKJshC1ObJGf2y7GbSg%3D&ddkey=https%3AStoreCatalogDrillDownView (https://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?catalogId=10001&freeText=64kx1&langId=-1&storeId=10001&productId=41662&krypto=xRkYkiWnWr9kvzd3ruQdQmb8PyDuigMVYRvh56BH6%2BAHF8vwAjsYZ94lTRTivmnZywpl%2BNgcfUfxjcJsggq7O78kdKJshC1ObJGf2y7GbSg%3D&ddkey=https%3AStoreCatalogDrillDownView)
250ns or faster is fine.
Bryce.
Quote from: Bryce on 15:16, 04 July 18
250ns or faster is fine.
Bryce.
Cool!
Ok, the I take it as a yes that the product I pasted the link for will work fine :)
I guess it would be wise to also buy sockets..?
Yes and yes. It's always a good idea to add sockets if you've removed RAM.
Bryce.
Quote from: Bryce on 07:49, 05 July 18Yes and yes. It's always a good idea to add sockets if you've removed RAM.Bryce.
Ok, don't seem to be having luck... I tired the bank swap in a very.. "artisanal" way with just a couple of wires I had, but when I power I still just get the grey screen. Just in case it doesn't seem like it in the picture, none of the wires are touching.
Ok, while I was waiting for replies, I went ahead and removed all the RAM chips.
I also installed sockets on the left side. This is as far as I got before the missus told me to stop qith all the smoke, so will have to do the second half tomorrow.
Since I already had the RAM I had purchased for the Apple II that I hadn't yet used, I decided to gove those a go.
I thought the left side was bank 0, so tried powering it on, same result :(
I then read actually that is bank 1,so I decided to do the remove PLA and bridge pins thing a try with hope, but unfortunately I still get the same resukt with that :(
I will do the other half anyway tomorrow, but any help on how to proceed with what I just experienced?
The next most likely suspects would be the input/output buffer ICs from the RAM.
Bryce.
Quote from: Bryce on 08:34, 09 July 18
The next most likely suspects would be the input/output buffer ICs from the RAM.
Bryce.
Which chips were those?
IC114 and IC115
Bryce.
Quote from: Bryce on 12:28, 09 July 18
IC114 and IC115
Bryce.
Great... more buying and worst of all waiting for it to arrive ???
Any particular variant, or any plain old 74LS244 & 74LS373..?
I would think the PAL would not be the culprit, being I did the remove and rewire trick, right..?
Unless of course the upper RAM is also damaged? But that would seem less likely as it would mean two things had failed simultaneously.
Bryce.
Quote from: Bryce on 20:17, 09 July 18
Unless of course the upper RAM is also damaged? But that would seem less likely as it would mean two things had failed simultaneously.
Bryce.
Well in any case I finished the second half today and installed all new RAM, same thing.
Try testing the output pins of the buffers mentioned above, but also the outputs of the multiplexers (74LS153's). All of these could could be the cause of the issue.
Bryce.
Quote from: Bryce on 20:40, 09 July 18
Try testing the output pins of the buffers mentioned above, but also the outputs of the multiplexers (74LS153's). All of these could could be the cause of the issue.
Bryce.
What should I look for? A specific voltage with a multimeter? Or do I need a fancier tool?
It's more difficult to be 100% sure with a multimeter, but you could at least check if any bits are "stuck": ie: any output that is always 5V or always 0V is suspicious. A changing bit will read a value somewhere in between (2V - 4V) depending on the ratio of 1's and 0's.A scope would obviously be a better choice of tool.
Bryce.
Quote from: Bryce on 12:28, 09 July 18
IC114 and IC115
Bryce.
Ok, FINALLY got a 74LS244 and a 74LS373 today (plus 2 sockets ??? ).
Guess I will try replacing these two and see what happens...
Quote from: Bryce on 12:28, 09 July 18IC114 and IC115Bryce.
Well, removed those two, soldered a pair of sockets and inserted replacements. Nada...
Same gray screen with black border :(
Bumpity bump...
Getting a bit frustrating buying and replacing chips with no improvement :(
Do you think I should try the last thing replacing the 74LS153's ..?
Sometimes when turning it on I get something like the attached, but 95% of the time it's just the grey screen with black border.
Anything else I may be missing..? :'(
Another variant
And a third variant...
Quote from: walterg74 on 13:25, 06 August 18
Bumpity bump...
Getting a bit frustrating buying and replacing chips with no improvement :(
Do you think I should try the last thing replacing the 74LS153's ..?
Sometimes when turning it on I get something like the attached, but 95% of the time it's just the grey screen with black border.
Anything else I may be missing..? :'(
Unfortunately you are trying to do a "blind repair" which often ends in disappointment and frustration. Normally you test first and change the required parts, but without proper test equipment, it's just taking educated guesses. Sometimes you get lucky and hit the jackpot with the first guess, but often it ends like this.
Bryce.
Quote from: Bryce on 18:38, 06 August 18
Unfortunately you are trying to do a "blind repair" which often ends in disappointment and frustration. Normally you test first and change the required parts, but without proper test equipment, it's just taking educated guesses. Sometimes you get lucky and hit the jackpot with the first guess, but often it ends like this.
Bryce.
Yes, that's unfortunately the case. Don't have any test equipment beyond a multimeter, but even so, it's useless without the knowledge of what to look for. I do have a 464, but besides the CPU (which I did test anyway just in case), I don't have anything else I can use it to test I think.
I was only going by the list detailed on another post for the observed symptoms. (40031,RAM, 244/373,etc.)
Would you say if it stopped worked suddenly (in front of my eyes) that in theory it should be one component?
What would you suggest?
Thanks, Bryce.
Electronics almost always dies instantly in front of your eyes, it doesn't necessarily point at whether one or more components have died, because one failure can cause further failures within milliseconds. I'd suggest that if you really want it to work again, you should send it to someone to have it fixed. If you can't find someone local you can always send it to me.
Bryce.
Quote from: Bryce on 17:28, 07 August 18
Electronics almost always dies instantly in front of your eyes, it doesn't necessarily point at whether one or more components have died, because one failure can cause further failures within milliseconds. I'd suggest that if you really want it to work again, you should send it to someone to have it fixed. If you can't find someone local you can always send it to me.
Bryce.
Thanks Bryce, I understand.
Today I won an auction for a second 6128 that is in working order. I think with this I should be able to at least check the components more properly. Although of course it wouldn't rule out things that can't be tested with this method. And then I will just have to send it to someone who knows... (there is definitely nobody local)
While I am very far away, where was it that you are located?
Germany, but it shouldn't be too expensive if you just send the PCB.
Bryce.
Gesendet von meinem Motorola DynaTAC 8000x mit Tapatalk
Hi All,
Posting back because I really hate when I see people asking things and then never posting back to say what happened / how things turned out...
I had gotten a second working 6128, and tried swapping the other non-standard chips with the same results...
I contacted someone I was refered to who was local, and had the expertise and proper tools to take a look at it. What they found was that simply some traces had been corroded under the mask (said probably humidity at some point?) and after they repaired them, the machine booted up normally again :)
When he sent me the pictures I first thought "how on earth did I not see this??" But then he told me it wasn't visible but was under the mask :D
Anyway, happy with my once again working 6128, and now I have two I guess :)
Thanks to all for the help!