News:

Printed Amstrad Addict magazine announced, check it out here!

Main Menu
avatar_biishop

6128 Voltage problem on Z80

Started by biishop, 15:36, 14 July 19

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

biishop

Hi,

I've got some trouble after playing with an serial extension on my 6128.
I'm now front of a black screen of the death. After pluging the device and begine to try it, i got some freez and some char automaticly type on the screen. A switch off, wait and switch on and a grey screen with black border appear.
My forehead was starting to sweat, i switch off and remove the interface. Wait couple of hours. My favorite blue screen reappear again.
I am a little sadistic and probably too curious. I check the hardware of the interface, plug it into cpc again. And switch on, normal boot, test again and suddently a new freeze with this time the motor of the disc drive switch on alone. All of this things make me restart again. And after this nothing more reappear on the screen now, the red ligth is up, the typical sound is here, but nothing on screen, no bip with the del key.


The first step I follow is the advice from Bryce, test the presence of 5v on all the IC and from this step I allready have a problem on Z80, CRTC, Gateway. ROM chips, HAL, 765 and Logic IC are with something close to 4.8v.
The most hotest chip est le Z80 and after the CRTC (the original but i can easyly keep my finger on it)
Ram chips are cold.

2.2v Z80
2.7v 8255& Sound Chip
1.98v 40010
0v CTRC ( this make me surprise because it a bit lukewarm)


I do a test with a spare Z80 but it still go in the same way.


My skill is a bit light in hardware, but I will try to at least find why and where could be at the origine of this low variable voltage on the CPU. It's variable and weaker and weaker. This morning for example, i was surprise to find 5v on it but after 1minute or less it went down to 2.5v. It can be a component that has failed or is about to collapse.


Do you have basic advice check list to follow in my case?
A list of components to check in the power ligne ?
A capacitor, a typical resistance known to be check in the standard verification.


Thanks for you help whatever it is.


CC

biishop

Btw, I've test the Z80 on a prototype board with a clock Arduino and Nop loop instruction.
The address pointer is correctly incremented.

gerald

Quote from: biishop on 15:36, 14 July 19
2.2v Z80
2.7v 8255& Sound Chip
1.98v 40010
0v CTRC ( this make me surprise because it a bit lukewarm)
Where (on which pin) did you measure these voltages ?


biishop

#3
Your question make me puzzled  :o
I just realized that I took as principle the fact that the connector +5v was like the logic component that I tested that opposite to the pin 1.

A take new measurment with data sheet this time,

So I reverify
Z80 pin11 - 4.95v
CRTC pin20 - 4.95v
GateArray pin9 4.95v, pin21 1.29v, pin38 3.42v
AY-9813 pin13 3.0v
D8255AC pin26 4.95v

I'm so a nooby sorry

gerald

Quote from: biishop on 19:24, 14 July 19
Your question make me puzzled  :o
I just realized that I took as principle the fact that the connector +5v was like the logic component that I tested that opposite to the pin 1.
You're not the first nor the last  ;)

Quote from: biishop on 19:24, 14 July 19
A take new measurment with data sheet this time,

So I reverify
Z80 pin11 - 4.95v
CRTC pin20 - 4.95v
GateArray pin9 4.95v, pin21 1.29v, pin38 3.42v
AY-9813 pin13 3.0v
D8255AC pin26 4.95v
5V rail seems good.
But you still have some pin wrong ;D
- AY VCC is on pin 3
- Gate array :
  - pin 6 and 25 on 40010
  - pin 33, 9 38, 22 on 40007/40008 (with a heatsink)

What are the current symptoms with no extension connected : black border and grey screen or all black ?

biishop

#5
So then

AY pin3 4.95v
40010 pin6&25 4.95v too

At the beginning I was doing some test with my serial interface and bizar symptoms appeared like freeze with writing in the memory screen, some series of special ascii char type randomly, I hear also the relay click from the tape.
At this point I had a gray screen with black borders. I had this state randomly but I didn't take care about it. I should have been wary of it but I wanted so much to make this interface work.
After one or two restart I the blue screen come back (when I think back now I probably mistreated the power button a little like in the old time it was such a practical but I think the ancestor has anyway not appreciate).
In the late evening the CPC was stuck on it, no more correct start.
I left the night and the next day everything was ok.I went back to test the communication with the interface I was able to load my backup on a floppy disk of a small program to initialize the communication and suddenly re blocking of the screen.
But this time with the reboot a completely black screen. I try the next day but nothing more this time.
I read a lot of thread with similar problems, with explanations for the CPC boot mechanics, the copy of the rom in the ram, I do not have a hot ram chip. But now with the black screen, I wonder what would be the tracks that I could follow.

Btw Thanks for you support

biishop


biishop

#7



I do some basic test with an Arduino Oscillo and here they are :


I suppose  :)  something pulse behind the Clock pin of the Z80


But for the INT pin16 the signal is flat  :(




Bryce

Hi Biishop,
      they are good signs. It means that your computer is basically working, the machine is powering and the Gate Array is doing what it should. I suspect that the RAM or ROM may have failed and the fact that you were testing a serial expansion at the time might just be a coincidence. Try connecting your Arduino scope to the Data in (pin 2) and Data Out (pin 14) of each RAM chip and see if data is moving around.Also try probing the Data pins (pins 11 - 13 and 15 - 19) of the ROM chip to see if it has problems. Any pin that always stay high or always low could indicate an issue.


Bryce.

biishop

#9
Sorry I miss again the correct data sheet and i modify my message .
I want to go too fast it's the problem of my life.

Now I turn arround the GA and compare to the other chip I feel it oddly cold ...

biishop

#10
Job done here is the result.
Each checked pin have no flat signal
Ram signal are some time different but its the same compare to the opposite one like (IC119 & IC127 - IC120 & IC 128 etc)

I have some doubt with ram14_1 signal.

Here is screens shot

biishop

Here is a large rom14 capture from


IC119-127-120-128-122-130 (picture 1)


IC121-129-123-131-124-132-125-134 (picture 2)


IC124-133 (picture 3)





gerald

Quote from: biishop on 16:12, 15 July 19
Job done here is the result.
Each checked pin have no flat signal
Ram signal are some time different but its the same compare to the opposite one like (IC119 & IC127 - IC120 & IC 128 etc)

I have some doubt with ram14_1 signal.

Here is screens shot
RAMPIN2_1 is highly suspect if it's what you seen on pin 2 of one of the RAM CHIP.
Pin 2 is the data input of the RAM and is directly connected to the Z80 data bus  (as well as to the PPI/FDC/ROM and ram read buffer).
While there is no time scale on the picture, it is very unlikely that this pin stays high like that.
If any of the driver force it to 1, there is no way the ROM can be read and therefore no way the CRTC can get initialised.
Do you have a chance to test the Z80 and confirm it is working ?



biishop

I do a Basic test with the Z80 and the Arduino, I simulate Clock with a DO,and a NOP operation and I saw the counter of the pointer program growing with 3 led on the adress bus thats all.
I also replace the Z80 by another one, and I got the same result. I mean on the Amstrad PCB.

gerald

Quote from: biishop on 18:35, 15 July 19
I do a Basic test with the Z80 and the Arduino, I simulate Clock with a DO,and a NOP operation and I saw the counter of the pointer program growing with 3 led on the adress bus thats all.
I also replace the Z80 by another one, and I got the same result. I mean on the Amstrad PCB.
OK, so we know that the Z80 is not the root cause.
ROM and CRTC are the next to check. Do you have any way of using an external ROM (XMEM ?)

Bryce

I'd agree with Gerald, but I have a hard time understanding that weird Arduino scope! What on earth is that scale on the side meant to be? The pin is at 800?? 800 what? Bananas? and there's no time base at all. The scope really isn't telling you anything. If it's possible to set up the scaling, then set the side scale to 0 to 5V or whatever the full scale is and set the sample rate to at least 16Mhz so that it cleanly displays bits.

Either way, I'd also say you have a stuck bit on RAMPIN2_1

Bryce.

biishop

#16
Hi,


I do not have any way atm to try an other ROM. I look XMEM quickly but I only found some Marvel information  :D .
edit: I found it, its an expension device, I dont have it.


I know, I will try to go deeper in the Arduino Occilo project to discover this crucial information too.


Take u aware. Ty

biishop


Some news about my problem because I'm still on it but waiting for hardware now.
I ordered a small oscilloscope to make quantifiable measurements this time and more reliable as well.
I ended up realizing the initial error made on this machine a voltage error of a transformer to the fact that it had to work for a good time in 7.5v instead of 5v.
This has to be due to some component.
For the most basic of them they are also in command: 2 rom with the programmer and the bank of ram as well as supports tulip.
I also have a new machine but this one is more recent and does not have the same PCB on the other, I still hope to make some comparisons thanks to that.
See you soon to hope for a successful rescue.

biishop

Hi mates,
I'm back with some new nooby investigations. I buy a 5Mhz oscilloscope and I followed different thread to reach a conclusion around my CRTC which receives a clock of 1Mhz
but which doesn't send anything in vsync or hsync. I changed the chip, strictly the same model. Same thing, the only difference on the news that I have no frequency on the D0, D1, D3 ... compared to the old one. Here I am still drying. What could be the tracks to follow? Change the component again? I have some material now and I can do some measurements.
Thanks for help, plz


biishop

#19
Here other information about what I try to do during this time. I change IC117 TC74HCU04P, check resistance like recommended, change 8xRAM Bank0 MN4 . Change IC110 74LS132 and like I allready say change the CRTC UM6845 IC108. A also find a plan for a 40010 but without any sureness it was functionaly ofc. My board is a Z70290 model. Is there any trick to test IC independently?

Powered by SMFPacks Menu Editor Mod