CPC 6128 - white screen, black border

[McArti0]

..

[McArti0]

Yesterday 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. 
Btw. If you don't have time, we won't anything.

[eto]

Yesterday 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? 

Dandanator? 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.

[Bryce]

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.

[eto]

The crystal divination ball for CPC duck diagnostics did not work.

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.

[abalore]

If you are willing to pay the shipping to Spain and back, I can fix it for you.

[McArti0]

The crystal divination ball for CPC duck diagnostics did not work.

So maybe (just maybe) (...) but now it has been worsened. We will never know.

Next crystal divinaton ball too

Unfortunately, I won't make anyone think.

[mmldrm]

If 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. 😊

[mmldrm]

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.

[McArti0]

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

[Gryzor]

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. 

[McArti0]

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.

[McArti0]

@mmldrm

You 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.

[abalore]

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.

If information is valued regardless of who said, the place where you lived is just an argumentum ab auctoritate fallace.

[McArti0]

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.
 

[mmldrm]

For better or for worse... I'm back. Apologies to anyone who groaned upon reading that. 

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.

*

[McArti0]

OMG! You are alive.

@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...