A1200 rev 1D1 CIA problem?

[Firestone]

Hi everyone.

I dug out one of my spare A1200 motherboards today, and my initial thought was to recap it.
It seems fine, no dull solderpads or anything. Boots fine and everything works.

So now I tested it with Amiga Test Kit 1.21.

All tests passed, including sound, video and memory tests.

But when testing CIA timers:



EDIT:
DiagROM 1.2:




Any idea where to start to hunt this one down? Defective CIA-chip?

EDIT: Board has never been recapped before, but can that cause this to happen?

[Kin Hell]

CIA chips A & B are the same part number. - 8520PL
See here: https://bigbookofamigahardware.com/b...t.aspx?id=1485

You could explore the PCB here. https://www.amigapcb.org/index.php
There are two versions of A1200 board there; 1D4 & 1D1 (beta)

[Daedalus]

Failed capacitors in themselves won't cause symptoms like that, but if they have leaked, then it's possible that the leakage has caused damage to the PCB somewhere. Such damage might not always be visible, though if the pads are clear and shiny then it's less likely I guess. The VSync signal comes from Alice, it's worth checking that that circuit is fully intact, and that the pull-up resistor on it is working as expected. I haven't seen a CIA fail in this way, but I guess it's possible. I'd rule out everything else first.

[Firestone]

Originally Posted by Daedalus

Failed capacitors in themselves won't cause symptoms like that, but if they have leaked, then it's possible that the leakage has caused damage to the PCB somewhere. Such damage might not always be visible, though if the pads are clear and shiny then it's less likely I guess. The VSync signal comes from Alice, it's worth checking that that circuit is fully intact, and that the pull-up resistor on it is working as expected. I haven't seen a CIA fail in this way, but I guess it's possible. I'd rule out everything else first.

Thanks for the reply.
I have replaced all capacitors now, and checked continuity on all of them. No leakage found under them, and all solderpads are okay.

I did some more measuring based on what you said.

U2 (Alice) -> E251R : OK
E251R 68ohm OK
E251R pin 1 -> U7 pin 21 (tick) : OK

According to AmigaPCB, this also travels to R250 on the bottom of the board.
I measured it, and according to AmigaPCB, it should read 4,7kohm, but I have measured it to 1,2kohm....

[klx300r]

got same error few years ago and installing new CIA chip solved the issue.

[Jope]

It's surprisingly common for the tick signals to fail in those surface mount CIAs. I've replaced a few in A1200s in the last decade or so.

[Firestone]

Yeah, I've come to the conclusion that it might be the CIA-chip itself.
And yes, it's strange how they just fail without any notice.

Where do you guys get sockets for these? I guess the CIA-chip itself must be sourced from ebay and other sites like that....

[Daedalus]

Yeah, sounds more like the CIA is acting up alright. Sockets are standard PLCC-44 surface mount sockets that can be found at any of the major electronics retailers. But fitting the sockets can be a fiddly job to do by hand and can introduce its own problems, so think carefully about whether you really want to socket it or not.

[wiz12]

There was a seller on Amibay selling these custom chips including the CIAs but I'm not sure if he is still selling them since his advert was from a few years ago. Analogic is the only other place selling them but prices are far from cheap.

https://www.analogic.co.uk/catalogsearch/result/?q=cia


These are the sockets I use and have had no problems.

https://www.mouser.co.uk/ProductDeta...XeZ%2Fbg%3D%3D

[Firestone]

Thanks a lot everyone.

[klx300r]

i'm +1 for no socket

[Firestone]

Why do you recommend against sockets?

[klx300r]

Originally Posted by Firestone

Why do you recommend against sockets?

I agree with what Daedalus said above as over the years I've found that sockets have been an issue time and time again. I also have a good friend of mine who makes a living as an electronics technician that tells me the very first thing he does when he gets equipment in for repair is get rid of all sockets as they're just too much of an issue i.e people not properly removing/inserting chips/ bad connection joints etc

[Kin Hell]

Originally Posted by Firestone

Why do you recommend against sockets?

Sockets were always an issue back in the day & Commodore were on it.
They get brittle with age.
Contacts on edges of the chips can miss the contacts on wall of the socket due to bowing/shrinkage.
You can exert too much pressure making sure they are seated & split the socket.
Less observant people can attempt to insert the chip in the wrong orientation for the socket, again making the socket split.
When Commodore released their later Amiga's, they opted for surface mount for better reliability (no movement with heat changes) & production costings.

In Engineering terms, it was just natural progression.

[Firestone]

Yeah that makes sense. I have experienced more than once changing sockets on the a500 and a4000. The busterchip gave me headaches on the a4000 several times too.

[Kin Hell]

Originally Posted by Firestone

Yeah that makes sense. I have experienced more than once changing sockets on the a500 and a4000. The busterchip gave me headaches on the a4000 several times too.

Indeed & the A4KD Buster revision changes from 9 to 11 for DMA issues went hand in hand as the necessity of C= wanting to go surface mounted for manufacturing costs.

As Daedalus mentioned earlier, sockets are very tricky son's of guns to re-work.