Amiga 500+ CIA B issue...

[kbj]

Hey All,

I'm repairing a 500+ that I picked up and with AmigaTestKit running the CIA tests, it's reporting an IRQ failure on CIA B - but everything is good with CIA A. When I swap around the chips, the failure stays with CIA B (even though the working CIA A 8520 is now in that socket).

Any ideas? Is there anything else I can check?

Cheers,

KbJ

[Daedalus2097]

CIA-B uses INT6, so it's possible there's an issue with the traces on that line somewhere. INT6 is pin 21 of CIA-B, and it goes to Paula pin 18 as well as the side expansion port. It should also have a 10k pull-up resistor to 5V, so check continuity between those pins and resistance to ground and 5V, see if anything is amiss.

 

The Paula socket sometimes gets corroded when there's been a leaky battery - the Gary socket is usually far worse, but it might be that the Paula socket needs to be changed.

[kbj]

@Daedalus2097 Appreciate that!  I was looking at the schematics just now and see that.  The Paula socket looks fine - the Gary socket not so much - that and the Agnus socket are pretty green in bits.  I'll meter it out as you say, but may just replace the Gary socket and see where I am.

 

Thanks for the advice!

 

KbJ

[Daedalus2097]

Yeah, it's worth replacing the Gary socket as all sorts of weird things can go on when Gary's not quite right. And there's a fair chance it'll stop working further down the line if it's bad. Likewise for the Agnus socket, though that's a little more fiddly to replace.

[kbj]

@Daedalus2097 Replaced the Gary socket as the Paula socket metered out just fine - and also replaced the Angus socket as that looked pretty corroded once I'd removed Agnus from it.

 

No change mind you - and I checked the continuity of the INT6 line - all good to everywhere it went.  There is a bit of corrosion on U10,11,12 and 13 - especially 12/13.  I think they're more to do with the data path to memory though?

 

KbJ

[Daedalus2097]

Yeah, they're the data bus to the custom chips (including chip RAM). Strange that everything's checking out. Do you have a scope? It might be worth checking if the interrupt line is actually triggering during the test. It's unusual but it does happen that Paula can be faulty.

[kbj]

Hey @Daedalus2097  - I replaced Paula from a working board and same issue.  It may be worth checking with a scope anyhow I guess in case there is some weird issue with the board itself - even though I already metered it out and it looked fine.  I could also put this Paula in the other board to make sure it works.

 

I'll give that a go.  Just waiting on new sockets so that I can replace the Paula socket.

 

Cheers for the help!

KbJ