Wednesday, February 28, 2018

Yet Another A500+ Motherboard - Part 1

This is getting to be habit... Another A500+ motherboard has arrived. This one just gives a black screen despite the best efforts of the previous owner.

Next patient please!

On first inspection the repair wires look to be correcting the classic signs of battery damage. They were pretty well done but a bit thick for my liking and, as I have some uber thin kynar wire knocking around, I decided to remove them with the intention of replacing them.

Previous Repairs
Then I inspected the board again. To my surprise, it looked very, very clean. There's not much damage from the battery. I poked around with a screwdriver to expose where the battery had leaked and remove the damaged conformal coating.  Where the copper was exposed I ran a smidgeon of solder across the tracks. But there wasn't really much damage at all.

One of the least damaged I've seen...
Continuity testing revealed that the only issue seemed to be from the cluster of resistors/diodes near the AUDIO FILTER section on the board. On the opposite side of the board one of the pads has fallen off probably through exposure to the battery leak and then a hot soldering iron. No biggie to sort this out but it shouldn't cause a black screen. Hmmmm.

Missing pad - no problemo!
I tried testing around Gary who is another prime candidate for causing a black screen, especially if his socket is corroded by the battery. No problems.


Black screens are listed as being a CPU fault but, in reality can be caused by any number of things. Dodgy logic chips, broken tracks at Gary's socket (as mentioned), busted ROM etc.

Anyway, just to be sure I swapped over the CPU with a spare but this made no difference. Then I swapped Gary with a spare but this still made no difference. I don't have a spare 8375 Agnus or a 2.04 ROM but I do have a couple of 1.3 ROMs. Just out of interest - I wasn't even sure the 1.3 ROM would work on an A500Plus - I swapped out the ROM for one of these.

And it booted.

Well, this is awkward..
This is a surprise to me as I would normally expect ROM faults to give a red screen. In any case, the board boots to the familiar KS1.3 white screen. This is good news but is a bit of a pain as the 1.3 ROM with only boot from an internal floppy drive i.e. DF0. The big advantage with 2.04 is that it will happily boot a bare board from an external floppy drive. To do any more testing I have to dismantle my own A500+ and swipe the drive from it for a while which is a bit of a pain...

Before I start dismantling I decided to carry on looking over the board in case of any other issues. There were a couple of minor things. The pads at jumper 9 and jumper 7A are gone. This is partly my fault. Slightly over exuberant cleaning, battery acid and the heat of a soldering iron were too much for them. They were there when I started cleaning and, well, you can see the result. They were in poor condition but had been that way since before the previous owner. Their non-existence is not really an issue for a standard Rev 8A board though.

JP7A - Gone, but not forgotten.

J9 - Missing, presumed gone.
The mouse and joystick ports have pins that are damaged. I'm not sure if I did this. Oops. Again, it's not a big deal as I have a big bag of these connectors that I accidentally over-ordered a couple of years ago...

It wasn't me...honest..(probably was actually)

In part 2 I will dismantle my A500+ and attempt to test whether there are any other problems on this board. I might also replace the mouse and joystick connectors. 😀

Thursday, February 08, 2018

Even More Memories...

Well it seems my detective work was correct. After testing the 512Kb memory expansion that was very kindly sent to me by Neil, I'd come to the conclusion that one of the four DRAM chips was dead.

After posting the blog post to the Amiga Facebook group one of the other guys there offered a DRAM chip to see if I was correct. I gladly accepted and guess what arrived in the post today? Yes, a bill from RS Components. But also, a small envelope with an expertly packed 20pin DRAM chip. Yay!

Well Wrapped DRAM
This DRAM chip is different from the others on the expansion board in a couple of respects. Firstly, it's made by Motorola (just like my phone). Secondly, it is rated at a speed of 70ns. That's a lot faster than the ones on this board which have a speed of 120ns. Hmmm. Not sure if that will make a difference or not..

Just to be sure I found the data sheets for the old DRAM and the new DRAM chips, and checked that they have the same pin out. Which they do. Nice.

Motorola DRAM Pinout

V53C DRAM Pinout (original on board)

So, time to just bite the bullet and put this sucker in there.

New Chip Inserted
Moment of truth....

No magic smoke released. Phew. White screen. OK, patience - there's no internal floppy attached to this test motherboard but there is an external drive. It boots and runs the test disk no problem.

Memory test shows 1.5Mb of total memory, just like before. This time though:

It works! This has been running through 84 iterations of the memory test with no problems. Nice! Huge thanks to Michael for the DRAM and Neil for the originally broken RAM expansion.

PS Memory test is now over 150 rounds with no faults. Oh, yes. ;)

Thursday, February 01, 2018

More Memories...

As I was perusing the Amiga Facebook pages I came across a post by a chap who had dragged his A500 out of storage and was going to give a young relative a lesson in retro gaming. Sadly, this was not to be as the A500 refused to start.

With the quantity of helpful people on the group several pieces of advice were given and it transpired that the memory expansion was the cause of the non-boot. The OP (that's 'Original Poster' for you non-hip and trendy people out there..ahem) was about to throw it in the bin but I asked if he could send it to me instead as I was looking for another repair project. And he did.

Package Arrive Safely
It's a fairly generic 512Kb RAM expansion as was the trend in the late 80's and early 90's. It has four 256k x 4 bit CMOS dynamic RAM chips and, bizarrely, three of them are socketed. The battery was left on too but this was quickly despatched.

Argh! Curse you Varta!

Varta removed - clean up on aisle 5

I took some vinegar and applied it liberally to remove the battery residue (fizz-tastic) and then washed away the vinegar with some distilled water and allowed it to dry thoroughly. After a bit more clean up I was ready to test.

First test...
I used my A500+ spare motherboard with an external drive and the excellent Kier's test disk. There is an intermittent fault with this motherboard in that it sometimes won't boot and I get a yellow screen. Normally, I have to re-seat Agnus and then it works. Sometimes it works just by breathing near it. That's for another time. At this point, the motherboard is working...

...and so is the RAM expansion. Well that was easy.

Before I got too cocky I decided to run the RAM test and see what transpired. It was not necessarily good.

Aw, crap.

I re-seated the three RAM chips and got the same result. It looks like there is one dodgy chip in there but the question is, which one?

So I labelled them 1, 2 and (duh) 3. Then ran the RAM test again and took a picture of the board and the results. Then I moved the chips around and did it again. Rinse and repeat. Very quickly, I honed in on chip number 3.

Okay so I did 2, 1, 3.

Error on FEDC

Order now 3, 1, 2

Error now on 7654 - it's either chip 2 or 3.

Order now 3, 2, 1

Error still on 7654 - It's looking like it's chip 3 at fault

Order now 1, 2, 3

Error moved back to FEDC - It's chip 3 alright..

Whenever chip 3 moved, so did the RAM errors. I think we have our faulty RAM chip. Although I am not an expert in this, what  happens is the chemicals from the battery leak and attack the metal legs of the chips. In some cases it's just superficial but sometimes the corrosion can creep up the legs and into the actual chip destroying it completely with no chance of repair.

Rather annoyingly these are of a later vintage than the chips I was playing with here and I don't have any others that might be suitable. I guess I need to get on eBay to look for a replacement chip and see if that recovers it. But in the meantime, I will be doing some more reading up on this type of chip, just to be sure that my rudimentary detective work is correct.

Special thanks to Neil for providing the RAM expansion. :)