Performa 6200 and the PSU of doom

[rubber_jonnie]

As some of you may have seen, the PSU of doom from my Performa 6200 decided to go full nuclear, and I've tried, but now deemed the PSU as unrepairable:

PXL_20221227_132523533.jpg

So I removed the wiring harness and did a bit of a janky hook up to my old bench PSU, a 12v supply hooked up to a PicoPSU:

PXL_20230113_142808221.jpg

The idea being that I could see if the machine was still alive and worth fixing after the 'nuclear event'. And it worked:

PXL_20230113_142801054.jpg

So I at least now know the machine works. Sure, there is some noise on screen, but I think that is likely due to my current lashed up PSU.

Since the PicoPSU and 12v supply are effectively spare right now, and also small enough to fit inside the old PSU case, I'm going to try this: https://68kmla.org/bb/index.php?threads ... 200.35759/

Thanks to the folks who pointed this out to me.

I have all the parts so will give it a bash as soon as I can.

[stephen_usher]

You will have to recap the motherboard unfortunately. The caps Apple used were even more prone to leakage than the ones Commodore used in the A1200!

[rubber_jonnie]

You will have to recap the motherboard unfortunately. The caps Apple used were even more prone to leakage than the ones Commodore used in the A1200!

I have already looked at the mainboard and the caps on it look OK with no signs of leakage (Unlike my Classic). I'll review once I have the PSU sorted as the display was fine before the PSU went foom.

[stephen_usher]

They generally look OK until the first power on and then they squirt juice over the PCB.

[rubber_jonnie]

They generally look OK until the first power on and then they squirt juice over the PCB.

I have some of the caps, just not enough, so have ordered what I need plus some spares. The caps have no signs of leakage, plus the mainboard is simple to pull out on a tray without disassembling the whole machine, so they can wait for now whilst I sort out the PSU.

It's one of those machine that 'got collected' quite a while back and that I pull out and play with once in a while, it'll be nice to be able to put it away when all this is done knowing that next time it's out, the PSU won't go boom!

[rubber_jonnie]

I thought I'd make a start on this today, and right away got lucky, finding an ATX socket I'd recovered at some point:

PXL_20230115_142218407.jpg

So I got busy with the soldering iron and knocked up a harness as per the instructions found at the link I'd posted earlier.

Currently it's just lashed up for testing purposes:

PXL_20230115_161425891.jpg

The results have been a bit mixed though, while it will power on and off as expected via the keyboard, sometimes it just doesn't and you have to leave it a moment unplugged before it'll play nicely again. Shutdown from the menu also works just fine and seems to all the time.

The how-to guide has a small contradiction though. The poster states to use a 74HC04 at the beginning. Further down, the wiring guide (That I used) states to use a 74LS04, so I'm wondering if the difference between the HC and LS might be enough to make it erratic.

I'll get an HC on order, see if it helps, but I feel that it's progress as the fact it turns on sometimes suggests that the wiring job is at least fine (I did check several times anyway), and it's just wonky signalling.

I also resolved the fuzzy display as well, it was the ground between the 12V PSU and the chassis of the machine. Once grounded the screen was nice and sharp again.

[stephen_usher]

I guess that the HC version will have a lower "high" voltage level. Probably marginal then. Need any pull-ups? ;-)

[rubber_jonnie]

I guess that the HC version will have a lower "high" voltage level. Probably marginal then. Need any pull-ups? ;-)

There's a pull up and a pull down in there as it goes, I'll go back and check the voltages there. It only uses 1 of the nand gates so I should double check the input and output when it works vs when it doesn't.

[rubber_jonnie]

So I accidentally ordered a 74HCT04, which it seems isn't quite the same as a 74HC04 :dizzy:

I did realise my mistake immediately after ordering the HCTs, so have an HC on the way too, but it was the HCT that arrived today. Bah Humbug.

Anyway, I swapped the 74LS04 for the 74HCT04 and it is much better at starting. It does occasionally put up a fight, however I'm really hoping the HC will do the job properly.

So next I really need to figure out assembling it all into the old PSU case, and by the time the HC arrives I should be ready to reassemble everything.

[stephen_usher]

The HCT should be the better choice anyway. The 'T' stands for "TTL Level", so it's more compatible with TTL chips. The HC ("High speed CMOS") use the lower CMOS transition voltage levels and are technically incompatible with TTL chips, drawing more current. (I can't remember which chip is stressed and will fail sooner.)