exxos's Atari, Amiga & retro forum

Badwolf wrote: 08 Apr 2025 16:03 This sounds like you've got a fixed low impedance path then the normal lower resistance in the other direction (this is assuming the lower value is with positive on the 5V and negative on GND, of course!)

The lower value is in reverse. 35R with positive on the 5V, 31R with positive on GND.

Badwolf wrote: 08 Apr 2025 16:03 Less than a watt of extra dissipation? I'm not sure you'd pick it up and it runs a risk.

I think if this involves a small passive somewhere - it will show. If it is one or more IC, it may just be invisible. If I cool the board and do this in a dark room it might help but will have to see.

The fact it looks resistive on both directions though doesn't seem 'normal' even if it seems like an ok dissipation for one of the ICs.

Badwolf wrote: 08 Apr 2025 16:03 I think I heard you mention a seven digit meter. Inject 1.5V in one corner and measure values across the board? All those decoupling caps should actually help you measure local variations stably, I'd have thought.

I will give that a shot first.

... a Atari Mega STE Mainboard has between +5V and GND 100 Ohm. So that is normal without connected the Power Supply to the Mainboard and all chips on the Mainboard.

frank.lukas wrote: 08 Apr 2025 18:04 ... a Atari Mega STE Mainboard has between +5V and GND 100 Ohm. So that is normal without connected the Power Supply to the Mainboard and all chips on the Mainboard.

So 35 ohms is *maybe* still in the right area... what about in reverse though? Maybe I should measure some other Atari boards to see.

[edit]

Forgot to add - I removed all the DRAM chips (16 of them) and the 35R didn't change at all.

- will do the 1.5v injection thing next (I set it up already but has to stop for more important things)
- will try 5v a bit later with IR camera, after cooling the board a bit

I have some stuff still to do just now but will return to this in the evening.

Injected 1.5v at one corner of the board (ACIAs) and measured it at the decoupling caps all over the board.

There is no significant drop anywhere - less than 100mv at the farthest corner only.

What I did learn is that a whole section of the board (components marked 9xx) seems to be missing from the schematics in the TT service manual. They're on the layout in that manual, but schematic for section 9 is missing. Decoupling caps in that area seem mostly tied to other rails. Looks like peripherals or video output related.

So I'm thinking of moving on towards applying 5v either with the chips missing or with minimal chips populated. I'm waiting for new socket for the ROMs but I'll decide if to try power tests before touching that...

Next-ish problem is the remaining PLCC sockets.

The black sockets look ok. Some of the brown ones have defects. A few pins are pressed in and a number of them are corroded/tarnished.

I cleaned the pins which helped but it looks like these sockets had silver-plated copper contacts and the silver has tarnished black. I was considering swapping the worst ones but checking with the microscope, they might be ok so long as the pins are all aligned properly.

(can't find an sdcard handy for the microscope so poor mobile photos from the screen only :p )


meanwhile I did mange to clean up the broken edge quite a bit and the intermittent semi-short went away

Did a bit more on this today...

Removed the corroded ROM sockets...


Fitted new turned-pin sockets...


Replaced a PLCC socket with some blackened/grey pins and a couple of pushed-in pins...


Next I'll try to confirm the DRAMs are ok and start putting things back together for tests.

dml wrote: 08 Apr 2025 16:55

Badwolf wrote: 08 Apr 2025 16:03 This sounds like you've got a fixed low impedance path then the normal lower resistance in the other direction (this is assuming the lower value is with positive on the 5V and negative on GND, of course!)

The lower value is in reverse. 35R with positive on the 5V, 31R with positive on GND.

That worries me a bit, although I'm trying to convince myself a chip could conceivably have a MOSFET style component with an effective reverse diode action and some kind of high impedance path to a rail -- would a CMOS 555 timer exhibit this behaviour perhaps? But it makes me feel uneasy still. :P

dml wrote: 09 Apr 2025 12:03 Injected 1.5v at one corner of the board (ACIAs) and measured it at the decoupling caps all over the board.

There is no significant drop anywhere - less than 100mv at the farthest corner only.

That's a fairly significant drop, mind. What if you took it down below 1V? Could you get any extra resolution? Try a non-opposite corner?

So I'm thinking of moving on towards applying 5v either with the chips missing or with minimal chips populated.

You'll probably be OK, but it's still a TT motherboard and I'd probably try everything I could before taking the plunge if I were in your boots! :lol:

FWIW I pulled out my (fully populated and then some) Falcon and I get near-as-damn-it 185 Ohms in both polarities between 5V and GND. 35 feels really low in that context.

BW

BW

Badwolf wrote: 12 Apr 2025 22:14 That worries me a bit, although I'm trying to convince myself a chip could conceivably have a MOSFET style component with an effective reverse diode action and some kind of high impedance path to a rail -- would a CMOS 555 timer exhibit this behaviour perhaps? But it makes me feel uneasy still. :P

I did think it seemed a bit weird.

Badwolf wrote: 12 Apr 2025 22:14 That's a fairly significant drop, mind. What if you took it down below 1V? Could you get any extra resolution? Try a non-opposite corner?

I could try it again at 1V and maybe solder wires to the board this time, use the filter to get a longer reading. Pressing probes onto pins and leads isn't very stable with such small values - they are drifting all the time and the exposed contacts are dull.

Badwolf wrote: 12 Apr 2025 22:14 You'll probably be OK, but it's still a TT motherboard and I'd probably try everything I could before taking the plunge if I were in your boots! :lol:

FWIW I pulled out my (fully populated and then some) Falcon and I get near-as-damn-it 185 Ohms in both polarities between 5V and GND. 35 feels really low in that context.

That's a big difference.

I'll see what else I can find - but I think if damage was likely from 5V it already happened before I got near it. If any chips failed they're probably already dead.

What I'm not sure about yet though is bringing up the 5V rail without the other 3 rails - so I can current-limit it to begin with. Usually other voltages are only needed by floppy drives, audio and peripheral/serial stuff but I'm not sure about the TT. Might have to study the schematic a bit to be sure it won't damage something.

FWIW I'm using this....

I got 3 of these dead for next to nothing and fixed them all. Well, if you ignore the fact I'm not paying to have them calibrated! That would tarnish their freebie status quite a lot.

Actually two 1062s and this 1071. The 1071 is the only one with the ohms option fitted. I think the other two have RMS/AC option or something, I forget...

The TT will be safe to power up with just the 5V rail. There's nothing which requires a bias voltage other than audio amplifiers and they just don't work if they don't have the other voltages.

Start with a lot voltage and gradually raise the voltage monitoring the current draw. The TT is quite thirsty so it should take quite a few amps at 5V.