19K6101 monitor or Tempest board issue?

lmorchard

New member
Joined
Feb 28, 2015
Messages
15
Reaction score
5
I've got a Tempest machine that seems to have a collapse issue on the X-axis. (I think that's technically the vertical, but of course the monitor is turned sideways.)

However, the collapse isn't down to absolutely a flat line. It's about a fingernail-width wide bar. When I play with the X axis pots on the PCB, the X CTR is the only one that does anything: it moves that thin bar of the game screen back & forth on the monitor. So, that makes me think the monitor might be okay and that this is actually a game board issue.

I've recapped the 6100, installed an LV2000, replaced & retested all the frame transistors, reflowed solder on every point on the deflection board.

I found this "Tempest Project" page that mentions replacing MC1495 ICs on the game PCB to address a collapse. Could that be what I'm looking at?
 
It certainly could be a game board issue. yes, the MC1495s could be bad. Did you look at the pots on the "Y" portion of the game board? Any broken legs? "Y" portion on game board emphasis.
 
It certainly could be a game board issue. yes, the MC1495s could be bad. Did you look at the pots on the "Y" portion of the game board? Any broken legs? "Y" portion on game board emphasis.

The pots all looked fine to me, no broken legs. Though, I had a replacement pot kit from Bob Roberts on hand. So, I swapped all of those in for the heck of it. Didn't seem to make anything better or worse.
 
If it was your pots, you'd have no signal. Distorted pincusioning like you're seeing is the 1495's.

The catch is it can be tricky to know which one. The thing to be aware of is that they are error amplifiers, so they cross-feed each other. The Y signal is boosted based on the position of the X signal (and vice versa), which is generated by another 1495 *in the opposite axis chain*. So the problem could be the one in the X axis chain, or the Y axis chain.

See the schematics for more.
 
If it was your pots, you'd have no signal. Distorted pincusioning like you're seeing is the 1495's.

The catch is it can be tricky to know which one. The thing to be aware of is that they are error amplifiers, so they cross-feed each other. The Y signal is boosted based on the position of the X signal (and vice versa), which is generated by another 1495 *in the opposite axis chain*. So the problem could be the one in the X axis chain, or the Y axis chain.

See the schematics for more.

If it could be any of them, I'm kind of thinking it might be worthwhile to just socket & replace all of them. Looks like they've been all replaced before, but the soldering is a little ugly. I tried reflowing the solder on the pins, but that didn't seem help. Hoping there's not some other buried problem that prompted the original replacement - no idea how long ago that might have been done.
 
Well, a bit of unsolicited advice: If they're already socketed, you might want to try just replacing them (or at least the ones that are socketed), and if that fixes it, leave it.

Atari boards are relatively challenging to solder, for a couple of reasons. First, the traces can be fragile, and lift fairly easily. And if someone has already done some work, this increases the chances even further, as the heat weakens the bond between the trace and the PCB. I repair a lot of boards with previous work done, and a lot of it isn't good, and they can be very tricky. The quality also varies from board to board, as it looks like Atari used multiple board houses, and some boards are more fragile than others.

Also, Atari didn't use solder mask on the top of the boards, which means the traces are exposed, and don't have a protective covering on them, which allows solder to stick to them very easily as a result. This makes it very easy to create shorts and bridges underneath the sockets, which can be hard to see. Just a nearly invisible wisp of solder can cause a short, and cause headaches.

This is especially true for the ROMs and RAMs, where the traces running between each pin, and there is very little clearance. Also, in the analog section, where there are some very thin traces running under the chips.

If you are confident in your skills and have the right tools, then by all means go ahead. But if you are new to PCB soldering and removing chips, etc, then you might want to hold off, and just leave the existing sockets there, as it's possible to really make a mess. I have a Hakko 808 desoldering gun, which makes things a lot easier, as safely removing the chips with a manual sucker is definitely a challenge.
 
Well, a bit of unsolicited advice: If they're already socketed, you might want to try just replacing them (or at least the ones that are socketed), and if that fixes it, leave it.

Oh, I'm entirely soliciting any & all advice, please & thank you! :D But, the chips weren't socketed whenever they were worked on last. Just looks like the old ones were desoldered & yanked out and new ones dropped in & resoldered. There's kind of a mess of (I think) flux and uneven soldering around the pins that doesn't look anything like the rest of the board

I am rather new to this stuff, but I think I might actually have access to a Hakko 808 at the local makerspace here (i3Detroit). I've just been using a cheap little manual sucker in my basement. I might try practicing a bit with some junk electronics first, though.
 
Last edited:
Also, assuming I do try to replace these MC1495 chips - anyone know a good source for them nowadays? Found some on AliExpress, but wondering if someone in the US has a stockpile of them.
 
Also, assuming I do try to replace these MC1495 chips - anyone know a good source for them nowadays? Found some on AliExpress, but wondering if someone in the US has a stockpile of them.

be sure to check i3's stockpile too! they have TONS of ic's (mostly TTL).
 
Back
Top Bottom