Atari Tempest - Playing blind - Need help

[multisystem]

Hi Everyone,

Richard Here working with Alex in the Arcade Archive Museum (UK).

We are working our way around all the machines and commissioning as many as possible for the launch, I hope some of you will be attending.

Anyway, Atari Tempest is proving to be most tricky and I'm just about stuck as to what to try next. Any Tempest repair experts out there?

Maintenance so far...

The wiring loom and most of the connections had quite poor contact, all cleaned up and now all the controls check out in the self test menu.

We originally had a good looking game image, all well aligned, set to free play we could start a game and play for a few minutes. Then Game play becomes very erratic a few minutes after first power-up, when things had warmed up the vector graphics often corrupted and some of the sound disappeared or became corrupt, usually ending in the screen blanking out.

I checked out both Pokey's and cleaned up all the socketed parts, Pokey's were fine and the cable connection between the math board and the game was very oxidised, good clean and everything seemed a lot better.

Edge connectors clean up - done

Checked 5v on the PSU and main game PCB - reading 5.05v - sense circuit if also working fine and the sense resistors are not burning out.

But now the game plays blind. All seems to be running with full sound but with no vector monitor image when in game play mode.

The spot killer LED is ON, so that's not good, something is upsetting the game play vector output?

But the monitor seems to be completely fine because it's running ok in test.

When switched to test mode, the spot killer LED goes out, we have a vector image, correctly aligned, and no error codes...

Today -

Most of the board caps have already been changed, I have now also changed the 2.2uF Tant cap as that was reading very low.

I have again cleaned sockets and buzzed out almost all the ROM socket signals just to make sure they are all making connection to the board.

What next?

I'm considering dumping all the ROM's so I can verify if they are okay, but before I do that, does anyone know what else could be making it run blind?

Any and all help and advice appreciated.

I'm off to look at Qbert's missing sound now, but will be back to Tempest if anyone has any suggestions.

Thanks,

Rich.

 

[Fantazia2]

Not an expert but this may be of some help.

https://www.arcade-museum.com/manuals-videogames/T/tempest-trouble.pdf

Also a link to some more help pages/manuals.

http://www.andysarcade.de/tempest.html

And a tip from another forum, it was for wavy graphics but may work.

"The three rows show the positions of the DIP switches at N13, L12, and D/E2. So the one is DIP #7 of N13, which control coin modes. See page 8 of the Tempest Manual (TM-190).

I'd start by doing the video adjustment steps listed on sheet 2, side B of the schematics (DP-190)."Fantazia22022-11-02 14:41:24

 

[ColinD]

Get an oscilloscope on the game board XY outputs (ideally an analogue one in XY mode) and that should help troubleshoot. You need to know whether the game board is working or not first of all and then take it from there. From your photo, it does look like you have a monitor issue though - Not sure you should have that horizontal line and nothing on the top half (Don't leave the monitor switched on like that for long if you can help it).

 

[Equites]

The TEST mode looks fine to me, thats what I would expect to see when the PCB is booted with TEST already on. Therefore, it would seem the monitor is fine and you have issues with vectors in GAME mode.

The number '3' in the lower right tells me you are running rev 1 ROM set on the board, I would expect that number to change via spinner.

The deflection is activating Spot Killer to protect itself.

I can't see any Maths Box errors in the screen shot so a bit of an odd one for sure.

I would definately take Colins advice and hook up an oscilloscope to the XY outputs from the PCB and put the oscilloscope in XY mode to see if you can see any vectors being rendered.

Equites2022-11-02 15:37:36

 

[multisystem]

Thanks,

The switches activate in test mode and the spinner rotates the number sequence.

No error codes above the line, there was a M error originally, but that was the cable between the two boards, not had that since solder reflow of the pins and clean of the board-to-board connectors and cable set.

I'll get a scope on it tomorrow and see if anything is being generated, I guess if it is it may just be a problem in the vector output stages. I was starting to think the Vector ROM's 111 and 112 may have gone bad.

The TEST mode looks fine to me, that's what I would expect to see when the PCB is booted with TEST already on. Therefore, it would seem the monitor is fine and you have issues with vectors in GAME mode.

The number '3' in the lower right tells me you are running rev 1 ROM set on the board, I would expect that number to change via spinner.

The deflection is activating Spot Killer to protect itself.

I can't see any Maths Box errors in the screen shot so a bit of an odd one for sure.

I would definately take Colins advice and hook up an oscilloscope to the XY outputs from the PCB and put the oscilloscope in XY mode to see if you can see any vectors being rendered.

 

[multisystem]

Thank you, I'll take a look tomorrow on the scope and see if any XY Vector's are being generated.

Get an oscilloscope on the game board XY outputs (ideally an analogue one in XY mode) and that should help troubleshoot. You need to know whether the game board is working or not first of all and then take it from there. From your photo, it does look like you have a monitor issue though - Not sure you should have that horizontal line and nothing on the top half (Don't leave the monitor switched on like that for long if you can help it).