RESOLVED: Tempest - bad output on "Y" & scaling issues..

Sprout

Well-known member
Joined
Jun 15, 2010
Messages
1,398
Reaction score
118
RESOLVED: Tempest - bad output on "Y" & scaling issues..

Just got the monitor issue resolved (see here).

EAROM wasn't working (not saving high scores), so after putting several games on the monitor to verify it was solid, I shut down, and powered back to confirm it was indeed still not saving scores.

As fate would have it, my board has now decided to go down - I'm getting this output, consistently - both on the monitor, and on my scope..

As per this post, it sounds like I can connect my scope leads at the first TL082 after the DAC, and before the MC1495L's to see if I get a good (albeit smaller) image. If that image is good, I know the issues are downstream. If they're not, then I'm guessing I'd suspect an issue with the DAC @ D10?

Would this equate to pins 9 & 11 on the 74LS04 @ J11, or should I go with pin 1 off of A11/D11?

All voltages in the area checked out as well... Going to pick up a bunch of TL082's (NTE858M) tomorrow; the one at D/E13 was already socketed/replaced by me when I was originally repairing the board, so easy enough to do a quick swap on it - others are not socketed however. Want to make sure whether I'm likely dealing with MC1495's/TL082's vs. the DAC, since if it's likely the MC1495's, I'll be putting an order in to get a batch from China which will take a while...

Appreciate clarification/insight! Pretty bummed that I finally get the monitor solid, and the boardset (which had been solid for a good while) decides to die immediately afterwards :(
 
Last edited:
That looks like classic bad DAC.

You can scope the YOUT signal at pin 7 of D11, and use the XOUT at the test point, and display the image on your scope. You're likely going to find there's no change in the output image, meaning the issue is upstream from D11. Test the inputs to the 6012 DAC with a logic probe, and if they are all toggling, replace the DAC. If they are not all toggling, it's a digital problem upstream from the DAC, but I'm guessing that will not be the case.

BTW, you don't want to run that signal into your 6100, as that type of distorted/offset signaling can kill it.
 
Thanks for the insight!

Only had the monitor online for that video, and strictly using scope afterwards.

Pins 2-10 on the DAC all pulsing, so just put in an order for some AM6012's (China, so a week or so). Will update once I get it swapped out; hopefully no other issues!
 
Pins 2-10 on the DAC all pulsing, so just put in an order for some AM6012's (China, so a week or so).

Since you have a scope I would take the time to probe pin 7 of the TL082 at D11. I agree it looks like the DAC (or the LF13201 switch) but I'd rather desolder 8 pins than 20.

Note that you can get the DAC312HPZ-ND by Analog Devices, available from Digi-key. They are $11.39 each but you know they are genuine.
 
Update,

Still waiting on parts from China, but remembered I had a few dead Asteroids Deluxe boards that I've not gotten around to repairing (don't have an AD any more, so no rush). Pulled a 6012 from one of them, and back in business.

Now - last (hopefully) issue. Before the board went down, I was noticing some funny business with the title screen; the "Tempest" logo had parts missing - it would draw most of the sequence, but skip some parts. I thought this may be down to a bad connection somewhere, and was going to work on it after the monitor was solid. It's now gotten to where it's skipping the majority of the animation - just jumping to specific sections. Note - this behavior was occurring with the original 6012, and is persisting with the replacement, fwiw.

Re-seating the interconnect cable on both ends, re-seating the socketed chips on the aux board - no change. Connectors were re-flowed a while back, and still look great.

In digging around, I found this post, with this video. Other than the EAROM consistently passing built-in diag's in my case (and seeming to be working fine now), and not getting the weird bend in "Tempest" towards the end of that sequence, it's describing exactly what I'm seeing - odd behavior with the stars between levels, and the odd scaling with the logo during attract mode. The odd behavior with the stars occurs during game play as well, but in some ways I actually kinda like it more - it makes it feel more 3D in some ways, oddly enough.

Given the OP replaced the DAC08 (D10) and the 74LS273 (D9) to no avail, I'm kinda hesitant to shot-gun either of those. Since the thread was never closed out with any additional troubleshooting/solution, has anybody else come across this particular symptom and/or have suggestions on the likely culprit?

The outputs seem to be active on the LS273, and in checking B9 (74LS32) for the "SCALELD" signal, pins 1,2, & 3 are all active as well. The 100k resistors seem ok (nothing lower than about 96.8k; most in the 97.5k range). Q11 and Q12 checked out fine via diode test on my meter.

Since the inputs DVY0 through DVY7 are also used elsewhere (ie, Analog Vector-Generator Color Outputs), I'm assuming the issues have to be from the LS273 and downstream, stopping at where Y BIP, Y REF, X BIP, and X REF are generated? Would the DISRST (which appears to reset the LS273 via pin 1) be a factor in this behavior (I'm guessing not)?

I remember playing machines in the wild with this behavior back in the day, so surely somebody else has come across this perhaps?
 
You should PM jkoolpe and ask if he ever resolved that board, as it looks like the same issue.

The slight bending in the logo in his vid could just be the monitor partly kicking into shutdown, which some 6100's will do if there isn't enough stuff being drawn on the screen (depending on the monitor board revisions, and how sensitive the spot killer circuit is on any given deflection board). I would consider that a separate monitor behavior, so it's irrelevant here.

I agree that it's scaling-related, but given that the 273 and DAC08 didn't fix it, it may be something in the scaling control further back.

But asking Jon if he ever fixed it will be the best thing to do right now, as it may save you a bunch of time. I'm sure he'll be happy to comment.
 
The outputs seem to be active on the LS273, and in checking B9 (74LS32) for the "SCALELD" signal, pins 1,2, & 3 are all active as well. The 100k resistors seem ok (nothing lower than about 96.8k; most in the 97.5k range). Q11 and Q12 checked out fine via diode test on my meter.

Since the inputs DVY0 through DVY7 are also used elsewhere (ie, Analog Vector-Generator Color Outputs), I'm assuming the issues have to be from the LS273 and downstream, stopping at where Y BIP, Y REF, X BIP, and X REF are generated? Would the DISRST (which appears to reset the LS273 via pin 1) be a factor in this behavior (I'm guessing not)?

I agree that trouble in the DVY bus would cause much more of a mess. I'd check the outputs of K8 and L8 in the Vector Timer Control and maybe piggyback a new LS191 onto L8 to see if it changes the behavior.
 
Ok, just got this resolved...

The other board having this issue (see the referenced post) was resolved by replacing D8 (LS161) - I checked mine, and it was checking out fine. Checked just about everything else in the "Vector Timer" and "Vector Timer Control" areas, but kept thinking about that DAC08. Swapped it, and that did the trick - everything working 100% now!

Thanks for the help folks!
 
Back
Top Bottom