Asteroids rev 5 PCB bad X and Y output voltages

[andrewb]

Yes, that is a speed mod board.

I'd remove it, at least for now, until you fix the other issues. Just remove and reinsert a 161 into C5, and the board should be back to stock. You want to remove any other potential variables when you're trying to debug.

There are also easier ways to speed mod these, without needing a daughterboard. Advanced Thread Search post bodies here for 'NMI mod' for more info. Though that board probably gives you more options, if you actually want to change the speed.

Also, I would replace that 161 at D7. That's a Signetics chip, and it's in the Vector Timer. It actually might fix your problem. But even if it doesn't, it's still an improvement to the board, because Signetics chips are such common failures. So you're just removing a weak point for the future, regardless.

 

[wugly]

just so we know why it failed do one thing at a time and test. the c5 is quick and easy and less chance of screwing up the board

 

[Dilbert]

I've removed the speed mod and installed a new socket at C5 as a precaution. I substituted the 161 from the daughter card for the time being and no change. I did notice that once in awhile the test screen flashes on the screen. Not sure if another clue or if I'm just overthinking everything. I'll probe around for any stuck pins and go from there. Probing around the vg I can hear the pulsing that matches the PUSH START but I'm not sure I can hone in on it or use that to find the glitch within it. Lots of garbled noise so far but I'll dive into some schematics and look for floating pins

 

[andrewb]

Did you replace the 161 at D7 as well? I'd do that next.

I highly doubt you're going to find any stuck pins. If something were that broken, the VG wouldn't be working at all.

You're going to want to try piggybacking stuff next. Starting with the six Position Counters, and then the four 161's in the Vector Timer. But I'd replace D7 first.

 

[mwiggett]

I've had that same graphic issue and I think it was a bad RIP carryout on one of the 191 position counter that fixed it. I may be mistaken.

 

[Dilbert]

Will do. I've got to order some chips because I can't find any laying around here. That's the only reason I substituted C5 for the 161 on the daughter board. Maybe if I swap the chip at D7 with C5 the problem will change (if those chips are the culprit)?

 

[Dilbert]

No change when swapping the chips. Could it be some sort of RAM or ROM issue since the test screen flashes once in awhile? I'm not sure how the test screen ties into everything

 

[wugly]

a wild ass guess if the test screen flashes every once in awhile might be a bad bidirectional buffer at e3 but it is not high on my list. I would expect a lot more wrong going or being tripped

 

[andrewb]

Usually when you see the test screen flashing at incorrect times, it suggests a problem in the Stack and Program Counter section (but again, this isn't a 100% rule). You could try piggybacking the three 193's in that section.

Get yourself one of each major TTL type on the board. I have one piece of foam with one chip of every type stuck in it, that I use for piggybacking. You can piggyback every chip on the board one at a time in about 10 minutes. Just go up and down each row and make a list, and get one of anything for which there are two or more chips on the board.

 

[Dilbert]

Ok. I have replaced a few more chips and now we're getting somewhere. I have gotten the glitch from being constant to happening less frequently, and it's reduced in severity. I've replaced the 74161's at C4, C5 and D7, which didn't do anything. Replacing the 139 at E4 is where I saw major improvement. When swapping the old/new chips you can tell a significant difference. Still, the score in the top left corner jumps around, as well as some asteroids and even the ship debris when you explode. I tried comparing the inputs/outputs of the 139 with a probe to see if I could tell a difference and maybe trace the problem back, but I can't tell much of a difference. I've piggybacked all of the surrounding chips and no change.

As of now, the 161's at C4 and C5 are warm to the touch, which I'm not sure was the case before. The clock signals on pin 2 of those chips are non-existent, is that correct?

 

[mwiggett]

Should have a clock on pin 2 at all those chips. Check continuity between all three of those chips to fit into.

 

[Dilbert]

Should have a clock on pin 2 at all those chips. Check continuity between all three of those chips to fit into.

My bad, C5 has a clock (that sounds weird on the logic probe), C4 has no clock

 

[mwiggett]

I was looking at the wrong part of the schematic. C5 may have a clock that sounds strange on the logic probe as that one is 3Khz. C4 should have clock on 11,12,13,14

 

[Dilbert]

I was looking at the wrong part of the schematic. C5 may have a clock that sounds strange on the logic probe as that one is 3Khz. C4 should have clock on 11,12,13,14

Ah I see. Clock signals are all there. They look different than each other on the o-scope but I assume that's normal.

 

[mwiggett]

The frequency should match what the schematic say on the scope. But they are probably fine.

 

[Dilbert]

I'm considering replacing every chip in the 4th row at this point lol. I had the two bad RAMs and a bad 139, and the 161 is replaced.

 

[mwiggett]

Honestly, I'd send the board out. You don't want to just change chips for the sake of it. You can cause more issues.

 

[Mattspad]