asteriods watchdogging

[wugly]

I am having problems with my vmem output. it stays high for quite a long time. in excess of at least 2 seconds then I get a single pulse down. this is on chip e4 of address decoding circuitry. pin 14 input looks decent input on pin 13 looks to be pulsing but around 3 1/2 volts. should I pull pin 24 on cpu from socket and check the output or do you think e4 is bad. my output on e8 but labeled l8 on schematic or all looking funky half of them are stuck high with a few being high mostly with brief spikes to 0. pin 12 is the only input that looks bad and that percolates back to a7 with an input of vmem from e4.
pulled mpu and bent it and saw good 4 volts minimum on signal. put it back in and am seeing a lot of pulsing on pin 13 and 14 but a lot of time on a 5 volt signal with only a stray vmem going to 0 volt. a14 from mpu was probably dirty. but I am wondering maybe not having enough vmem pulses. any ideas?

 

[wugly]

pulled e4 and put it into a tester and its good. so something else is causing mpu to not put out the correct combination of signal to make vmem

 

[wugly]

when running the board in test I get what appears to be a decent cross hatch pattern but it has a bright dot in it on the right side about midpoint maybe a little higher. when running in game mode it has a diagonal line running from right to left looks about as far down from center on left as it looks high to the right not sure if its all the way to the left like it is to the right

 

[ArcadeTechGW]

Maybe your brightness is up too high. That sounds like retrace lines.

 

[wugly]

it works good with a known working board. the watchdogging board has these problems as if it is holding on at this spot for extended time and the game is watchdogging in the video section.

 

[wugly]

Guess I am going to have a problem without having hooked up. Going to connect and see if I blow a fuse. Wanted to test in small pieces. First time repairing vector monitor and have heard the stories about them destroying themselves if things are wrong.

 

[andrewb]

Not retrace lines. This is a board issue. When it's retrace lines, they're between all objects, and consistent. If you have stray lines, that's a board issue.

Do you have a logic comparator? I would look at the VG Address Selector section. Those 157's can be something that will cause stray lines. Not a guarantee, but just where I'd be looking based on boards I've seen before. But you can only go so far with a probe and/or scope. At some point you need to see what data is on the lines, as things can be toggling, but not toggling correctly. That's when I break out the comparator and/or signature analyzer.

As for connecting a monitor, you want to test the DC and AC voltages on the XOUT and YOUT first. DC needs to be between -1V and +1V on both, and AC should be 2-4V, roughly. It's the DC that will kill a monitor, as when there are board issues, either axis can get stuck at several volts, and that's what fries the deflection system (as the beam isn't designed to be deflected and held that far for long periods).

B/W monitors are a little more forgiving than color vectors though, when it comes to this. They can tolerate a railed signal for a while (not that you want to do it regardless). But color vectors will nuke in seconds.

 

[wugly]

Not retrace lines. This is a board issue. When it's retrace lines, they're between all objects, and consistent. If you have stray lines, that's a board issue.

Do you have a logic comparator? I would look at the VG Address Selector section. Those 157's can be something that will cause stray lines. Not a guarantee, but just where I'd be looking based on boards I've seen before. But you can only go so far with a probe and/or scope. At some point you need to see what data is on the lines, as things can be toggling, but not toggling correctly. That's when I break out the comparator and/or signature analyzer.

As for connecting a monitor, you want to test the DC and AC voltages on the XOUT and YOUT first. DC needs to be between -1V and +1V on both, and AC should be 2-4V, roughly. It's the DC that will kill a monitor, as when there are board issues, either axis can get stuck at several volts, and that's what fries the deflection system (as the beam isn't designed to be deflected and held that far for long periods).

B/W monitors are a little more forgiving than color vectors though, when it comes to this. They can tolerate a railed signal for a while (not that you want to do it regardless). But color vectors will nuke in seconds.

don't know if you got side tracked by the message from arcade tech but my description was of the video board that is watchdogging because I have a fault in the video side of my board and I stated the images I got with the board running to maybe get a good direction to look for the fault that is on the board. wish I had a signal comparator. I have a signal analyser that I have not had a chance yet to figure out how to use it though. going to take a look with the scope and failing that I might give the analyser a shot at it.

 

[wugly]

still tracking down the problem and had to go to work so haven't progressed as far as I want. to andrewb advice and looked at vector generator memory address selector. had all kinds of nothing for input on k3 and j3. seems someone beat me to it and was running this problem down before. and had already socketed and replaced these chips. Pulled them out of sockets and put them into ic tester and they are good. H3 and f3 at a quick glance appeared good so I will bypass them for now and if need be later take a closer look. k3 and j3 are fed by a counter at j5 which has input problems pin 9 and 10 are at low pin 15 high and pin 1 is pulseing away. when I get back from work I am going to check j6

 

[ArcadeTechGW]

still tracking down the problem and had to go to work so haven't progressed as far as I want. to andrewb advice and looked at vector generator memory address selector. had all kinds of nothing for input on k3 and j3. seems someone beat me to it and was running this problem down before. and had already socketed and replaced these chips. Pulled them out of sockets and put them into ic tester and they are good. H3 and f3 at a quick glance appeared good so I will bypass them for now and if need be later take a closer look. k3 and j3 are fed by a counter at j5 which has input problems pin 9 and 10 are at low pin 15 high and pin 1 is pulseing away. when I get back from work I am going to check j6

So now check to make sure they soldered the sockets in correctly.

Do all the pins connect where they should?

Are the Vias still in the board?

(Do you know what a via is?)

 

[wugly]

may have found the problem. dvy8 is low. need to triple check. removed f7 and tested good so lifted pin 19 and got a distant spaced signal hopefully this is correct. thinking possible chips on this dvy8 might be bad. possible f8 pin2 f9 pin 15 or j6 pin 10. thinking about nipping a pin, thinking j6 pin10 because it is a driver and have had them give problems on there input before. off to work I go.

 

[wugly]

still working on the problem and I might be getting close but am now chasing my tail. in the program counter on chip h5 pin 10 load 7 I see input. grounded output on pin 6 adma out. can lift pin 6 and see signal. now it gets wierd thought it was h3 pulling it down so I socketed it. I I lift pin 9 on h3 I can see signal on pin 6 of h5 I can put pin 9 back and lift pin 10 on h3 and see output on pin 9 h3 and pin 6 output on h5 I figure pin 9 looks like what is on n/p3 rom. I I pull rom n/p3 with all the other pins back in I see output on pin 6 h5. I have swapped the n/p3 with another rom and replaced both the 193 an d157 chips. need some advice here