Atari Video Music - morphed shapes

[Shockwav3]

Hello there,

I'm currently servicing an Atari Video Music that was busted, so far I'm making quite fast progess (busted opamp replaced, modded for 230V and composite) and got it back to basic operation.

There's one problem that remains and I'm having a bit of trouble to pinpoint it ... thing is whenever I select the horizontal divider of 5 (9th button from the left ... basically making 5 shapes appear in width, problem persists regardless of the vertical setting) the shapes get morphed instead of being perfectly symmetrical (making them look like 'arrowheads' instead of diamonds). A photo of the output is here: http://i.imgur.com/SFNDuqF.jpg
If my judgement is right the falling edges get smeared, while the rising ones are fine. Other settings work as expected.

 

Any idea what specific part might cause this or is this likely a problem with the C1010280 custom chip? I've checked the voltages and test points pointed out in the schematic and those are fine. So are the voltages forwarded by the 'horizontal'-switches.

Sadly I don't have a second machine available to confirm/decline a problem with the custom chip, but a good multimeter + a decent scope are available.

 

Resources that might be helpful:

Patent: http://www.google.com/patents/US4081829

Original schematic (bad quality scan): http://i.imgur.com/7pfBnlP.jpg

redrawn (unconfirmed) schematic: atariage.com/forums/index.php?app=core&module=attach&section=attach&attach_id=413189

Edited May 26, 2017 by Shockwav3

[+-^CrossBow^-]

Way more interested in knowing how you modded it for composite? Because that has been discussed many times before with no one ever really coming forth on how they did it or if they were able to do it. In fact Ben Heckendorn even looked into this in one of his episodes on YT and wasn't able to figure out how to get comp from it.

 

So..yeah...what did you have to do to get that going?

[+Stephen]

I'm not much of a hardware guy. This part of the patent seems relevant to the issue you are having:

 

Comparators 36a and 36b have as their second inputs a common signal on line 37 which is produced by the ramp generator 18 (FIG. 3). Ramp generator 18 referring to FIG. 6 includes an integrated circuit which produces on its VA through VDoutputs four signals which count down (from 1111 to 0000) and then up in a binary fashion. These correspond to a binary digital representation of V-shaped ramp signal for each frame. The most significant bit (MSB) normally has a period of approximately 1/60 of a second (or a multiple thereof) to thus correspond to the vertical sync frequency of a television receiver. The four digital signals are converted to an analog signal by the common technique of summing through appropriately weighted resistors to provide a V-shaped waveform indicated at 41. The values of the summing resistors are indicated and correspond to the digital weighting value of each binary signal of the integrated circuit which is 1, 2, 4, 8. Similarly, the HA through HD outputs provide a V-shaped waveform indicated at 42 with a fundamental waveform corresponding to the horizontal trace interval of approximately 63.84 microseconds. The two waveforms 41 and 42 are summed at 43 and after shaping and amplification produce the final waveform on line 37 which is utilized by the comparators 36a and 36b to compare with the energy waveform of the audio signals to provide channel 1 and channel 2 luminance signals.

[Shockwav3]

Way more interested in knowing how you modded it for composite? Because that has been discussed many times before with no one ever really coming forth on how they did it or if they were able to do it. In fact Ben Heckendorn even looked into this in one of his episodes on YT and wasn't able to figure out how to get comp from it.

 

So..yeah...what did you have to do to get that going?

I found someone else who had modded his AVM via google and got instructions from him after contacting him via his website.

I was asked not to spread the instructions ... so you're on your own

[Keatah]

Well what good is that? It's not like it's a state secret or anything..

[+-^CrossBow^-]

I found someone else who had modded his AVM via google and got instructions from him after contacting him via his website.

I was asked not to spread the instructions ... so you're on your own

 

Really..so this is a trade secret now? Huh.....I see....

 

Maybe I shouldn't share anything I do hardware related either anymore. But then, it wouldn't make much sense to have this community then would it?

[Shockwav3]

It's called loyality. I was asked not to share it and so I'm not out of respect for the one who helped me out. Lemme quote the E-Mail I got back then:

I'm not going to say that its the definitive way to do it, but it works for me. I'd prefer it if you didn't share the modification publicly..

I pointed into the general direction I had to go to find it ... I'm afraid that's the best I can do.

 

Also nice way to hijack this thread But since this thread is already going down the drain at a rapid pace ...

In fact Ben Heckendorn even looked into this in one of his episodes on YT and wasn't able to figure out how to get comp from it.

I'm 99.8% certain Ben either ran out of time or interest with the AVM, ...

[+-^CrossBow^-]

It's called loyality. I was asked not to share it and so I'm not out of respect for the one who helped me out. Lemme quote the E-Mail I got back then:

I pointed into the general direction I had to go to find it ... I'm afraid that's the best I can do.

 

Also nice way to hijack this thread But since this thread is already going down the drain at a rapid pace ...

I'm 99.8% certain Ben either ran out of time or interest with the AVM, ...

Fair enough. I wish you luck in the repairs and hope someone is able to set you in the right direction with this.

[Shockwav3]

Post that was just in the making:

Phew ... I just checked all caps/resistors/diodes related to the relevant parts and as said, testpoints look alright under the scope.

Only potential issues I could find would be the CLK pin being slightly asymmetric with the falling edge being less steep than the raising one ... but just very slightly. Frequency itself is perfectly fine.

Also pins 13&14 have slightly wonky levels when the horizontal divider is set to 5 (otherwise they're nice -5V signals) ... the attached resistors check out fine.

 

Turns out the recommended substitute for the op-amp has a slightly different frequency response ... I've previously put in a MC3401P which showed above issues (the original CA3401G was busted) ... now with an CA3401E popped it in works nicely.