new 65XE and poss video circuit (passive component caused?) glitch

[Beeblebrox]

Just working on another recently acquired 65XE board. Few issues including a video circuit. 

 

FYI the board was bought off it's previous owner who is fairly experience and has removed all 40pin ICs, the OS rom, basic Ic and also MMU, (as well as the cart connector and ECI port). He had removed all the latter chips very cleanly with a desolder gun. 

 

He said he thought the ram chips/os were an issue and was just selling the pcb on cheaply - after removing all the valuable ICs.

 

I had a dead donor board for the latter connectors which I transferred over. I also added precision sockets to all the missing IC locations on the new 65XE PBC.

 

After adding in all the ICs it powered up on Din to S-video but it was clear there were video issues and also Ram issues as it just went into self test memory test.

 

I desolded the Ram chips and installed Lotharek's 64k Sram module. It booted to the ready prompt but still had the video glitching.

 

Got some bad signal/glitching which I am assuming is most likey transistors or a failing/bad pot, or bad connection to maybe Antic or GTIA?:

 

 

I've replaced the Q2 and Q3 transistors so far, and also replaced 4050, Freddie, tried another GTIA chip. Antic chip works when I pop it in another machine.  I've also adjusted the pot. 

 

You can plug in SIDE3, access the loader and run games. 

 

S-video cable is fine on other A8s.

 

I will double check the GTIA and Antic sockets and may replace the pot at some stage but just wondered if anyone has ever come across this issue? Could a bad pot cause this? I've cleaned it with contact cleaner.

 

I could replace more transistors in the video circuit as well I guess. I once had an issue with screen glitching on a 130XE which had a problem with Antic pin 2 as well as the Q3 transistor and I sorted that. (I've checked these two things with this board so I can rule these out I think).

 

Any ideas? 

 

I've also checked sockets for any shorting on pins and nothing. 

 

I wonder if I need to check any other passive components in the video circuit. I have Sam's 130XE field manual to hand. 

 

Thanks for any suggestions. 

 

FYI it's a PAL XE. 

Edited January 22, 2023 by Beeblebrox

[Rybags]

Antic isn't really involved in the actual end product video generation other than telling GTIA when it should do HSync and VSync.

 

Is the luma reasonably consistent with text in those multicoloured lines?

This looks more like the colour is being screwed - the 4050 I don't think is involved there.

Maybe check the 4.43 MHz PAL colourburst crystal?

[Beeblebrox]

21 minutes ago, Rybags said:

Antic isn't really involved in the actual end product video generation other than telling GTIA when it should do HSync and VSync.

 

Is the luma reasonably consistent with text in those multicoloured lines?

This looks more like the colour is being screwed - the 4050 I don't think is involved there.

Maybe check the 4.43 MHz PAL colourburst crystal?

Thanks for the clarification and suggestions. 

I had wondered if checking the crystal and occilator was worth it. I have spares for both also. 

So is it worth replacing the other transistors or looking at any other passive components in the video circuit? 

 

Edit:, I'll check the luna re text later. Think it is fairly consistent. 

 

By the way I have an occillator installed bottom left, Y1. Then the colourburst in y2. (co16801). Doing this from memory as out and about.

Edited January 22, 2023 by Beeblebrox

[Rybags]

I'm no expert on this stuff - the pic initially gave me the thought of bad RF output or TV not tuned but of course you say you are using the monitor port.

 

I guess another thing to try might be contact cleaner on the monitor socket.

But yes, I suspect a component or 2 has gone bad, maybe transistor or cap.  By the looks the luma portion is probably fine, but the colour isn't.

If you could make up a connection where you have luma only going to the monitor you could confirm that theory.

[Beeblebrox]

1 hour ago, Rybags said:

I'm no expert on this stuff - the pic initially gave me the thought of bad RF output or TV not tuned but of course you say you are using the monitor port.

 

I guess another thing to try might be contact cleaner on the monitor socket.

But yes, I suspect a component or 2 has gone bad, maybe transistor or cap.  By the looks the luma portion is probably fine, but the colour isn't.

If you could make up a connection where you have luma only going to the monitor you could confirm that theory.

Good point re the din5 jack itself. Hadn't thought of that. Might check it's solder points to the PCB too. 

 

If all my investigations turn up with nothing and the Din5 is ok I think I can just install ugv, but I'd rather get to the bottom of it. 

 

I have used contact cleaner on the board prior to and after the sockets went on. 

Edited January 22, 2023 by Beeblebrox

[Beeblebrox]

Lil update. 

Din5 Jack fine. 

 

I replaced the oscillator with the smaller crystal version and a it made a big difference but still issues. Tweaked the pot, which I had previously done before but at the time hadn't improved things. This time is found a slight difference and decided to replace it. 

 

Replacing if didn't improve things much but I had a spare so least I rulled another thing out. 

 

Looks like interference issues are still present and chroma issues. 

 

Took drastic measure and decided to transplant my VBXE XE install from another XE onto this board so spent several hours doing all the prep work, installing din13 jack, creating dupont connectors, wiring things up. Ran outta time today so will see if worked. 

 

Just wanna see it stable running solely off the VBXE.  Once that is running I'll revisit the stock video. (Machine will have both din5 stock and din13 rgb). Have done this install before in an XE. 

 

BTW the current output for stock video :

 

 

 

 

 

 

Edited January 23, 2023 by Beeblebrox

[Beeblebrox]

Ok further update. VBXE transplanted successfully onto this 65XE board. VBXE confirmed working but even on it's separate din13 RGB output there is still on screen interference.

It is less pronounced and a major improvement over the Din5 output which is still bad. Remember that the VBXE is providing the crystals onboard for both the stock Din5 video output as well as the Din13 RGB out directly wired to the VBXE board.

 

Just to confirm:

 

I've tried on both brick Atari main power supply as well as USB power - makes no difference - issues still persist irrespective of which power supply used.

 

I've also checked the S-video cable on my 600XL stock din5 video output and no issues with the cable.

 

Something on the pcb is causing this, especially now since the VBXE is providing the system clock/crystals and you get issues even when outputing to the Din13 (although not nearly as bad).

 

I wonder if a general grounding issue is a possible cause for some of this? (Edit: Alongside some issues with Luma and Chroma)

 

FYI this VBXE install which has been transplanted from my other XE and implemented in exactly the same way saw no similar graphical interference on the other XE. (same length of wiring used, etc etc).

So it is peculiar to this 65XE.

 

Incidentally the RGB Din13 cable will be routed underneath the board as before and also the cables will be tidied up at some stage.

 

      

 

Stock output on S-video (but obviously driven by the VBXE board now), - really bad still:

   

 

 

 

  

 

By contrast here is the RGB output via VBXE on the Din13 RGB, ( still issues but a much better output):

    

 

 

     

 

  

Edited January 23, 2023 by Beeblebrox

[woj]

I am no specialist, and really have no idea what it could be, but I probably would (a) take the schematics and check what's on the way from GTIA to the DIN-5, probably concentrating on the transistors, (b) re-cap the board out of reflex and habit  Oh, and also (c) without removing the whole thing, desolder the RF-modulator signals from the board.

Edited January 23, 2023 by woj

[marauder666]

I’d try disconnecting the modulator, it’ll lose composite video, but s-video will still be ok.   See if that clears up interference.