Another VCS with no colour after composite mod

[Starglider01]

Ok thanks. I don't think the Jack is the issue, but I'll try that as a last resort. I'm just waiting a new PSU which is currently the #1 suspect. [emoji1696]

[Starglider01]

After replacing the ribbon cable that connects the switch unit on this heavy sixer with a new Dupont assembly, for some reason the picture quality improved fairly significantly. My only question now is whether I should be expecting better composite video quality on a 1977 revision 2 heavy sixer than this?: https://photos.app.goo.gl/7MU1NqU7JqrEuEW46

[ChildOfCv]

That blue Matrix streaking shouldn't be happening.

 

If you put the meter in AC mode and measure the +5V at various points, what do you read?

[Starglider01]

That blue Matrix streaking shouldn't be happening.
 
If you put the meter in AC mode and measure the +5V at various points, what do you read?

Solid 5V at the regulator and pins of each chip. Any other points specifically to test?

[ChildOfCv]

6 minutes ago, Starglider01 said:

10 minutes ago, ChildOfCv said:

That blue Matrix streaking shouldn't be happening.
 
If you put the meter in AC mode and measure the +5V at various points, what do you read?

Solid 5V at the regulator and pins of each chip. Any other points specifically to test?

In AC mode?  No way that should happen.  Ideally it should be 0V everywhere.

[Starglider01]

Sorry didn't notice you said AC mode.

 

What does that test for specifically related to the picture noise? AC interference?

 

 

[Starglider01]

Here are the AC readings at various +5V DC points:

• Voltage regulator: 0.005V AC
• TIA: 0.005V AC
• RIOT: 0.001V AC
• CPU: 0.004V AC
• Inductor: 0.005V AC

What does this tell us and how can I use it to help the noise? Any other places to try? Thanks!

[+DrVenkman]

I think the suggestion to try AC mode is to measure any noise in the signal. I really don't think noise that low should affect the video that badly, but I've never used one of those TFW8B mod boards, either. My Atari consoles are modded with UAV boards, either from @Bryan when he first designed and sold them several years ago, or more recently from @MacRorie's shop, The Brewing Academy, where he produces them under license with permission from Bryan. 

 

Anyway, you definitely have some noise in that signal but honestly, I would not be surprised if a lot of that is coming from the switching power supply. Modern PSU's are surely more energy efficient but they are noisy as crap. I played around a lot last winter/spring building 555 timer circuits and using my oscilloscope to measure the capacitor charge/discharge cycles and how they relate to the timer duty cycle - long story short, most of the cheap DC power supplies I could find to power my little breadboard produce TERRIBLE noise that is plainly visible on the scope. The best one I could find is an Elegoo-branded one that came with a big Arduino kit I bought to play around a few years ago. That little thing is generic as it comes, but it produced an absolutely rock-solid 9VDC with no ripple or noise when connected to my timer circuits. Unfortunately, it has a fixed plug rather than interchangeable ones, and it won't fit a 2600. 

[Starglider01]

Thanks. This has been my suspicion as well based on advice from Luke/Console5, so I have a new PSU en route from him as we type. ?

[+DrVenkman]

2 hours ago, Starglider01 said:

Thanks. This has been my suspicion as well based on advice from Luke/Console5, so I have a new PSU en route from him as we type. ?

BTW - watched your first video on this project today. I enjoyed it quite a bit, but whatever dark magic you used to make yourself and Lady Fractic look like 10 year old kids freaked out my eyeballs.  

[Starglider01]

BTW - watched your first video on this project today. I enjoyed it quite a bit, but whatever dark magic you used to make yourself and Lady Fractic look like 10 year old kids freaked out my eyeballs. [emoji4] 

Haha glad you liked it. That was a Snapchat filter I found, along with some bluescreen fun. [emoji64]

[ChildOfCv]

12 hours ago, Starglider01 said:

Here are the AC readings at various +5V DC points:

• Voltage regulator: 0.005V AC
• TIA: 0.005V AC
• RIOT: 0.001V AC
• CPU: 0.004V AC
• Inductor: 0.005V AC

What does this tell us and how can I use it to help the noise? Any other places to try? Thanks!

Those are good low values.  If they were too high, as Dr. Venkman says, they would contribute greatly to noise.

 

But unless your meter is filtering out high frequency AC when it takes its readings, that also indicates that the power adaptor may be okay.  Of course it might be doing that.  Fingers crossed.

 

Ya know, there is one other possibility.  I think it had been suggested to try on a different TV back when you were having the color problems.  Now that the color problems are solved, does the non-digital TV look any better?  I'm curious because that weird blocking looks similar to MPEG-based issues.  Perhaps the TV is creating its own problems, then trying to filter them out?

 

[ChildOfCv]

8 hours ago, DrVenkman said:

BTW - watched your first video on this project today. I enjoyed it quite a bit, but whatever dark magic you used to make yourself and Lady Fractic look like 10 year old kids freaked out my eyeballs.  

Wow, I didn't know he was THAT Starglider!

[Starglider01]

 
Ya know, there is one other possibility.  I think it had been suggested to try on a different TV back when you were having the color problems.  Now that the color problems are solved, does the non-digital TV look any better?  I'm curious because that weird blocking looks similar to MPEG-based issues.  Perhaps the TV is creating its own problems, then trying to filter them out?
 

I have tried 3 TVs... [emoji45]

[ChildOfCv]

A few more spitballs:

 

1) What is the voltage of the composite video where you connected the wire?  If it's too high or too low, it could be that you're seeing the result of clipping.

 

2) Do you have a pick or something you could use to remove a pin from that ribbon cable replacement wire?  (this is definitely a long shot, but...)  there is a phenomenon called reflected energy, which happens when electricity goes into an unterminated or poorly terminated wire, such as what used to go to the RF modulator.  You get a "reflection" of that energy back at the branch point.  You could pull out pin 12 to see if that makes a difference.

 

3) Do you have an oscilloscope?  If so, you can look at the composite signal before and after the adaptor to make sure it looks similar.  Output should be a copy of input.  The adaptor is just an amplifier.  Also, wherever its lowest point is, you should see about 1V difference to its highest point on a white screen.

 

BTW, whoever suggested removing R215 may have thought you were doing an S-video mod.  In that mod, color is separate from luminance, so you'd remove the resistor that mixes them together.

Edited October 4, 2020 by ChildOfCv

[Starglider01]

1. 2.47V DC / 0.138V AC

 

2. I'd rather not mess with that cable right now but it's an interesting thought

 

3. Sadly no oscilloscope here, only a logic probe

 

Thanks!

[ChildOfCv]

That's a good middle of the road voltage.  So no problems there either.

 

Well, I suppose if the new power supply doesn't fix it, there's always the possibility of testing the jack.

 

Oh, it may be worthwhile to untwist the Dupont cable too.

[Starglider01]

Thanks. Jack is brand new and Dupont untwist didn't help.

 

Fingers crossed for the PSU!

[Starglider01]

On 10/1/2020 at 10:41 AM, -^CrossBow^- said:

Another suggestion I was going to toss out was to wire in an actual RCA jack temporarily off the composite output and tie it to ground along the mainboad somewhere to see if that would rule out the AV jack currently being used.

Okay I have tried this. Short story: Same result.

 

I went one step further and bypassed the composite mod completely, touching my makeshift cable composite directly to R216, and GND to R237 or the TIA GND. Although the colours were weird, I got a picture, but the same artifacts/interference. So I think we can safely say it's not the mod and is something with the VCS... (although I didn't desolder the cables going to the mod as the VCS pads are getting fragile, so it was technically still connected to the VCS albeit not to the TV.)

 

I've now also tried 3 PSUs, but no improvement.

Edited October 6, 2020 by Starglider01

[+-^CrossBow^-]

Okay... so at this point I'd have to lean on the TIA being the issue unless that was tried already and I missed that further up?

 

[Starglider01]

33 minutes ago, -^CrossBow^- said:

Okay... so at this point I'd have to lean on the TIA being the issue unless that was tried already and I missed that further up?

 

Yup TIA was swapped... I've actually tried 2 TIAs.

 

On some level I'm prepared to say this is 44 years old and entitled not to be perfect... But if you have any other ideas, I may never give up. Feel free!

[+-^CrossBow^-]

What happens to the picture when you do these two things:

 

1. Does the picture change when you move the controller around itself while you are holding it? Seems the picture flickered a bit when you set the controller down on the counter next to the VCS just after the quick play of DK?

 

2. What happens if you start to wiggle the plug and wires that actually plug into the small mod board itself?

 

[Starglider01]

1. All recent tests have been without a controller. With that DK test, moving anything would cause changes. That has stabilized a bit since I've been troubleshooting. But even now, just moving the mod's wires, or the VCS itself, causes the interference pattern to change/improve/worsen.

 

2. As above.

 

Thanks.

[+-^CrossBow^-]

And you have ruled out the wire harness having loose wire connections within it then? I know you said further up that you soldered a wire for the composite from the pin header on the mod board to the VCS, but what about the other wires?

 

 

[Starglider01]

I haven't specifically desoldered/replaced those as I didn't want to damage any pads on the VCS, but I believe that the bypassing test I did ruled that out. However I will probably give it a go soon just to be sure. Pray for my pads!