Atari RGB Light Sixer Repair

[+Nathan Strum]

1 hour ago, ZeroPage Homebrew said:

Thank you SO MUCH to @Nathan Strum for doing an above and beyond job of restoring and improving the RGB Light-Sixer VCS!!

 

We showed it off on the ZeroPage Homebrew stream last night and the picture output was incredible, I've missed it so much!

You're welcome! I'm glad I could help out, and that it got back to you safely. It was nice to see it in action on the show again! Looking forward to seeing it playing many more games for years to come! 

[Retro_Game_Lover96]

I was working on my Atari RGB mod, I tested it with composite, but I got garbled graphics. Maybe it might be the bad chips, or it could be anything wrong.

[+Nathan Strum]

On 7/12/2020 at 5:34 PM, Nathan Strum said:

Unfortunately, my own S-video modded 2600 stopped working (again) today. So the problem that mysteriously fixed itself, wasn't really fixed. Pretty sure it's a capacitor somewhere. I won't have time to deal with it for awhile though.

 

Well, I finally got my 2600 working again. You can read that epic saga over in my blog: 

 

[+-^CrossBow^-]

I know this was done over the summer originally, but not long ago I had a similar saga with the same RGB kit into another member's light sixer. The client purchased the works when they ordered the kit. So they had the SCART connector that you put together, all the different jacks etc. They also purchased the component output kit too. I actually did wire up everything and install AV jacks for EVERYTHING. Meaning RCAs for composite and audio. The audio RCAs are for use with the s-video and component outputs (They do work together but best to only use one output at a time and disconnect the cables for whichever you aren't using). I did use the included s-video and 8-pin connectors along with the breakout board so that meant I did have to install and wire in the included 1/8 headphone audio jack. Anyway, the purpose of this post was to state that the same issues that plagued the RGB setup originally in Nathan's document process were happening with this one as well only it started the issues pretty much as soon as I first connected the RGB up.

 

I had been testing the RGB install over stages using pretty much just the composite and s-video as that is all my small bench PVM will accept. Once I got the component installed, I also tested that using my old CRT in a spare room and that was working. Once I had everything wired up (I also used connectors for easily removing the mainboard without desoldering, I then made the mistake of connecting up the RGB, S-video, composite and both audio from the 1/8 headphone and the RCAs I had spliced off the headphone jack. When I first powered on the 2600 to test through my RGB testing, all I got was a loud hum from the audio and NO picture. I shut it off and tried it again. Same thing. I then took it all down, connected just the s-video and still no picture. 

 

Long story short, the exact same issue on this RGB was present as Nathan had on the one he worked on for ZPH. I found that some component on the RGB board had somehow failed and was creating a dead short across the ground and +5 on the RGB board. Removed the RGB board and put the console back to stock and it worked just fine. OH I also had to remove and replace the dc-dc switching regulator that was provided because that was also burned out the in the process?

 

BTW, I think I know the reason for the DC-DC regulator being used here. There is so much being added component wise in this upgrade that it would cause too heavy a load on the original 2600 PSU since that is ONLY rated for 500mA total current. With a DC-DC regulator installed, the entire setup with the RGB in place only requires between 320 - 350mA of current. So, the DC-DC regulator was provided so that the original 2600 PSU could still be used and not require have to upgrade to a larger output capacitor PSU. The large 220µf filter cap is needed on the DC-DC regulator because without it, the switching noise from the regulator will interfere with the RF output should you decide to still use it.

 

I replaced the Dc-DC regulator that came with the kit (Since it burned out anyway) with one of the Traco branded 2-2450 DC-DC switchers I use on 7800 consoles. I found that I still had to add the 220µf cap off the input side of the regulator to clean up noise on the RF though. Only I installed the cap on the bottom side of the switch board PCB when I did it.

 

I never figured out what went south on the RGB board since it was fine during all initial stages of testing. Since I was using connectors/disconnects for the wire harnesses it wasn't a misplaced wire either. I never found any shorts or anything in my wiring. The replacement RGB board (Yes I purchased a replacement), installed in the exact same manner, I used the exact same wires and wiring locations to solder to and the second RGB board seemed to work just fine with the setup.

 

BTW the +5 is added to the 8-pin RGB out because most SCART devices require that +5 (And in some cases it is a +12v) to know that a signal is present and switch over to that video source. Mainly needed for auto switching SCART boxes. So I did wire up the +5 on the second go around as well to make sure that was present in case the client needed it for something in the future that required it.

 

Near as I can tell, my connecting pretty much everything at once to try and test in sequence, was causing too much of a load on the RGB board itself and internally fried something but I can't imagine what or how honestly. But yes, exact same issue with the first one I installed as was documented here happened with the one I was doing.

 

BTW I did reach out to Tim about my experience with this but I really hesitate to put his answer in this posting and will answer anyone privately about it.