Forums

Recently I've been collecting SMS board variants in order to document them for the ConsoleMods wiki, and I noticed that the original NTSC revision (837-6067) seemed to have fairly dim RGB output which also affected the composite out. This was especially noticeable with my RetroTINK-5X which also had trouble "locking on" to its composite signal.

So tonight I was working on an 837-6067 board and I decided to experiment with replacing the series and pullup resistors in the voltage divider used for RGB input to the video encoder. I noticed that Sega, for some reason, used 3.3k ohm resistors for the series resistors and the same value for the pullups. This is different than the later NTSC revs which have 2.2k ohm for the series resistors and 4.7k ohm for the pullups. I had the appropriate resistors on hand so I swapped them out, and sure enough the RGB output is just as bright as my later consoles! This also had the benefit of improving the composite signal and it now gets along well with my RT5X.

If you have good desoldering tools, this is fairly easy to do. Remove resistors R6 through R11, then replace R6 through R8 with 2.2k ohm resistors and R9 through R11 with 4.7k ohm resistors. Obviously an RGB bypass can fix the RGB issue too, but I think this is a decent option for improving the RGB output from these early revs.

I've attached a couple screenshots from my RetroTINK-5X to show the difference.

EDIT: I'm starting to think this was a mistake on Sega's part. I noticed when looking at the Mark III, they also used 3.3k ohm resistors on the RGB voltage divider circuit. This would make sense for the BA7230LS, which expects lower video levels (0.7 Vpp vs. 1.0 Vpp on the Sony encoders), but I believe Sega may not have been playing close attention and simply reused the same values for the V7040.

Screenshot_2024-05-17_19-16-00.png (423.84 KB)

Before

Screenshot_2024-05-19_23-23-19.png (482.97 KB)

After

Awesome. Nice find and looks good

I have quite a few revisions myself, including all the Techtoy sms3 boards (even the Portugal version which is different from the Brazil version) and all the sms3 "collection" systems

Personally I wish Sega went with all unamplified rgb like the mark3 then I would just mod in ths7374 on them and leave them attenuated internally like the Japan SMS did rather than in the cable like export systems

You’re welcome to post things our Development wiki for preservation purposes when all other websites are gone.

Author	Message
ApolloBoy Joined: 30 Jan 2007 Posts: 330 Location: San Jose, CA	Fixing dim RGB on the 837-6067 board Posted: Mon May 20, 2024 6:58 am Last edited by ApolloBoy on Thu May 30, 2024 6:41 pm; edited 2 times in total
	Recently I've been collecting SMS board variants in order to document them for the ConsoleMods wiki, and I noticed that the original NTSC revision (837-6067) seemed to have fairly dim RGB output which also affected the composite out. This was especially noticeable with my RetroTINK-5X which also had trouble "locking on" to its composite signal. So tonight I was working on an 837-6067 board and I decided to experiment with replacing the series and pullup resistors in the voltage divider used for RGB input to the video encoder. I noticed that Sega, for some reason, used 3.3k ohm resistors for the series resistors and the same value for the pullups. This is different than the later NTSC revs which have 2.2k ohm for the series resistors and 4.7k ohm for the pullups. I had the appropriate resistors on hand so I swapped them out, and sure enough the RGB output is just as bright as my later consoles! This also had the benefit of improving the composite signal and it now gets along well with my RT5X. If you have good desoldering tools, this is fairly easy to do. Remove resistors R6 through R11, then replace R6 through R8 with 2.2k ohm resistors and R9 through R11 with 4.7k ohm resistors. Obviously an RGB bypass can fix the RGB issue too, but I think this is a decent option for improving the RGB output from these early revs. I've attached a couple screenshots from my RetroTINK-5X to show the difference. EDIT: I'm starting to think this was a mistake on Sega's part. I noticed when looking at the Mark III, they also used 3.3k ohm resistors on the RGB voltage divider circuit. This would make sense for the BA7230LS, which expects lower video levels (0.7 Vpp vs. 1.0 Vpp on the Sony encoders), but I believe Sega may not have been playing close attention and simply reused the same values for the V7040. Screenshot_2024-05-17_19-16-00.png (423.84 KB) Before Screenshot_2024-05-19_23-23-19.png (482.97 KB) After

Lifeisasimulation Joined: 30 Sep 2023 Posts: 20	Posted: Mon May 20, 2024 7:50 am
Lifeisasimulation Joined: 30 Sep 2023 Posts: 20	Awesome. Nice find and looks good I have quite a few revisions myself, including all the Techtoy sms3 boards (even the Portugal version which is different from the Brazil version) and all the sms3 "collection" systems Personally I wish Sega went with all unamplified rgb like the mark3 then I would just mod in ths7374 on them and leave them attenuated internally like the Japan SMS did rather than in the cable like export systems

Maxim Site Admin Joined: 19 Oct 1999 Posts: 14770 Location: London	Posted: Mon May 20, 2024 8:19 am
	You’re welcome to post things our Development wiki for preservation purposes when all other websites are gone.

Forums

View topic - Fixing dim RGB on the 837-6067 board