• Good day, Guest!
    Flowers in Chania

    OG is raising money to form a non profit community interest company enabling us to preserve more history than ever before!

    Unfortunately, like most things in life this is going to cost money and we're asking for some help from the community. All users will be given donator status, along with early updates and depending on how much you help us with we'll try to chuck in some free merch!

    This will cost us roughly $850 for two years including business address/phone number etc which is needed to register a business bank account within the UK.

    You can help us by making a custom pledge, or by making a general pledge as it all helps us greatly. We're happy for a one time donators via Ko-Fi, Patreon asks monthly so please be aware. To prevent charge backs we're unable to accept Paypal at this time. Although once live this will be sorted too!

    One time donations can be sent via our Ko-Fi,



    Or for monthly please visit our Patreon, Ko-Fi is preferred as they charge us less fee's. Become a Patron of Obscure Gamers!
  • The Obscure Gamers 2020 Light and Dark themes are now live! Please report any bugs and make any suggestions you may have to a member of staff.

Issues with RGB bypass board on a PAL 1-chip SNES

MangledLeg

Registered
Registered
Joined
Feb 11, 2020
Messages
4
Reaction score
1
AG User Name
MangledLeg
Absolutely boggled with this one.

I have a 1-chip PAL SNES that I was previously having issues with. Taking the RGB output into my old CYP-based transcoder had some sync issues and since upgrading to an Audio Authority RGB to component transcoder I’d get a rolling picture. All my other consoles running in RGB look amazing however, even my Master System and 32X.

I figured the issue was around sync and saw the results of the SNES RGB bypass boards. I figured I’d knock out two things at one - fix the sync and make sure the RGB looks as good as it could be.

I grabbed a board a while ago from oldskool consoles which was based on borti’s design. I bought the one set to TTL sync (first mistake maybe?) and performed the prep work as described in borti’s documentation:
  • R15, R16 and R17 removed
  • R28 and D1 removed (frees up pin 3 on the AV connector to use the CSync from the board for PAL systems)
  • C11 replaced with a 470nF cap (value TBC - currently not able to check my parts!) to fix the ghosting issues
  • Wires soldered to the nearby spots close to R15-16, and R28
  • While not in Borti’s documentation, I upgraded the SNES’ main regulator to a 1.5A one and I also desoldered and removed the RF modulator. I had previously put in a decoupling cap on the regulator to fix the centre line issue and left that in there (one of the solid Panasonic ones recommended years back on Assembler to resolve that issue in another SNES console I used to have)
I’ve triple-checked all my soldering and everything’s coming up clean in terms of continuity checks.

The symptoms are:
  • No blue signal is coming through
  • Sync seems stuffed - has trouble locking to the signal I think, you can see distortions in the screens attached. Black screens create a no signal error on my CRT
  • Tried at both 50hz and 60hz
  • Took apart the RGB cable and moved sync to pin 3 (stupid me at first didn’t change it - it was wired to composite sync!), and also moved to luma. Luma seemd a bit better than CSYNC off the bypass board in terms of not rolling the picture too much when the resolution seemed to change, but in practice once it locked onto the signal the screenshots are virtually the same as what I’ve attached
Photos as follows (apologies for the quality - taken on an iPhone and I’m writing this on an iPad...)

DBE593BE-23AE-478E-BE02-B3B5EEEC4B0F.jpeg


86E7AFA6-4858-4808-BED8-6C1233231159.jpeg


F3EEBB9C-3BC1-45E2-8A6D-29C5F27D9A1D.jpeg


I’ve tested with my NTSC N64 RGB cable (same behaviour - no blue, but blue is coming through on the N64) and ran continuity checks along the scart connector back to check the blue signal, disconnected sync on the SCART cable to make sure I wasn’t getting confused with the pinouts (can confirm no sync resulted in no picture). I also cleaned up my soldering on the PCB compared to the above (which, despite the blurry image, had no continuity errors between pads in the initial fitting).

Thoughts from here -
  • Any ideas if there’s an easy fix?
  • Would the fact it’s TTL sync explain some of the issues? All my other consoles are not TTL, but the fact that it’s still bust when grabbing sync from luma has me double guessing this
  • I didn’t do anything around the decoupling caps - is this worth investigating? I’m unsure if the bypass PCB has them fitted (I don’t think it does...)
  • Borti’s documentation suggests I can grab the RGBS signals from the S-CPUN IC - is this worth a shot?
  • Should I just break out my desoldering station and remove the current bypass PCB, grab a board from Videogame Perfection and have another go at it? He has a PAL-specific option so perhaps this will help
  • Have I killed the SNES??
Thanks in advance if anyone can help!!
 

Attachments

bond.san

Registered
Registered
Joined
Jun 30, 2019
Messages
21
Reaction score
4
AG User Name
bond.san
AG Join Date
06/25/14
Can you elaborate on your equipment setup please.

SNES -> NTSC RGB scart cable -> RGB too component transcoder -> [component in] CRT TV (Hitachi)

Is this your transcoder?

As i understand NTSC rgb scart cables use capacitors inline on the RGB lines
and a PAL rgb scart cable uses resistors inline.

your pictures are great.

Links:
borti4938's RGB bypass information
 

bond.san

Registered
Registered
Joined
Jun 30, 2019
Messages
21
Reaction score
4
AG User Name
bond.san
AG Join Date
06/25/14
Wild speculation here; ( my electronics kung-fu is beginner level )

personally I would replace the vid cable caps for resistors in line with this diagram,
( as this might reduce the (too hot) 5v peak to peak TTL sync too 0.7v ? peak to peak )

the datasheet for the THS*** amp IC should give the output voltages.
snes-scart-pal.png

I don't think you have borked any internal components.

Your PCB looks like an older version not available on OSH park?
What's the jumper on it do?

I have not installed this RGB bypass board but to further clarify your mod;
was it these components that that you removed?
RGB bypass component removal.png


Hi-res PCB scans from "STRIP_CLUB_PCB_SCANS"
 
Last edited:

MangledLeg

Registered
Registered
Joined
Feb 11, 2020
Messages
4
Reaction score
1
AG User Name
MangledLeg
Hi Bond.san!

Alrighty, answers:

Video chain is correct - there’s also a Kramer matrix in the middle for video capture, but even taken out of the mix (i.e. component video from transcoder to TV) the issues remain. That’s the transcoder as well.

Thanks for reminding me RE cable as I forgot to mention - resistors on the RGB lines previously on the SCART plug were removed and I don’t have the 220uF caps on there (as per standard NTSC spec, so I know this is an issue). Will look at adding these to the cable and putting an inline resistor on the sync to bring the levels down on TTL sync to see what difference it makes.

For the SNES PCB, you’re correct - those are the components I’ve removed. The only other change not shown on the PCB is the replacement for C11.

In terms of the age of the PCB, I can’t say but I have noticed oldskool consoles aren’t selling them anymore. The more I think about it the more I’m leaning towards a new PCB from Videogame Perfection, but if there is an easy fix I’m keen to try that first to save time and a few $$ :)
 

bond.san

Registered
Registered
Joined
Jun 30, 2019
Messages
21
Reaction score
4
AG User Name
bond.san
AG Join Date
06/25/14
Im also looking to restore csync on my PAL 1CHIP; have been reading assembler-games.com threads.

This page gives a detailed electrical background about the on-board video encoder

will need to re-re-read the posts as to deduct(!) how to do this or order a bypass board! :)

SNES RGB Bypass Board Vers. 4.1a

@Damien - where the forum attachments from assembler archived? I have tried copy and modifying the attachment address with no luck.
 

bond.san

Registered
Registered
Joined
Jun 30, 2019
Messages
21
Reaction score
4
AG User Name
bond.san
AG Join Date
06/25/14
Dang-it ! giving you conflicting information regarding the scart cable !! :(

The bypass puts CSYNC too pin 3 on the AV port.

As you mentioned oldskool consoles and in turn VGP's pcb are designed from Borti's design;

Quote;
"Key points:
1. This amp has been configured for use with an NTSC spec RGB SCART cable. Please use a SCART cable designed for an NTSC Super Nintendo/Super Famicom console, even if you install the mod in a PAL machine."

Quote - borti;
"Make sure that you don't have (electrical) contact between SNES mainboard and your modding board. If you are unsure, simply use isolating tape."
- don't think this is an issue as the silk screen tends to provide insulation

where are you getting csync from? -> rgb bypass pcb -> SNES AV port pin 3 = TTL (5v p-p) or buffered (0.7v p-p) ? -> transcoder input requires TTL or buffered sync ?
5578_F3EEBB9C-3BC1-45E2-8A6D-29C5F27D9A1D.jpeg

srgb csync out top.png

rgb csync move wire.png

So can I suggest moving the current csync wire to this via! :)
 

Attachments

Damien

Administrator
Staff member
Administrator
Moderator
PSH Staff
Refugee
Registered
Joined
Jul 26, 2018
Messages
279
Reaction score
815
AG User Name
Demon
AG Join Date
Aug 11, 2008
@Damien - where the forum attachments from assembler archived? I have tried copy and modifying the attachment address with no luck.
Hey,

We couldn't get them to function correct as XenForo stripped the filename type form the filename, I have a mostly complete attachment archive somewhere just had no time to fix due to ongoing family issues.
 

MangledLeg

Registered
Registered
Joined
Feb 11, 2020
Messages
4
Reaction score
1
AG User Name
MangledLeg
Edit - post will be updated, hadn’t refreshed the thread and there’s more discussion!
 

MangledLeg

Registered
Registered
Joined
Feb 11, 2020
Messages
4
Reaction score
1
AG User Name
MangledLeg
@bond.san, thanks for the csync tip - was having trouble chasing down a suitable location on my PAL 1chip!

I’m thinking of taking out the current bypass board and starting again with the newer board if the alternate csync spot yields no difference. I’ll update the post pending the outcome :)
 
Top