SNES 1CHIP PAL + VGP 4.1B i° SCART NTSC = no blue signal

[archimage]

Hello,

I've installed the VGP 4.1B board, based on Borti's.

Followed every step of each page :

Remove C44, C45 AND C47
remove R15, R16 and R17
Remove R28 and C46
removed D1

closed 3 jumpers J1 J2 and J3

using 470ohm attenuation NTSC packapunch cable

I'm getting no blue signal

Am I missing something ?

There's also this thread covering the same problem but still nothing found out

https://www.obscuregamers.com/threads/i ... snes.1716/

Each page VGP Borti's GITHUB point to Retrorgb installation guide which doesn't cover PAL 1chip

There's something not correct if we follow each guide

I have found a via where to get csync but still something is missing in my setup

I am also confused about this :

"Decoupling caps at the MultiAV

There are some 47pF ceramic capacitors at the MultiAV at pin 1 - 4, namely C44 (pin 1), C45 (pin 2), C46 (pin 3) and C47 (pin 4). There are also some footprints for 47pF ceramic capacitors on the modding board , namely C14, C24, C34 and C44.
Make sure that you either have for each pin at the MultiAV just a single cap installed. If you unsure, just left the footprints on the MultiAV free as these caps are optional.

To sum up:

1Chip-01/02 versions have all capacitors C44-C47 on board. So left C14-C44 on the modding board unpopulated.
1Chip-03 versions does not have C46 installed. So you can put C44 onto the modding board. C44 is directly next to R43.
SNES2 version do not C44-C47. This version even does not have footprints for them. So assemble C14-C44 on the modding board."

I have a 1chip 01 and I can't find these C14-C44 reference for the caps neither on the OSH on the GITHUB or on my VGP board.

Seems likke C14 and C44 references are outdated

I'm getting no blue color so I think the mod is pretty untested for PAL

Anyone has a suggestion ?

Cheers

[kel]

If there is only one color missing then it is more likely a bad connection or short circuit somewhere. Did you remember to insulate the underside of the board before installing?

I have installed these boards on PAL consoles and they are fully compatible.

As you mentioned, the capacitor numbers on the amp do not match the one's on borti's github page. This is because the board has been redesigned and is only based on borti's original design. It was designed to fit all models with the removal of any duplicate caps from the snes due to it being mass produced, whereas borti's original design is for DIYers that make a few boards in which case it makes sense to just not solder the duplicate caps to the amp PCB depending on which snes version it is intended for.

All the duplicate caps that need to be removed for each model are listed on the VGP install instructions and it looks like you have followed them for your SNES so I wouldn't really worry about it to much.

[archimage]

Thanks for your answer,

Yes I took care insulating everything.

I also tried to remove the amp and redo the installation, still the same problem.

Now when I measure the voltage on the via for the blue on the Snes it shows zero.

I suspect the blue signal is dead from the cpu, is it possible ?

[kel]

I don't think it's likely to be a problem with the output from the CPU, it's the last place I would look.

Do you have any pictures of the install?

I'd still bet my money on a short circuit somewhere. It can be so easy to leave a hair width of solder behind by accident and not be able to easily see it with the naked eye.

EDIT: test between blue and GND with your multimeter and disconnect the blue connection from the snes to the amp, this way you should be able to determine if the short is on the snes or the input/output of the amp.

[archimage]

actually that's what I did, I'm testing directly pin 157/gnd without the amp and it gives 0.01mv

[kel]

What value does red and green come to and are you outputting a full white screen when testing. I ask because testing these signals with a multimeter is not the most reliable method so we would at least need to know that we are getting a better reading from red and green.

Also have you checked blue to GND? You should get a reading of about 150-160ohms. If it's a lot less then there is a short somewhere.

EDIT: Sorry I'm just assuming that you only have access to a multimeter, please correct me if I'm wrong

[archimage]

I get 160 ohms on the green, and nothing on red and blue when I test resistance on amp input

I get 400mv on green and red, and nothing on blue

The screen has a yellow tint

[kel]

I think we are getting crossed wires, no pun intended.

I didn't mean what color your screen looked. What I meant was that you need to be outputting a constant full white screen to check signal values, ideally. There's no point in testing while running a game as the signals will be all over the place. 240p test suite has a full white screen pattern that will work.

A minute ago you were testing at the pins of the CPU with blue disconnected from the amp and now you are testing at the amp. We really need to stick to one place at a time if we are working together otherwise it gets a little confusing.

Please check those signals to GND again as that really doesn't sound right. Make sure that you have a good connection with the probe as sometimes flux residue can interfere with getting a good reading.

[archimage]

That's what I did, I tested on 240p suite with white screen.

I took the measurements actually from the vias that connects to the cpu/amp so it wasn't the same thing as disconnecting it from the amp you're right, however if it's not connected to the amp I have no picture on the screen so I took this route.

Now that we talked about shorts I suspected the board installation, or the board itself. So I removed the amp board completely to make sure it doesn't interfere at all. Also the SRGB chip has the three legs lifted.

So I can't test with the white pattern anymore because I get no picture, however with a cartridge inserted with the sd2snes screen blue is still close to zero.

[kel]

Ok, looks like we were just getting crossed wires.

With the board removed this is the perfect opportunity to test the snes and board separately for shorts between blue and GND. I really wouldn't concentrate on the signal value at the moment until we have determined if we have a short or not and also what the values are for each colour to GND on the SNES board. No value for red and blue to GND seems highly improbable and if correct then although unrelated to this issue it would still be a problem.

[archimage]

Ok, I get 160 ohm on each color when the console is off measuring from the gnd to each color.

[kel]

Doesn't sound like there is a short on the snes then. Just to be sure check blue to GND on the multi out pin also, it should be open circuit.

If everything is ok on the snes then test blue to GND on the input and the output side of the RGB amp PCB.

[archimage]

no short on multiout, neither on rgb amp in/out however the amp is completely dead now, it's been destroyed while removing it

[kel]

Damn, that's a shame. I can't say that I ever managed to get one out without destroying it, you were lucky to even do it once.

That's really strange that there is no short anywhere on the snes or amp board. Well I suppose all that is left now is to put the snes back to it's original state and see if it still works with the S-RGB amp or if blue is still missing.

[archimage]

I don't have the resistors/caps anymore,

Is there any way to test without those ?

[kel]

If you are up for it you could take the resistors and caps from the now destroyed amp board, they are a little smaller than the ones originally in the snes but it's doable with a bit of extra solder to bridge the gap. Don't worry to much about the caps for now, they are not super important just for testing purposes, it's mainly the resisters you can't do without.

[Link83]

Which legs of the S-RGB chip have been lifted? The RGB input or RGB output pins?
Keep in mind bending the pins up and then down again can cause them to snap off.

[archimage]

I lift the output ones, but I managed to solder them back on the pins.

Which resistors should I transplant ? There are many on the board.

Also what about csync ?

My cable has csync with 480ohm attenuation. How can I manage this ?

[kel]

It's the 3x75ohm resistors to the right of the THS chip and the pale colored capacitors close to them. Ignore the resistors and darker caps to the bottom right.

Unfortunately there is not a lot you can do about C-sync right now as the parts on the amp board are not interchangeable with the snes. The best you can do for now is just connect the C-sync pin to the C-video pin or Luma pin on the multi out. Just make sure that you don't put the original parts back on that circuit if you still have them otherwise it will send 12v down your cable.

Also bear in mind that your video and sync levels will be off unless you have a spare pal scart cable but it should be ok for testing purposes.

EDIT: on second thought, if that's the only cable you have then 480ohm attenuation may be to much for C-video or luma. It shouldn't do any harm trying but if you get no sync then you may need to modify your cable if possible.

[archimage]

i actually have an og snes cable and ntsc csync, I know I won't put back the 12v line, I understood it's the D1/R28/C46 line.

But i'm still confused on the difference.

I'm going to give a try to the resistors/caps transplant

[kel]

Looks like you replied before my EDIT so disregard what I said as it doesn't matter if you have a pal snes cable.

Sorry, you don't understand the difference between what exactly?

[archimage]

I managed to solder back all resistors and caps, tried to hook up with the PAL cable, picture is back but still no blue in the air...

I have found a defective resistor btw which I replaced

[kel]

This really is a difficult one

There's no short to GND anywhere so either there is a short to something else other than GND somewhere, there is a break in the blue circuit somewhere or you were correct originally and the blue output from the CPU/DAC has blown.

The only things I can think of now is to check every inch of the blue circuit with a magnifying glass, do a continuity check from pin 157 on CPU to the input of the S-RGB chip and if everything checks out ok then unfortunately it looks like the CPU/DAC output may have blown. The only way to know for sure would be to test at pin 157 with a full white screen and preferably an oscilloscope.

Did you solder to vias when you installed the amp? If so are you sure that the trace to the via wasn't accidentally damaged as they are quite thin and it is possible. I've damaged a few vias in my time, it's just something that can happen sometimes. I've accidentally ripped vias right off the board before and I've also had to repair installs for others where they have accidentally ripped a via from the board.

[Link83]

Keep in mind that the stock PAL SNES video circuit is completely different to the NTSC SNES.

Whilst the NTSC 1CHIP SNES basically just uses three 75ohm resistors in series from the S-RGB chip, the PAL 1CHIP SNES uses transistors (Q1/Q2/Q3) and 39ohm resistors (R15/R16/R17) and expects the SCART cable to have three 75ohm resistors connected to ground inside the SCART plug (Instead of the three 220uF capacitors as used in NTSC cables)

Plus as you are already aware, MULTI OUT port Pin 3 carries +12V instead of CSYNC on PAL SNES consoles.

[archimage]

I was hoping it was the via so I connected cable directly to the cpu blue pin. No success. It could be the via got shorted with gnd and breaks the blue line but in that case it would beep when tested with gnd. It doesn't.

[kel]

Really sorry to hear that I was hoping we would get this issue resolved. I can't think how it could have blown but at this point it doesn't look like there are any other obvious causes.

Let us know if you do manage to figure something out though, really would be unfortunate if there isn't any way to salvage the SNES.

[archimage]

Yes it's a shame everything else works fine