Saturn model 1: no video, no sound

Hello,

I'm trying to repair a model 1 Saturn (model 837-11892-01: PAL).
Here is what I tried so far:

• checking the power supply values: they are higher than needed (3.6V, 5.4V and 9.7V), so I believe I don't have a blown capacitor
• one of the capacitor was slightly crushed because of the top shield, I replaced it to be sure (it was the CE58, which doesn't exist on the US board?)
• I resoldered the audio/video connector's pins
• the TV shows a black picture instead of "No signal", though I find the video cable quite hard to plug inside the connector
• I guess it was opened before because it was missing two screws, but other than that the board is pretty clean

Also I did all the tests without pluging in the CD drive or the controllers board, which works on my model 2, but I'm not sure about the first model.

Any ideas to fix it?

 

check the caps and inductors next to the video output. one of the caps on some pal units are known to go bad. sometimes the trace gets burned to.

 

Are inductors the row of black squares? How can I check them?
I checked all the caps and none of them seem to be blown, I was able to read a value when I checked on ohmmeter mode.


But I found something fishy on the VDP1:

I cleaned it but the two pins stayed dark, is it possible that it's dead?

 

o that might be bad. hope it isnt blown.

my bet is still the cap you replaced ce58. i think thats the one whos trace often get burnt. if i remember correct, one side of it goes to a resistor on the other side of the board and then to the CXA. check for continuity between the cap and the resistor and the resistor and the cxa.

the threlegged squares are EMI filters, they can be bad to. especialy the sync one.

the inductor i have found bad is a small black square. it should read as a short not open. its in the same area.

 

So I checked the traces for the CE58: the - side goes to the ground, the + side goes to pin 1, which is used for the +9V on PAL units.
It explains why it doesn't exist on the US variant; but at the end it's pointless, the RGB cable doesn't use it, pins 8 and 16 on the scart are connected to the Saturn's pin 4 (+5V). I didn't pay attention before but my TV recognises a 16/9 signal but I configured it to display it at 4/3.

I removed the resistor on the pin 16 to display the composite signal: it works on my v2 Saturn and it's ugly as expected, but still nothing on my v1.

I also checked all the inductors from L1 to L9 (I believe L7 is the one close to the EMI filter, though it's not labelled) and all of them read as short.

I start to believe it's a bigger damage, because even if the video was not working, I would still be able to here the beeps when I try to set the language and time.

 

bummer. might be the vdp then. time to check voltages and maybe use a scope on the vdp and cxa?

 

I never did an advanced troubleshooting like that but I would be happy to learn.
Do you have links explaining what I should check and where?

 

retro rgb did a video on checking sync with a oscope. that might do. if not. just try comparing waveformes with your working one. or give up. seems like a lot of work for a console that migt be dead for good. but then again.....

 

An O-scope will help a lot here. I would troubleshoot a little further first, u still have a lot of variables.

Have u tested your working Saturn with the same TV and cable setup? This will rule out the cable.

Does the TV say normally say 'No signal' when nothing is plugged into that input?

I would do tests with both the cd drive and controller port attached. Some consoles require them to be attached to function properly. Does the cd drive start up and seek the disc?

Do u hear any sound, turn volume up.

With the scope, start from the psu as this is usually a cause on non functioning IC. Is the 5V line clean.

Probe the CVBS (composite video) output. If the signal is poor then it could be a bad cap. Bad caps don't always look bad and only testing with an ESR meter will tell for sure (resistance test doesn't work.)

Good luck.

 

Is it this one?
But I don't mind to work on it, even if at the end it doesn't work I will have learned new things.

Yes, same TV, same cable, I only have one.

Yes, it says "no signal" when nothing is plugged, or when the cable is plugged but the console switched off.

So when I start it it should goes to set language and date settings. If I get it right, I have to press A, A, Left, A to go to the menu.
When I press them blindly, with the lid closed, the CD doesn't start to spin.

No sound at all.

I will do that tomorrow
But I don't believe it's from the PSU, because I swapped it with the working one and it didn't change anything.

 

Ok. That helps a lot. So the psu can be ruled out.

The cd drive should still spin up a disc and initially home in the laser at power on.

Does your Tv normally play sound if it doesn't detect a picture? If you have RCA connectors maybe plug the audio connectors into a hi-fi to check for sound.

Looking at the Saturn block diagram, the sound is processed separate from the graphics so the problem could be further upstream from the VDP chips (as these don't affect sound). Could be issues with the SH1/2 or PLL chips. Double check all the pins are clean and remove any crust bridging them. Check all the caps as well. Maybe a full recap might be needed.

Good luck.

 

My TV does try to display something: instead of saying "No signal", it shows a black screen, I believe because it receives +5V on the pin 8 and 16 from the scart.

So I made a mistake when I said it's the VDP1: the chip is labelled 315-5688, while Sega Retro used its picture to talk about the VDP1 for some reason, it is in fact the SCU.
And if I believe the block diagram (page 18 on this pdf), it's basically the heart of the Saturn.

 

If you get no audio/video, but the display is switched on and stays black, then the system itself failed to run its boot code. No point in checking for things like the cable or the video encoder CXA chip, since the system just didn't get far enough to send any output to be shown by the a/v encoder ICs, so you get black screen and no sound. If you got sound but no screen, then it could be the cables fault or the video encoders fault, but that is rare.

If the cd drive centers the optical pickup, then the CD block (the whole SH1 subsystem) is working and can be ruled out.
The WORK RAM LOW (IC31 and IC32) can also be ruled out, since the boot code doesn't use it, although games do.

You won't get anywhere probing the video output, the system dies before that, in fact the video output is probably functioning correctly - the console is just not running meaningful code to display anything.

I'd try:
- inspecting the board for any dead traces, broken solder joints, etc. See if it had any modification attempts on it.
- recap the system, or at least the area around the PLL.
- if you want to start probing, I recommend trying the PLL outputs first. If those don't output a proper signal, the rest of the system may not run or desync, resulting in the code halting and nothing being output by the system.

 

That's really interesting, where did you learn that?

So I tried to probe some chips, here are the values corresponding to some pins, compared to my working Saturn in (brackets) if they differ.

PLL (315-5746 (page 46)): here they wrote it generates two clocks, I only found one interesting. The quartz runs at 17,7341MHz.

• 5: varying from 1.75 to 1.82MHz (1.346MHz)
• 17: varying from 3.27 to 3.55MHz (2.68MHz)
• 18: 26.6874MHz
• 23: around 3.6KHz (varying from 1.7 to 1.9kHz)
• 28: 4.43358MHz

Except some lower values with the v2 I would say the PLL works ?!

Then on the overview manual they describe the SMPC as "resets the entire Saturn system when the reset button is pressed or the power turned on" (page 62), so I checked it.
SMPC (315-5744 (page 33)):

• 10 & 11: 4MHz
• 12: GND
• 13: 32.768kHz
• 14: 1V
• 15: 5V
• 30: 784.92kHz (varying from 760k to 810kHz)
• 50: 9V
• 51: 9V (varying between 0V, 5V, 9V)
• 52: 9V (GND)
• 56: 9V (GND)
• 61: GND

Some data pins are different, but maybe I couldn't properly see they were switching?

Then I checked three data pins on the SCU on the v2:

• 69: varying from 250k to 275kHz
• 71: around 300kHz
• 73: around 300kHz

I could see a lot of data on the scope, but with the v1 it was a fixed 5V.
So either I didn't check properly the PLL (two clocks?) or the SMPC (which pins?) but the SCU is definitely not working.

 

All the signals coming out of that PLL chip should be stable - if they are wandering around, it's probably bad.

 

I had a unit with one of the LOW RAM chips dead (it booted up fine, but SOME games didn't work and some had corrupt graphics), and I once tried running the system with the SH1 desoldered (no boot, cd drive didn't move the pickup).

Actually you may still have CD drive issues regardless, for example if the data returned from the drive is bad, for example if some data traces are broken. But those could be in many other parts of the system too. And either way, if the pickup moves, then the SH1 at least goes as far as running its own firmware and telling the drive to move the pickup... so it is less likely to be broken.

Looking at your values, you did not specify which board types your other unit is. You said it's a model 2, so I'm assuming that it is either a VA5 or a VA9 unit (they are the most common PAL model 2s); those are pretty close to the VA1, closer than any other model 2s, so the values should be comparable.

PLL pins 5 and 17 generate clocks using pin 6 as the source, which they run through a phase detector. Test PLL pin 6; if it is a stable clock, then either the caps around the PLL are bad and brake the phase detection, or the PLL itself could be broken. This is my guess.
If PLL pin 6 is already unstable, then it may be something in the CD drive being broken, since that's what inputs that specific clock.

I think the SCU should be fine. It mostly just shoots data everywhere around the system.

I haven't actually tried to see if the system boots with the 68k removed... I should do that some time.

 

Thanks for the explanation

My bad, it's a PAL SD - the one where it's really easy to add a 50/60Hz switch

I tested everything without the controllers board or the CD drive connected. On my v2 the Saturn starts, though the sound is slower and there are red balls instead of the green squares, maybe the v1 is not as tolerant?

There was nothing coming on the pin 6 (pull-down?), I would guess the clocks generated on pins 5 and 17 are generated from the noise on pin 6. This pin is only used by the CD drive?

 

But, you measured the VA SD model without the drive too, and it had stable clocks? Or you measured that with the drive connected? I wonder where it gets those clocks from, without any input.

On all the models that use the 315-5746 PLL, pin 6 comes from the CD drive, if it is missing, then the entire sound block would be out of whack (which is why the blocks on the cd player are red and the sound is bad).

Still, the bad clocks on the PLL is what I'd concentrate on fixing.

 

The guy is that video is terrible at explaining a scope and what ac/dc coupling does.

 

just for reference, the clocks I get on the PLL are:
pin 5 - 11.28 MHz (1.34 MHz without the CD drive)
pin 6 - 8.464 MHz (0 without the CD drive)
pin 17 - 22.57 MHz (2.68 - 2.60 MHz and dropping, without the CD drive)
pin 18 - 26.86 MHz (since it is in 320px mode)
pin 28 - 3.578 MHz

all measured on a VA0 jp board.

note that I measured them with my multimeter, so these numbers are probably not the most accurate.