Home	Forums	What's new	Resources

Saturn: Low pitched and slowed down audio after recap

Xeauron - Monday at 9:23 AM

		Xeauron	Monday at 9:23 AM
		Hi all, I've just recapped a VA0 Saturn - made a bit of a hash of the through hole caps as the board is really thick and the holes are attached to ground planes thus making it even more difficult to clear the solder - So I did make a mess that area. I managed to get it done in the end with liberal use of hot air, capton tape and tin foil as a heat shield, though it did use a lot of heat on that area of the board. For the rest of the board I used solder tweezers and kept heat exposure to a minimum. 99% of the caps popped off nicely (one or two of the larger ones put of a bit of a fight, but nothing to write home about, all pads were in tact and cleaned. The board is fully recapped now and powers on, however there's an odd issue I'm puzzling over and maybe someone here has come across this? Everything about the system seems fine, except sound effects produced by the onboard sound chip are pitched down, like a cassette tape or record playing at a quarter speed. CD audio plays perfectly and at the correct pitch, only onboard sound effects / music are played really pitched down and slowly, it's a bizarre thing to see. If I boot Daytona USA, you wouldn't know there was a problem until you get into the select screen where the voice is pitched down which says "Please select a vehicle" So Saturn is stable and can play games via a Satiator, just the audio is pitched down. Controls: Good CD Audio: Good DAC Chip: Good (would CD audio be a problem if this wasn't working?) Onboard Audio: Faulty - Pitched low and slowed down VDP1 & 2: Good (Booted and played Daytona USA, Radient SIlvergun and Sega Rally successfully) CPU1 & 2: Good (I think - not sure what each of them do) Video - 50Hz / 60Hz: Good (within range, switched fine - signal stable, colour and refresh fine confirmed via OSSC) PLL Chip: Good (as far as I can see) CD Block: Good Real time clock: Good - appears to be keeping time fine (I timed it with my phone stopwatch) Games appear to be running at the correct speed, though I can't do a side by side comparison!) I've traced pin 16 of the Motorola MC68000 to the resonator component X3 using a multi meter in continuity mode. Could this be the culpret? (it's the blue package in the middle of the pic below the big sliver video clock crystal.) My reasoning is that clock crystals are apparently quite heat sensitive, and crystal X1 (video timer) seems fine, X4 (RTC timer) apppears to be keeping time, X3 is connected to the sound chip and apparently the CPUs (I asked AI - though it loves to contradict itself hence why I'm asking here). I traced this using the service manual (attached). The pins around all the chips including the sound chips look tip-top, shiny and test positive for continuity with the pads. Is there anything I can do to keep this Saturn out of the donor bin? Anyone have any ideas?

		rorirub	Tuesday at 4:06 AM
		SCSP or 68k is not getting proper clocks. Check SNDCLK on TP21, SCPCLK on TP22 and EXTCLK on TP92. EXTCLK should be ~ 8.464MHz, this is the clock signal output by the CD drive, Saturn receives it through the ribbon (pin 9 or 12?) -> PLL pin 6 to generate SNDCLK. SNDCLK should be ~22.58MHz, from PLL pin 17 -> IC24 pin 1 -> IC24 pin 2 -> TP21 and SCSP pin 14. SCPCLK should be ~11.28MHz, from PLL pin 5 -> IC24 pin 11 -> IC24 pin 10 -> TP22 and 68000 pin 16.   Xeauron said: CD audio is entirely handled by the CD drive and it is digitally mixed into the SCSP output by a Philips chip (IC33). VA6+ units omit this and use the DCC (IC6) to do the same thing. The SCSP/68k could be toast and you'd still get CDDA music assuming the machine boots.   Xeauron said: Possibly, because X3 is the SMPC clock, it should NOT connect to the 68000 clock input. Are you sure you didn't just accidentally measure some ground lines? If you really have continuity between X3 and 68k pin 16, I think you buggered something.

		Xeauron	Tuesday at 6:55 AM
		Thank you for the info! Really appreciate it. I'm going to need an oscilliscope aren't I... Any recommendations on a good one for a hobbyist thats accurate and doesn't need a remortgage of a house? (Up to about £450 - £500 max I reckon - do I need to spend that much?) I'm not above buying one here if you think thing is salvagable. I'm putting together a video to run you through what I did and what I plan to do and how it looks and runs now to give you a better idea. I'll edit this post with the vid ASAP, just need to shoot and edit. EDIT: Removed (vid and links redundant)

		rorirub	Today at 4:45 AM
		Xeauron said: Some multimeters can read clock frequencies. You don't need to make accurate reads for this, just to make sure the clock signal is there. But I can already tell the clock signal is missing, because you are running the machine without a CD drive. The 8.464MHz signal comes from the CD drive on VA0-9 units. No CD drive = no sound clock = the sound issues you mentioned (which I can reproduce by running my units without the CD Drive). So put the CD drive back and test it again. Even if the drive has a broken laser, as long as it still has working chips to output that clock, it will be enough. The x1/3/4 crystals are fine, the console wouldn't even work if they were not.

		Xeauron	Today at 5:41 AM
		Mate........ It's working perfectly. Now I'm embarrassed for a different reason, I feel so stupid lol. That's amazing, thank you! I just hastily reassembled it with the CD drive like "Surely not...." then boom, sound at the correct pitch on the BIOS. I can't thank you enough. You've just made my week. Thank you. Once you start taking to chat bots, they basically just reaffirm what you're thinking, hence why I wanted to speak to a human. I know the Saturn is a complicated beast (that's why I love it), but I didn't bother to reassemble that far as I thought it had an issue. The CD drive....omg.   rorirub said: That's what was playing on my mind, I was thinking how could the sound not be recieving the correct clock but the CPUs and everything else is fine. I also didn't think I hit it with anywhere near enough heat to cause issues.

		rorirub	Today at 5:52 AM
		Xeauron said: I did say that the clock path goes from the cd drive, to the pll, to the sound chips...

		Xeauron	Today at 6:57 AM
		Yes, for some reason I thought you meant the CD block daughter board. I know you mentioned the ribbon - I was in distress thinking I'd just ruined a Saturn, and didn't read your post properly. Again, thank you. I do still need to recap the CD drive board, if you know of anywhere I can find a disassembly instructions - let me know. I'm waiting on one of those bolt action desoldering tools before tackling that.

		rorirub	Today at 7:17 AM
		The CD board only uses THT caps from what I remember, you can easily remove those even with a chisel tip soldering iron. I don't know what you mean by instructions - disconnect the drive, identify the caps, solder them out, solder in replacements, reconnect it.

		Xeauron	Today at 8:30 AM
		rorirub said: I mean disassembly instructions - it's fine I found a video. I've never done it before so it's nice to have steps - where to pinch the plastic, where tabs are, where screws are - how to get the board seperated ect. Yes I could work it out, but it's quicker to have the steps already laid out that's all. Thank you again for all your help, I really appreciate it.

		rorirub	Today at 9:40 AM
		But... you already have the unit disassembled, I don't understand. Either way, it's just unscrew -> take top off, then pull stuff out one by one.

		Xeauron	Today at 1:00 PM
		rorirub said: What? No I don't... The self contained CD-ROM assembly I plugged back in is still in one piece. Needs cleaning for a start, there's bits of fluff and hair inside from wherever it's been (Ebay purchase), then it needs recapping - caps that are under the metal plate which are awkward to reach.