Repairing a VA15 with a Strange Black Screen Issue

[pastperfectgames]

Hi Everyone,

A fellow modder and I had encountered a strange set of issues with a VA15 board where with the lid open/CD drive unplugged the console would display the date/time screen but with no audio while toggling the values with the cursor, and would then freeze at the non-animated BIOS splash screen and never proceed to the CD player screen.

With the lid closed, the console would black screen. 

On top of that, the laser would not focus (bop up and down to try to detect a disc), although it would slide down along the rails.

Using another VA15 donor board I had (which, incidentally had black screen'd after I had replaced all the electrolytic capacitors - but that's another story), with a help of a fellow modder I started to systematically replace one chip at a time to try to see if anything could be done to at least save one of these boards.

It turns out replacing IC38, the CD-ROM RAM chip, allowed for the laser to start focusing again (this is the same chip as IC19, the sound DRAM).

Figuring out the black screen/freezing at the bios took a lot more time and effort (hot airing each chip off, soldering them on, ensuring all the pins were making a solid connection, reflow as needed, etc.), but finally after getting to IC46 and IC47 and swapping those the Saturn sprung back to life!

These two chips are 74HC161A and are some binary counters (which are still sold), which in VA15 are tied to IC17, the audio chip. For *earlier console revisions*, analogs may present as IC23 and IC24, which would be tied to IC20, the PLL chip. Maaaaaaaybe these fail in those consoles with black screens where a recap doesn't help?

Definitely something that needs more research, but I'm really happy we were able to save at least one VA15! 

[zyrobs]

Thanks for the info! Must've been a mammoth of a task to reflow all the components one by one. Respect.

IC46/47 are only present on VA13 and VA15, plus on some VA10/11 boards but there only sometimes, as a sub-board. These boards have the sound parts combined into a single IC. Evidently something was not working properly there, so they came up with that fix mid-production and then incorporated it into VA13. I don't know what the chips do exactly, but from memory they are tied to the RESET line of the SCSP. It would make sense, that if those chips are broken, the audio section would not function at all and this would also prevent the booting process.

If the pickup is not moving its motor or bobbing up/down in an attempt to read, then it's the CD Block components that need to be inspected / replaced. These work completely independently from the rest of the system - they'll do their job even if most everything else is broken.

[pastperfectgames]

Yeah, we started with replacing IC37, IC5, IC1, IC14…all chips with 150+ pins! Sooo much tacky flux to clean up afterwards!

The big issue was we had absolutely no idea *where* to even begin to look, since there’s so little proper troubleshooting documentation available, besides the observation that recapping early revisions typically fixes black screens. This doesn’t *appear* to be the case with latter revisions, but again it all seems very anecdotal since there isn’t any sort of pooled knowledge source anywhere, really.

It would be amazing if these generic counter chips are what’s causing certain black screen issues, since those are of course much more readily available than Sega’s custom chipsets for replacement. And IC23 and IC24 appear to be responsible for timings as well in at least the VA0 and VA1 models, though they’re a completely different chip (different number of pins) and are tied to a different Saturn function.

Yup yup, the CD-ROM issue was a completely independent event; it was just an interesting coincidence that both areas used the exact same RAM! This finding may help those who believe they have a dead laser, when in fact it’s just bad RAM.

Overall, what makes all of this difficult is that even a loose pin on almost any chip can cause a black screen. So perhaps the first order of business would be to just reflow everything first and see what happens. Even then, it’s a lot of work.