VOGONS


First post, by cjreha

User metadata
Rank Newbie
Rank
Newbie

Hi folks,

It's been a while since I sat down and played with Wintel-era hardware, but the itch recently came back and I've been scratching it by finally finishing off a nice Pentium III build I started years ago. One of the things I always love doing when putting machines together is finding and using interesting period disc drives, and for this build one of my picks is a Kenwood 72x "TrueX" CD-ROM drive from the height of the drive speed wars. It's kinda-sorta cheating to get that rating, it's not a "raw speed" 72x like you'd normally associate with its disc-shattering contemporaries, it simply uses multiple lasers to read multiple tracks at once while spinning the disc at a much saner speed. It's a drive I'd sought for years, and finally last year found in a box of drives at one of the various VCF swap meets.

Anyway, to get to the point, I plugged it in for the first time today and was met with absolutely nothing. Unfortunately, it's stone dead. Won't even push out the tray, the status LED doesn't light up either. I cracked it open to see if anything was obviously blown up, but unfortunately aside from the usual age-related grime, everything looks fine. I've attached some good photos of the control board for anyone interested. Kenwood were actually fairly nice and silkscreened a basic block diagram of the chips on the board as well as labelling voltage/RF test points, so I was able to confirm the internal voltage rails (10/5/3.3V) are all present and correct. Unfortunately, beyond that, I'm kind of in the dark. As is kind of typical for late-era drives like this, most of the logic is packed into completely unobtainable ASICs, and my current guess is that the main Zen ASIC has failed (P/N L5A9394). It gets uncomfortably hot to the touch after about 15 seconds, while all other chips on the board either remain cold or only get very slightly warm. There are a couple other possibilities floating around in my mind, like the Winbond flash memory having failed or the early 00's SMD caps making it play dead, but I have absolutely nothing to support these hypotheses and will need to do more testing.

My question is, does anyone have any experience at all repairing these drives or at least diagnosing them? They have a fairly crappy reputation for dying in various ways, even when they were new; lots of forum posts talking about how they worked well, but many people needing to RMA them and ultimately going with other drives for reliability. So I may be fighting an uphill battle here... 🙁 If anyone owns a working drive and is willing to evaluate the heat levels of the chips and/or fill in some of my knowledge gaps I would be incredibly thankful as well. I am, of course, also interested if anyone has one they'd want to part with...

If nothing else, I hope this ramble was entertaining to read. If people were interested I'd love to post more often about some of my adventures fixing oddball CD-ROMs, it's been a very long time since I was active on the forum 😀

Reply 1 of 6, by Horun

User metadata
Rank l33t++
Rank
l33t++

With the circle on the ASIC it appears it uses part of the metal case as a heatsink (have a NEC that has similar mark when bottom is opened up)
That is all I know.....

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 2 of 6, by cjreha

User metadata
Rank Newbie
Rank
Newbie
Horun wrote on 2025-02-16, 02:18:

With the circle on the ASIC it appears it uses part of the metal case as a heatsink (have a NEC that has similar mark when bottom is opened up)
That is all I know.....

Indeed - the bottom plate has a bunch of indentations where it was pressed up against the chips to act as a primitive heatsink. I cleaned the old compound off so I didn't get messy working on the board 😀

Reply 3 of 6, by Deunan

User metadata
Rank l33t
Rank
l33t
cjreha wrote on 2025-02-16, 00:50:

My question is, does anyone have any experience at all repairing these drives or at least diagnosing them?

My experience with those "dead" units is pretty much 0% success rate. Usually the voltages are good, clocks are there, but it just doesn't do anything. Usually the conclusion is:
- If there is any RAM/ROM activity on power-on then it's either dead SRAM chips or Flash has glitches - possibly due to many years of being unpowered.
-If there is no activity then most likely the main MCU is toast - thermal death I guess. Or possibly some other ASIC is responsible for RESET control and it died, doesn't really matter as I have no spares of either.

I could swap the SRAMs but quite often these are old, by today standards unusual packages. Maybe even DRAMs for more capacity. I usually have no spares on hand, and ordering them is too expensive and not worth the trouble. And if the PCB has parts on both sides and these particular chips are glued as well, that's it for me. I'm not wasting all that time to fight the glue.

I did fix two HDDs by replacing faulty SRAM chips but only because I could get spare parts at reasonable price, and I could see activity from the MCU. Feel free to post any success stories, maybe I can learn something new.

Reply 4 of 6, by cjreha

User metadata
Rank Newbie
Rank
Newbie
Deunan wrote on 2025-02-17, 22:57:
My experience with those "dead" units is pretty much 0% success rate. Usually the voltages are good, clocks are there, but it ju […]
Show full quote
cjreha wrote on 2025-02-16, 00:50:

My question is, does anyone have any experience at all repairing these drives or at least diagnosing them?

My experience with those "dead" units is pretty much 0% success rate. Usually the voltages are good, clocks are there, but it just doesn't do anything. Usually the conclusion is:
- If there is any RAM/ROM activity on power-on then it's either dead SRAM chips or Flash has glitches - possibly due to many years of being unpowered.
-If there is no activity then most likely the main MCU is toast - thermal death I guess. Or possibly some other ASIC is responsible for RESET control and it died, doesn't really matter as I have no spares of either.

I could swap the SRAMs but quite often these are old, by today standards unusual packages. Maybe even DRAMs for more capacity. I usually have no spares on hand, and ordering them is too expensive and not worth the trouble. And if the PCB has parts on both sides and these particular chips are glued as well, that's it for me. I'm not wasting all that time to fight the glue.

I did fix two HDDs by replacing faulty SRAM chips but only because I could get spare parts at reasonable price, and I could see activity from the MCU. Feel free to post any success stories, maybe I can learn something new.

It turns out you were pretty close - I actually managed to get it working!! In this case it was failed solder joints on the main ASIC, a high-density QFP, and the TSOP flash chip. I'd actually pulled the flash chip entirely to dump with my EPROM programmer. Upon resoldering I decided on a lark to test the drive, and it ejected and started showing signs of life! I had to clean the lens and put a bit of grease on the rails before it started reading reliably, and I also replaced the SMD electrolytics in the interest of reliability. So far it's got about ~6 hours of power-on time and use on it with no problems! I can't wait to start installing CD games to see how it performs!!

Thank you all for the musings and advice, I'm over the moon this story has a happy ending 😀

Reply 5 of 6, by maxtherabbit

User metadata
Rank l33t
Rank
l33t

Nice work! I love a happy ending

Reply 6 of 6, by d3jsp

User metadata
Rank Newbie
Rank
Newbie

Nice, I have an old one that is failing the same way. Would just reflowing the solder joints help or did you recap it as well?