PC-Engine Duo with video problems and stuck laser

[Honky_Bonk]

Back-Story:
I bought a PC Engine Duo from Japan. The seller stated that HU-Cards are playing, but CD-Rom Games would not load.
It came without any cables or controllers. So I had to make my own AV-Cable and power supply (used a 9v 1A).

Symptoms on arrival:
Right from the start it showed garbled graphics on the BIOS screen, but I had to wait for my Turbo ED and a Controller to check more. I noticed that when I quickly turned the console off and on again, the BIOS screen was perfect. So I assumed it had to do with the caps.
After replacing all the caps the problems remained the same unfortunately. (I soldered in C525 the wrong orientation at first, but corrected it later without any difference.) When my Turbo ED arrived the menu was always showing correctly, but games sometimes had glitchy graphics, sometimes not. It almost felt as if the console needed time to "warm up".

The CD spindle actually only needed a little push to make it work again. Seems like it just got a little stuck at some point. I didn't bother too much adjusting the laser (it would only read to around the 50 min mark), since CD games also had glitchy graphics.

What I did so far:
None of the caps seem to have leaked badly, though some had corroded pads. But I was not able to find any bad traces on the board. So I decided to reflow all chips in case one of the legs somewhere wouldn't have contact. I might went a bit to far here re-soldering literally every "chip", even the smaller IC's and mosfets. I used a soldering iron, lots of flux and only quickly touched the legs. After that the symptoms changed / got worse a bit. I accidently bridged two pins on the largest chip on the bottom (the NEC D91317GD somewhere between 70 and 75 I think). I don't know if this caused any damage.
I also removed another bridge between Pin 41 and 42 on the HuC6270 later on. After this the graphics are not messed up any more.


Status now:

1. Somewhere in the process the laser got stuck on the outer edge a would not move anymore. I had to bring it back to the center by hand, but it won't read discs anymore (no spinning). The BA6290A on top of the board becomes extremely hot.

2. I get some distortions on screen that are best visible on black background (attached a video), that start to appear a minute or so after turning on the console and getting worse. The colors are also a bit off. Like the text on the BIOS screen turns from blue into some sort of cyan. This also applies to HU-Card games.

No idea if the 1. problem causes the 2nd. I can not find anything obvious on the board anymore. Any ideas where to start fixing the problems?


(It's getting worse at 0:36)

[Keith Courage]

Graphical glitches can be caused by 3 things. Usually it's bad traces or connections in between ram and HU chips but sometimes it can be a bad Ram or HU chip as well.

I'd start by looking under the card slot on both sides of the board for corroded traces or via holes.

[Honky_Bonk]

@Keith Courage

Thanks for your answer. The thing is, the board looked almost pristine to begin with. I was not able to spot any bad traces or vias, altough I checked it several times. I tested coninuity between the HUC-Chips and VRAM but they seems to be fine.

Meanwhile after poking around quite a bit the situation got worse. I rarely get a BIOS-Screen or Hu-Card boot. Usually after I let the board sit for a while. When the BIOS comes up, it's only for a second or so before the screen goes blank. Most of the time it's just a white, purple, blue, etc. screen. Wiggling and moving / slightly bending the board sometimes brought back the BIOS. So I was very confident, that there is a intermittent connection somewhere, but I was not able to find it yet.

I just checked continuity on all HUC Chips to check for shorts of close by pins. The HUC 6260 gets slightly warm (maybe 30 degrees celcius) and I found (not visible) shorts between 42&43, 45&46, as well as 52&53, 53&54 and 55&56. According to the Turbo Grafx Service Manual the later three should be normal, but the first two not.

- Since I do not feel very secure reading those datasheets, could somebody verify that 42&43 as well as as 45&46 shouldn't have continuity?

- Does anybody have any idea what else than a bad HUC 6260 could cause these shorts?

[Honky_Bonk]

Update:

A thorough cleaning of the board with lots of IPA and a toothbrush made the graphic errors go away. Have been playing Neutopia for a few hours without issues and BIOS also comes up.

My problem is the CD-Section now. Whatever I do, I get the message "Please close cover". The switch works. I get pulse on pin 8 of IC104 and when I hit the switch I get pulse on pin 33 as well.
But it seems the BIOS does not get the necessary signal from IC104. According to a post by theSteve that I read somewhere, the signal should be send through IC105 to IC501. It seems that he posted a pinout for IC501 at some point, but I'm not able to find it anywhere.

BTW, should the "Please close cover" message on a working Duo come up right away, when the system is turned on (switch not pressed down). Because in my case it only appears after I hit Run.

[Keith Courage]

Yes, if the cd lid is open, that error message is supposed to display immediately if you hit run.

Unfortunately I can't tell you where to check to fix the issue you are having. I've never seen that happen before where the system always says please close cover even if the switch is good.

[Honky_Bonk]

I see 

I replaced the diode D102 on the bottom, casuse I could some weird voltage levels there, but the problem remains. As I said, my assumption is, that the BIOS does not receive the info from the chip that the lid is closed. It should pass through IC105 and IC501 (the large chip on the bottom). Has really nobody saved the pinout that theSteve made for this chip?

If I knew, which traces to follow, I might find, where it gets lost...

[Honky_Bonk]

Update:

I got rid of that "Please close cover". Reason was that through-hole capazitor C136. I might have bent it a few times too often. I replaced it with a new one and the problem was gone.

Now I get the following behaviour: Right when I turn on the system the laser moves to the far edge and stays there. When I hit the lid-switch I get "Just a moment", but nothing else happens...

[Keith Courage]

Man, you are having some of the strangest symptoms ever with that DUO. I have never seen a lens motor do that when first powered on either. At least that I can remember anyways.

I wonder if maybe one of the Large BA chips is faulty. They are the Larger chips right next to the two 100UF caps you just mentioned.

[Honky_Bonk]

@Keith Courage

The system indeed acts absolutely weird. The capazitor in fact did not make a difference, cause the "Please close cover" came back. Then I found a broken leg on the BA6290A on the bottom. (Wiggling on the capazitor might have caused it to get contact.) Sure enough I fixed the leg, the message was gone once again for a while. But the now system thought the lid was closed all the time.
But, of course, the problem with the "Please close cover" message came back. Only progress is: The motor does not push to the outside anymore. It does nothing at all. Though the lens moves up and down when starting the system.

You're right, the BA6090A chips might be bad. Since they're not cheap where I live, I'll see if I can harvest some from an old CD-Player.

Then again, the Bios sometimes refuses to boot. So I might have very well broken traces somewhere I'can't find, which prevent signals from reaching the Bios...

[Keith Courage]

Usually bad traces or vias that cause the no Bios on boot are under the card slot. Trace all hu card pins to their destinations to make sure all have continuity.