Dead pc engine duo

[Mrwoodchuck]

Hi all,

I recived 3 duos from japan all with issues, i have so far fixed 2 of them..  O:)

But this last one has me stumped, I did some cleaning, replaced all the caps and all i get is a white screen mocking me when i power it up.

I replaced both 7805's and it reads 5v when the unit is powered on, i checked the cpu and all 5v pins are reading 5v.

There is no cards in the slot it should show me the load cdrom but it does not.

So i'm at a loss here, any input would be welcome..

Thanks

[pulstar]

Have you tried a HuCard/CD to see if you are getting any activity from the unit itself. Could be an AV port or output problem.

[Mrwoodchuck]

Sorry I failed to mention that,

I tried both jp, and usa cards nothing..
I hit the run button to make the drive start up like normal when you turn it on with no card in it, nothing....

thank you for the reply!

any other ideas?

[SephirothTNH]

Try doing an svideo type mod on it.  You only need to do the luma and see if you get an image.  That will bypass all the av stuff in the console since you solder straight to the Luma pin on the gpu. 

If you still don't get a picture than I would try replacing the ram chips.   Then eventually the big chips HuC62XX.

[Mrwoodchuck]

I think your on the right track with the ic's..

I tried to region mod one of my working ones and now it has the same issue as this one, white screen..

if i turn the power on and off quickly it changes colors, blue green white.. so on and so on..

I think i might have fried something when i was attempting the region mod, the fuse blew and everything..

So ya now i have 2 dead systems.. 

[BlueBMW]

Id wait for thesteve to chime in before you junk any systems.  There are other components on the hucard data lines that have beenknown to go bad and cause the same symtoms youre describing.

[thesteve]

Its changing colors/white screen, the 6260 looks good
Its likely just a broken connection
Logic probe if ya got it... start with the card slot
All logic pins must read a valid logic state

[Mrwoodchuck]

Thank you for the reply!

I Dont have a logic probe..
Right now all I have is a volt meter, I know not helpful in that situation.

I guess I will go and check for continuity on all the ic's starting with the 6260.

[Mrwoodchuck]

so I spent the better part of sunday tracing and coming up with nothing, I get payed on friday so i think i will spend a little bit of cash and get eather a usb logic analyzer or a logic probe.

what do you recommend?

I'm assuming the analyzer because i can save the data and easily compare it to another working console.

but it would be a cheap $50 and below one..

[thesteve]

a simple probe is all you need unless your reverse engineering a data bus

[Mrwoodchuck]

Ok, I will hopefully get one this weekend.

Is there any guide to see what signals i should be seeing?
I was thinking of comparing my working system, but it is acting a little funny too..

thanks for all your help thesteve, I appreciated it alot.

[Mrwoodchuck]

Well I am back, I got bought tools i figured i saw some of the saleae knockoffs and i would try one because they were so cheap.

I get my probe tomorrow but i tried the knockoff to see if there was data on the lines.
So my first power up and i found there was no data on pin 20 of the hucard connector.

i compared to my good one and on boot there is data on that pin.

Now the fun part, I reconnected it and wanted to make sure i was right on pin 20, well no pin 20 now had data but another pin "pin 22" had no data... and every time i power cycled it would bounce around..

so i am not sure if the logic analyzer is junk or what.

I will get my probe tomorrow and will report back what is going on then.

[thesteve]

lol, no they do that
your concern first is the data bus D0-D7 (the pins you clip for a region mod)
if any of them are stuck you have a shorted chip
if stuck there is a small bus chip connected by the slot (i have seen several die) pull it and test with card as bios will not be connected

[Mrwoodchuck]

Ok so i got my probe today, I have a pin stuck low every power cycle so it is down to that bus chip.

What is the ic number so i don't accidentally pull the wrong chip? Lol

In addition my second dead one has all high on the 15 - 23 pins not 18 as that is gnd..

[thesteve]

all high says the 6280 isnt addressing
your looking for IC520
note all 3 hudson chips and the ram chip by the card slot share that data bus and 1 or all of them could have a short do the same
pull IC520 first as its small and splits the system into a TG16/PCE and SCDR2 som you can work on the core system first

[Mrwoodchuck]

So I pulled ic520 from both units. They are still acting the same as before, checked around the cluster of Hudson chips still can't find anything.

Man these are a pain in the butt.. Lol

[thesteve]

Yes they are
Without a cart, and 520 removed, probe all the address and data pins at the slot (pinouts on line)
The data pins (D0-D7) should be high
The address pins all pulse

[thesteve]

If good,  insert cart
Recheck data pins (now pulsed)

[thesteve]

If all good replace ram chip

[Mrwoodchuck]

this is gonna be a long one.

system 1

without card

1 h
2 dead
3, 4 l
5 - 10 h
11 - 14 dead
15 - 17 h
18 gnd l
19 - 23 h
24 dead
25 h
26 dead
27 h
28, 29 l
30, 31 h
32, 33 l
34 dead
35 - 37 h
38 5v h


with card

1 l
2 dead
3 l
4 - 6 h
7 - 10 l
11 - 14 dead
15 - 17 h
18 l gnd
19 - 23 h
24 dead
25 h
26 dead
27 h
28 , 29 l
30 , 31 h
32 , 33 l
34 dead
35 - 37 h
38 5v h

system 2

without card

1 h
2 d
3 pulse
4 l
5 - 12 pulse
13 dead
14 pulse
15 - 17 h
18 gnd l
19 - 23 h
24 - 29 pulse
30 l
31 pulse
32 l
33 pulse
34 h
35 l
36 h
37 h
38 5v h


With card


1 l
2 dead
3 pulse
4 l
5 pulse
6 pulse
7 - 12 pulse
13 dead
14 - 17 pulse
18 l gnd
19 - 23 pulse
24,25 pulse
26 pulse
27 pulse
28 pulse
29 pulse
30 l
31 pulse
32,33 l
34 pulse
35,37 h
38 5v h

System 2 IS ALIVE!!!  apparently pin 13 was dead the connection from ic906 and pin 13 was damaged even though i looked and the trace was fine still no continuity.. I popped in my only jap game and it worked, so at this point put chip 520 back in and try again i assume? and if the 520 is back in and it does not work i assume replace that chip.

System 1 is just all over the place, it seems like the 6280 is dead?

[thesteve]

sys1 isnt  clocking
check clock pulse (likely good)
replace 6280

[Mrwoodchuck]

Awesome, thank you for all your help thesteve!!

I don't have my oscilloscope its in storage I will have to grab it later.

Is there any location you know of to buy the 6280 or is that a situation where i will have to pull one from a dead system?

I feel like I should bake you cookies or something for all your help. and i'm being serous.

[thesteve]

gotta pull one

[Xenogears]

I have the same issue on a duo my friend asked me to check. Bios won´t boot, HUCards work finde, if I use a System Card and press Start, the screen shows "Please wait..." without the Detector switch for the Lid closed. I would like to start checking Traces from the Bios-Chip but do not know, where it´s stored. Could you help me out ? THX a lot !

[thesteve]

xeno your system has a fault in the interface
it could be a dead IC501, but more likely is a bad trace to IC501 or 1 of the small logic chips near the cart slot not running

[Xenogears]

I thought so but it just a simple VIA next a CAP. C519 Ground VIA leading to IC 517 was dead, so the bios did not start to boot. Connection reestablished and it worked. Once again a via issue... but thanks a lot Mr. TheSteve. Without your posts a lot of Duos would have been faded away 

[thesteve]

 thats what i was saying about the logic chips not running

[trapexit]

thesteve, thanks for this info. Are there any service guides for the Duo, Duo-R, Duo-RX floating around? Original or like the info in this thread? A step by step of what to check?

I just picked up a Duo-R that at the moment I don't have the history of unfortunately but it appears to be a botched video mod. The 6280 and 6260 were wired up and partially covered in silicon rubber. The wires were snipped so it's possible the mod was fine but the Duo-R stopped working so they just snipped the wires to get the components back.

Either way... the device turns on. Fuse and 7805's look fine. When on it displays a white or occasionally off white or blue or pinkish screen. Placing a HuCard in does nothing. I can post images of the 6260 or 6280 if interested but as far as I can tell there aren't any shorts or unconnected pins. I removed all the silicon rubber (that was a test in patience) and desoldered the wires left over and made sure to reflow the pins.

I don't have an oscilloscope or probes but with my multimeter it looks like the data pins are high as are the address pins.

Suggestions?

If I'm reading the thread correctly it was recommended to disconnect IC520?

[thesteve]

removing IC520 splits the system into PCE and CD system
that lets you rule out the cd system in troubleshooting
its not where i start, but if everything looks good it helps
also i have seen IC520 dead causing the issue

[trapexit]

And the IC520 on the Duo-R the same as Duo? Must it be removed or can a pin or two be lifted to accomplish the same?

I noticed C947 is not populated. Anyone know if that's expected?

[thesteve]

same chip on both
missing caps are expected (often replaced by chip caps under board)

[thesteve]

if the chip is good yes a pin lift, but when its bad it wont help

[trapexit]

Noted. Thanks.

The damage to the main ICs seems like a plausible location for a problem. HuC6280's pin 16's trace to ground is damaged for example. All the silicon makes it rather difficult to get a good look at whats going on with the pins. However, without knowing when it stopped working it's difficult to know where to start.

I've already invested more time into this than it's probably worth. May end up being a parts board.