PCE DUO Controller problems.

[DeshDildo]

The quest to repair my Duo continues.  I finally have sound again after many nightmares and a few hours everyday for the past week looking into it.  I couldn't wait until I had time to play it tonight.  I fired it up and, go to play and the right d-pad and II button are not working.  WTF!  I jiggled the connector and reflowed the solder joints for the connector on the board.  None of that worked so I robbed a known good TG16 controller and swapped the cords.  Still, the exact same issue.  I checked continuity in the controller cord itself and checked few things on the Duo board.  I'm at a loss.  For the little info I've found online I believe both of my failed functions are on pin 6 of the controller port.

This bastard is really starting to piss me off.  The controller worked perfectly before I dish washed the board and had to make repairs to get sound back.

Anyone have some ideas?

[CGQuarterly]

Did you clean the conductive rubber pads with alcohol and clean the contacts on the controller PCB with an eraser?  What did you put in the dishwasher?

[thesteve]

trace pin6 back to the HuC6280
odds are you have an open trace/via

[schweaty]

Edit:  reread the original post.  nothing to add really

[schweaty]

whoops... double post

[DeshDildo]

Did you clean the conductive rubber pads with alcohol and clean the contacts on the controller PCB with an eraser?  What did you put in the dishwasher?

Yes, I did that as well just in case.  I put the whole Duo PCB in the dishwasher to thoroughly clean it after I was having intermittent static issues in the sound.  This was all after I did a full cap replacement.  I must say it worked well other than every small item that was a bit weak became non-operational afterwards.  Hopefully, this means that this Duo will be reliable as all hell when I get it figured out.  I also am not too butthurt as I got this one super cheap and, it appears someone has already repaired the disc portion so that works flawlessly.

trace pin6 back to the HuC6280
odds are you have an open trace/via

Yeah, I was working towards this a bit and followed it into an IC on the top of the board and was unsure of where to go next.  Also, I didn't spend a ton of time on it due to my frustration level.  I will go back at it tonight with a calm mindset.  Thank you again for your masterful knowledge!

[thesteve]

Sometimes the board has corrosion in vias that is still resistive
When washed they go open

[DeshDildo]

Traced all inputs to PCB back to hu6280.  I did not find anything open so I double checked power and ground (I figured it would be OK since the other buttons work and it was).  I then ran jumpers one by one to each corresponding pin to hu6280 with no change.  Since then while poking around I must have messed something up as now nothing will work.  I really love working on this Duo.

[DeshDildo]

OK, I repaired my mess up and am back to all functions EXCEPT right D-pad and button II.  According to everything I have found online these two functions correspond to pin 3.  I did run a jumper direct from this pin (after FL903) and had no change.  I also checked voltage on this circuit all the way down to pin 24 of HU6280.  I measure 5VDC to any ground on the PCB and the voltage drops to zero when I press button II.  I also get a small drop (about .2 VDC) when I press the right D-pad.  I've measured this even at pin 24 of HU6280.  This leads me to believe the input is working fine but something inside or beyond HU6280 is screwed?  I'm out of ideas for now.

[thesteve]

6280 is as far as it goes
i have seen a few 6280 chips with bad controller pins

[DeshDildo]

6280 is as far as it goes
i have seen a few 6280 chips with bad controller pins

Well... shit pickle.  If worst comes to worst, I have a spare TG16 that is rough on the outside but works.  There aren't any differences between a North American and Japanese HU6280 right?  The only difference I know of is how pin 27(?) is wired for region lock.

[ToyMachine78]

Did you clean the conductive rubber pads with alcohol and clean the contacts on the controller PCB with an eraser?  What did you put in the dishwasher?

Yes, I did that as well just in case.  I put the whole Duo PCB in the dishwasher to thoroughly clean it after I was having intermittent static issues in the sound.  This was all after I did a full cap replacement.  I must say it worked well other than every small item that was a bit weak became non-operational afterwards.  Hopefully, this means that this Duo will be reliable as all hell when I get it figured out.  I also am not too butthurt as I got this one super cheap and, it appears someone has already repaired the disc portion so that works flawlessly.

trace pin6 back to the HuC6280
odds are you have an open trace/via

Yeah, I was working towards this a bit and followed it into an IC on the top of the board and was unsure of where to go next.  Also, I didn't spend a ton of time on it due to my frustration level.  I will go back at it tonight with a calm mindset.  Thank you again for your masterful knowledge!

Can you really dishwasher a PCB or is this sarcasm?

[BlueBMW]

I may have a spare bad TG16 board (wont read hucards) if you need one for parts.

[DeshDildo]

I may have a spare bad TG16 board (wont read hucards) if you need one for parts.

Thanks for the offer.  I will let you know.  I am going to double check everything again tonight and make sure I'm not missing anything.

[NightWolve]

Can you really dishwasher a PCB or is this sarcasm?

'tis true. I thought the idea was absurd when I first heard it, and I still wouldn't do it either, I would wash it in a sink with dishwashing liquid soap, not throw it in a machine that'll be squirting pressurized water in all sorts of directions and what not.

[OldMan]

I've tried both. The pressurized water (which isn't much pressure) ensures everything gets removed. I have been told to use distilled water to rinse with, though. So I do.
And low heat with a long drying period makes sure all the water gets evaporated

[NightWolve]

I've tried both. The pressurized water (which isn't much pressure) ensures everything gets removed. I have been told to use distilled water to rinse with, though. So I do.
And low heat with a long drying period makes sure all the water gets evaporated

Yeah, if I ever did this, I would leave it to dry for at least like 2 weeks or so afterwards. Still don't really trust the idea, though...

[thesteve]

I run almost everything through the washer
Let it dry for a day after a good shake off

[ToyMachine78]

I've tried both. The pressurized water (which isn't much pressure) ensures everything gets removed. I have been told to use distilled water to rinse with, though. So I do.
And low heat with a long drying period makes sure all the water gets evaporated

Do you use cascade? Totally being serious here lol. Detergents have abrasives in them to blast off dried food and what not. Those things can etch glass. Not sure if I would trust it with a PCB.

[BlueBMW]

Ive washed dozens of pcbs this way and always had great results.  Ive fixed intermittent wierd issues that caps didnt fix by washing them.  Im a believer

[OldMan]

Dawn liquid dishwashing soap, actually. About 1 tablespoon in the bottom of the dishwasher. Not too abrasive, yet good at removing oily/sticky stuff (like excess flux).

[DeshDildo]

Dawn liquid dishwashing soap, actually. About 1 tablespoon in the bottom of the dishwasher. Not too abrasive, yet good at removing oily/sticky stuff (like excess flux).

I'm glad this method is approved as this is exactly what I did.

[DeshDildo]

Well, after work I sat here and double and triple checked everything.  I don't see any other answer than replacing HuC6280.  This is not what I was hoping for.  Something must have happened in the dishwasher as the controller worked flawlessly before that changed.

I still feel the need to save this board as I have so much time invested in it I would would pissed at myself to throw it in my electronics scrap heap.

[NightWolve]

Something must have happened in the dishwasher as the controller worked flawlessly before that changed.

Yeah, that's why I don't completely trust the idea. Maybe you didn't let it dry long enough before you reconnected it and powered it up... Just a little bit of water somewhere, trapped underneath an IC or something could've done it, who knows.

I still feel the need to save this board as I have so much time invested in it I would would pissed at myself to throw it in my electronics scrap heap.

Just send the PCB to steve when you're ready to give up.

[thesteve]

the dish machine didnt do it
wet board didnt either
please refer to the TG16 repair manual for the pin functions
use a logic probe at the 6280 and look for the pulses from the controller

[turbokon]

I got a super cd-rom2 that I will try washing. Even after cap changes, I'm still getting some crackles and rattles in the sound department.

[DeshDildo]

the dish machine didnt do it
wet board didnt either
please refer to the TG16 repair manual for the pin functions
use a logic probe at the 6280 and look for the pulses from the controller

I don't have a logic probe but I suppose I should grab one.  There are times in the future where I am sure it will be handy. 

Besides testing continuity in the original circuits to HuC6280 (all were fine) and testing continuity through the jumpers I installed (also fine), I did check voltage on the circuit.  I looked up the HuC6280 pin out and checked each controller input.  At idle I measured 5VDC on each pin (This was using ground at multiple places on the board).  When a button is pressed it goes to 0VDC because the button press is basically sending input power to ground (I think this is how it works.  This is how NES controllers work).  I'm still trying to grasp "high" and "low" inputs but I surmise that one press sends all power to ground and the opposite function (on the same input line to HuC6280) only sends some power to ground.  I assume this is the difference in high and low but I really don't know I'm just guessing.  You electrical engineers and software folks would know the answer here.  Anyways, pin 3 of the controller port handles input functions of D-pad right and button II.  This circuit goes to pin 24 of HuC6280.  When I measure power directly from pin 24 of HuC6280 to any ground on the PCB this is the result:
Constant 5VDC when no button is pressed
0VDC when button II is pressed
4.8VDC when D-pad right is pressed

Is this the same thing a logic probe is testing for just on a more accurate scale to pickup pulses and whatnot?

[OldMan]

pin 3 of the controller port handles input functions of D-pad right and button II....

Constant 5VDC when no button is pressed
0VDC when button II is pressed
4.8VDC when D-pad right is pressed

The pad and the buttons go through a chip in the controller that switches between them.
The 4 directionals are read, then the controller lines are changed, and then the buttons are read.
It's not a matter of partial voltage being a different button. That's why you should use a logic probe - it will light on a high (~5v) and be off on a low (~0V).

Check the voltages inside the controller. It sounds like either the conductive pad isn't making good contact, or the chip in the controller died.

...and try another pad, if you can.

[thesteve]

The system feeds 2 pins pulses
1 pulse enables the controller
The other pulse switches between d-pad and buttons
The 4 inputs back to the 6280 will receive pulses from any input button synced to the select pulse from the system

[DeshDildo]

I'm back at it with my new (old) Micronta logic probe.  I seek the advice of you almighty smart fix it guys.

I tested pulse and there is a constant pulse on pin 31 () at all times.  I measured pulses at each pin with the corresponding button presses.  My circuit in question for D-pad R and button II goes to pin 24 (K1) of HuC6280.  When I press either of these I do get a good pulse on pin 24. 

This seems to tell me that all circuits are good and the controller is doing it's job but, there may be an internal failure in the 6280?  Is this correct thinking or is there more to the story/ more tests to be done?

[thesteve]

that sounds like a chip fault to me
all other buttons/directions working?

[DeshDildo]

Yes sir.  Everything else works fine.

[thesteve]

with all else ruled out its the chip

[NightWolve]

Bummer.

Dishwasher + 6280 chip did not mix well...

[BlueBMW]

Seems like Ive seen a lot of huc chips fail here recently.  Im not convinced the dishwasher did it.  Its possible it was on its way to failure already.

[DeshDildo]

Well, enough time has passed and the rage I had with this duo has subsided.  Tonight I decided to pull the donor CPU off of a parts TG16 that BlueBMW was extremely kind to send to me (seriously dude I owe you big time).  I am trying to make time to install it in the finicky Duo tomorrow.  I have to say it is damn tough to desolder one of these guys without bending the pins all to hell.  I think I did an OK job but I am already worried about them lining back up with the pads.  I'm sure I'll have plenty of time in tweeking them to line up. 

For future reference what is your guys' method on pulling one of these things?

[SephirothTNH]

I've never removed a 62x0 chip from a nec system but I have removed snes PPUs before which are the exact same size with 20 more pins squeezed in.  I used a product called Chip Quik and got them off without bending any pins.  At least they weren't bent enough for me to notice or make transplanting them to another system difficult. 

I hear a lot of people use hot air to do it though.  If I knew I had to do it a lot I would probably try hot air but for a few one offs I'm more than happy with Chip Quik.

[thesteve]

halogen lamp

[DeshDildo]

halogen lamp

Being a DIYer who learns as I go I have never heard of this.  What's the process?  I assume you have a specialized flexible stand of some sort with a powerful bulb.  Do you flux the pins?  Can you usually work the chip up at one time or use something to gently lift pins while under the light?

[esteban]

halogen lamp

Being a DIYer who learns as I go I have never heard of this.  What's the process?  I assume you have a specialized flexible stand of some sort with a powerful bulb.  Do you flux the pins?  Can you usually work the chip up at one time or use something to gently lift pins while under the light?

Place halogen bulb directly on PCB. Wash your hair in the sink. Towel dry. Return to work bench. Sweep up ashes, deposit in trash bin.

[thesteve]

Using 2 50W lamps with foil covered reflectors
1 above the chip about 3/4 in
The other below board about 1.5 in
Wait for solder to melt
Remove chip

[DeshDildo]

The brain transplant has been competed.  I have ALL controller functions back!  This at least confirms the old 6280 was indeed bad.  I do have some messed up colors and no sound though so I just need to double check some connections but there are signs of life.

[BlueBMW]

Awesome!

[DeshDildo]

Ugh!  So close I can taste it.  I spent some more time with it and got everything working great.  The only thing giving me trouble is I have very little left side audio.

[DeshDildo]

OK... it seems like fixing one problem breads another on this infernal machine.  I've double checked the left audio circuit and note I have a weak signal but, can't figure out why.  I double checked my connection at pin 17 of Hu6280 and that's all good.  I used a small home made probe to test the audio sound at different points in the circuit going to op amp 4558.  Left audio input is pin 2 of 4558.  Using this probe on pin 6 of 4558 (right audio in) the audio is nice and loud.  Check left audio in at pin 2 it is pretty faint.  I have this faint audio all along the left audio circuit.  I tried to bypass all of the circuit and vias by running a wire directly from pin 17 of 6280 to C878 which is a small smd cap.  This yielded no change.

It seems to me the circuit itself is fine but where is runs through all of the small smd stuff must slightly boost the signal before going into the op amp.  My only guess at this point is one of the small smd components has failed.  If the op amp failed I suspect I would have low or no sound on left AND right.  Also, I noted that the left channel has much lower sound than the right going INTO the op amp. 

The rage is beginning to build again.  Any suggestions?

[thesteve]

sounds like the typical audio issues
if both sides are good off the CPU move on
check cd sound

[DeshDildo]

sounds like the typical audio issues
if both sides are good off the CPU move on
check cd sound

Will this affect sound even when testing with a Hu and no CD games.  I

[thesteve]

Sometimes

[NightWolve]

I've never removed a 62x0 chip from a nec system but I have removed snes PPUs before which are the exact same size with 20 more pins squeezed in.  I used a product called Chip Quik and got them off without bending any pins.  At least they weren't bent enough for me to notice or make transplanting them to another system difficult.

That's pretty hardcore, dude... It looks way WAY too difficult for the PPU or CPU on a SNES... I'd replace the whole unit if I thought that was the problem. I will soon finally attempt replacement of one of the SRAM chips on my SNES... It was damaged as the 7805 voltage regulator was dying and sent high voltage to the motherboard. The SRAM pins are far enough spaced and it seems workable. I was thinking of using Steve's method with a halogen lamp to desolder one off of the Sony Sound module that BlueBMW sent me. I don't trust the chips on the motherboards he sent me, but it looks like the same type of 32K x 8 bit SRAM chip is used in the sound module, so that's probably the best place to take one from.

And Desh's saga with his Duo continues... Good looking out once again, BlueBMW! He already got somewhere at least... Just noticed I was a little late to the thread, so did Desh finish it off ?? All working 100% ?

[DeshDildo]

Nightwolf, the saga continues.  I had it working by swapping the CPU.  I still had an issue with sound on the left channel.  While continuing to diagnose I developed a direct short somewhere that fries fuses on power up.  In the mean time I purchased a slightly beat up DUO from KC and just swapped my nicer shell to it. 

I have been waiting to for a part (from a furnace believe it or not) to fall into my lap.  While diagnosing the short I just soldered in a cartridge fuse receptacle to make changing fuses easy during this process.  Instead of going through fuses, some older Armstrong furnaces had a 2 amp breaker wired into their 24v control signal.  I've been praying that we would be replacing one of those units at work so I could rob the breaker off of it to further my DUO diagnosis.  I haven't even taken the time to search but maybe I can source some sort of 1-2 amp breaker by itself for cheap.

Once I get this thing fully operational I will have no use for a second DUO so I'll probably do a raffle or something with it.