We use cookies to provide our visitors with an optimal site experience. View our privacy notice and cookie notice to learn more about how we use cookies and how to manage your settings. By proceeding on our website you consent to the use of cookies.
Good morning you clever people, perhaps you can help. I have a small PCB from a Prosthetic hand that is faulty. Initially I noticed that there was a misplaced/broken capacitor on the board possibly during manufacture and I sent it to a company that a friend recommended. They replaced the SMD capacitor as requested and I had hoped this would correct the operation however this hasn’t worked. They couldn’t test that board as it needs to be fully assembled to work with the motors, Bluetooth, sensors and so on. Now I may be clutching at straws but I have noticed that these components marked are getting very hot when powered (supply voltage to the board is 7.4 volts and draws about 400 - 500 milliamps) i’m pretty sure that’s way too much. I have identified one component which as far as i can tell is a PIC and is most likely flashed with custom instruction/programming. Not a great deal I can do about that however I thought that perhaps it may be another component that is causing the issue. I’m a prosthetic technician and know a tiny bit about electronics but this is way out my league. I have googled though and think that some of the components are inductor, transistor (possibly a Toshiba but then hsa a q code on the board for Triode?) the PIC. The other chip has no markings at all so no idea. The inductor and the chip get very, very hot (inductor very very very hot) the PIC and the transistor less so. I totally get that the issue may be elsewhere and this is the result. Fully prepared to be ripped apart for my lack of knowledge, will answer what I can.
I would like to try repair if possible (will buy a rework station if you think it might be worth it may buy one anyway). The manufacturer will be of no help and certainly wont supply a new board. apologies for the poor image quality (cheap microscope the wife bought me). Have multimeter and bench PSU but thats it. Sorry for the long winded post and thanks for getting this far… Hope you can help.
That’s a bad sign. Under normal circumstances a microcontroller of the DSPIC sort may warm slightly to the touch; anything beyond that is typically an indication that something has gone wrong to the extent of breaking things.
As mentioned, the microcontroller would have carried custom firmware. If its code-locking mechanisms were used it would be tough (by design) to get at that, but if the chip has been physically compromised in addition you’d need some extremely advanced forensic techniques to get at it, assuming that it hasn’t been burnt away by the malfunction.
Long story short, one’s chances of successfully repairing such a device with minimal equipment and limited background knowledge are…not good… With no offense intended, it’s probably fair to compare the task to attempting a hip replacement using a dollar-store exacto knife and a bottle of peppermint schnapps.
I know there is nothing I can to about the PIC and i have no doubt its instructions are secured, I was hoping that it might be some of the lesser components. The pic does get quite warm the yellow and green component get very, very hot. The red slightly less so. The prosthetic hand is a very expensive bit of kit and would have been great to have got it running. I was given it by the manufacturer as non running for display and they may have just written it off as not worth repairing but thats just such a waste.
If its not repairable then fair enough… thought I’d ask, also you would be surprised at what i can repair with an exacto dont like schnapps though…
The analogy is rather extensible though… Just as disassembling cadavers is part of an orthopedic surgeon’s training, so also can reverse-engineering a busted piece of kit be useful training in the electronics space if one’s expectations are properly calibrated.
The green/yellow/red seem likely to part of a DC-DC conversion stage. If one’s putting 7.4v to a board stuffed with chips that go poof north of 4, such a thing would be needful… One cookie says that if you get a part number off the green device, it’s datasheet will have an example circuit that bears a close resemblance to what you’ll find on that area of the board.
Tracing connections can be tricky if a board has more than 2 layers, and any parts left on it will confuse the issue.
This is helpful. There isn’t a number on this chip that I can see. Laserd off perhaps? Wonder if it is the same chip as the other one further to the left. Might be worth a look.
At this point there is nothing to lose and yes It is a multi layer board unfortunately.
Flux residue/conformal coatings/random schmoo can often make markings difficult to read. A light scrub with an isopropyl alcohol swab or something similar will often help reveal them in such cases, and it’s often necessary to play a bit with one’s angle of view and lighting in order to get decent visibility.
Different pin count and no adjacent inductor or transistor next to U2, so not likely the same.
I got some images of the chip with my cheapo microscope changing light directions and generally messing around with the angles. I have a partial number. Very very hard to see. Going to put the images into photoshop to see if I can enhance or bring out the detail. From the partial number I do have it looks as though it’s a Texas Instruments component. Perhaps this… hopefully if I can get the rest of the numbers and letters to show up in photoshop.
TPS62111MRSAREP step down converter. Values look to be in range hopefully know more later
Hi @william.mackinnon ,
One thing you could try is to remove the switcher chip and feed an inductor’s pin (whichever OK) from a 3.3V power supply to see who is left heating up…
I believe what Heke is suggesting is to remove the TPS62111RSAT. So instead of the inductor getting the 3.3 volts , 1.5a from the chip, you would use a power supply to feed that output into the inductor. Then trying to see what or if it is still heating up. Though I would be careful as this might not be easy to remove.
@rick_1976@heke@Verna_1353
Sounds like an idea. I wondered if the voltages in the component you suggested would be correct as board supply is 17 volts. Hence why I thought it might be looking at the
TPS62111MRSAREP.
I think I will order a rework station. Removing the chip might make it easier to see and then Try as suggested.
The 303 0 component is also hard to pin down would this be a digital FET? It’s about a close as I got.
Yes, that is why there is that TPS62111. It will chop the 17V down to 3.3V for the MCU and friends.
You should certainly obtain a rework station. IMHO every household should have one.
Hi @william.mackinnon ,
If there is space available, you could replace the TPS chip with a 3.3V module. Then you would not need a rework station. Just a soldering iron to solder short wires from the module to the +17V input, the +3.3V output and ground.
dont like schnapps though…
cheers