Barco S1/S2 Touchpanel Failures (shorted capacitors on input power)

[Steve Guttag]

I've just started to notice a trend on Barco S1/S2 (particularly S2...the ones with the larger plastic outer shell) are failing. They won't boot up. I went from 0 to 3 within a month (over 2 sites). They're old and I figured on the first site that perhaps that is just their lifespan. In some respects, the longevity of this stuff is undetermined until now.

But seeing one fail at another site, with similar vintage, I decided to take it back to the shop. What I found was disturbing but not too uncommon in the electronics world. The power rail (12VDC) was shorted. I'm amazed that the projector can deal with that so well. Note, if you plug one of these panels in while the projector is powered, up, the IMB/IMS will reboot as the 12VDC power rail will momentarily drop.

The problem was a shorted SMT capacitor and there are about 9 of them across the input power rail as it feeds different parts of the board. There is one main electrolytic (it was fine) and then they seemed to place the little SMTs in groups of three in different parts of the board.

So, if you do have a failed Barco touchpanel...DO NOT LEAVE IT PLUGGED IN!

[Bruce Cloutier]

So you pulled the SMT caps until there was no longer a short? Sadly they don't mark those to know if they were of the proper voltage. What size? 0603? 0805? What type? Just curious. I have product out and operational from 2004.

Those Series 3 JNIOR have dead batteries. The original design soldered those in. I know, dumb. There is a small amount of Flash dedicated to file space. While there is enough for files the components suffer a lot of wear (depending on the application). We have had those components fail. There is only one Flash chip and that carries the OS. That leads to boot issues. But those are the known failure causes.

But... we haven't had the caps go bad to our knowledge.

I have a telescope that I want to bring out of storage. There is a list of caps that they say that I will need to replace.

How old are these touchpanels?

[Steve Guttag]

I'm not quite that random. When I realized I had a short, I wish I had a thermal camera handy (they normally will point you right to the "hot spot" that is causing grief. I use a jiffy test (incandescent based test er with two AA batteries) to home in on it. Using a jiffy instead of a DMM will find noisy switches and these sorts of problems better as it will put more current through. Then, visually, one was more discolored (darker). Dabbed it with some flux...popped it off and ta-da, no more ring-around-the-collar.

I don't have a great feel for SMT sizes. It seemed small to me so I'm guessing 0603. As for Touchpanel vintage...2012-2013 is my guess.

I'm noticing that CP750s can also get shorted power rails too...so again, having an IR camera with a suitable power supply can be a handy aid in seeing who is getting unusually hot.

Poor-man's thermal camera...IPA and swab. See who evaporates fast.

[Bruce Cloutier]

Interesting. I can add the (not so wise) finger sensor test to your list. I also have a IR temperature pen-light thingy (RadioShack) that I can run around to locate hot spots.

There is probably a solidstate fuse upstream protecting the system.

I can understand why old electrolytic caps (and similar) fail. A ceramic cap is another matter. There was likely an ESD event. Someone got a real charge out of life and then gave it to the touchpanel. We're in weather where the RH is very low. People don't think humidifiers are important.

I hang TVS diodes (Zener things) on the rails to limit the voltage and handle those events.

[Steve Guttag]

If you ever hang on YouTube...check out Louis Rossmann's page (you'll have to go back years now since he is mostly on right-to-repair)...he has 100s of videos on fixing MacBooks...it is doubtful that someone gave it an electrical jolt unless it came through the keyboard. But shorted SMT ceramics are more common than you would guess.

As for the the projector and a fuse. When I was troubleshooting, in the field to see if it was the touchpanel or the cable, I swapped between theatres to see what moved with the problem. So, whatever fuse it might have, it is a self-resetting one. I did notice that the IMB in the theatre rebooted itself during the TP swap...I figured the IMB didn't like the blip in power. I didn't know it was presenting as a short, at the time.

Dolby has been using resettable fuses since the CP650 and the CP750 definitely has them too. They do get hot like an PTC. Perhaps Barco has something like that on the various fees off of the supply.

[Jason Raftery]

But seeing one fail at another site, with similar vintage, I decided to take it back to the shop. What I found was disturbing but not too uncommon in the electronics world. The power rail (12VDC) was shorted. I'm amazed that the projector can deal with that so well. Note, if you plug one of these panels in while the projector is powered, up, the IMB/IMS will reboot as the 12VDC power rail will momentarily drop.

I'm starting to see this same mode of failure though I'm unclear on what's causing the caps to short. The one I have in front of me now from 2016 will actually trigger an abrupt board reset on both the IMB and ICP in a DP2K-23B if connected while the projector is powered up.

I swapped the E1907DP80D3RT motherboard with one from a donor unit and the touchpanel works fine with the same Z3499792K cable, so it doesn't appear as though a faulty cable blew the board. I'm used to seeing CF cards go corrupt (or folks drop a touchpanel and crack the display) but the motherboards starting to fail is unfortunate.

[Steve Guttag]

I suggest getting a thermal camera and an adjustable DC power supply to locate the faulty capacitor. If you do some probing with a jiffy test (uses a 3V light bulb and a couple of AAA batteries) or a DMM to find where the 12VDC power goes on the board, you'll see a couple of spots where they put three ceramic SMTs in parallel. The one that is getting hot is your problem. If you are on a budget, use some IPA on the suspect capacitors and see who evaporates it first. On the one I worked on, visually, you could see that it was darker than its neighbors.

Unfortunately, I didn't get its value (and just left that component off the board) but I'm guessing it is on the order of .01µF. It is a very tiny one.