REPAIRED: Barco/Folsom ImagePro IP-2003SDI (first post!)

[clickykbd]

Hello all!

UPDATE: REPAIRED  , see post:
https://www.eevblog.com/forum/repair/repair-barcofolsom-imagepro-ip-2003sdi-(first-post!)/msg1228926/#msg1228926

Long time viewer, first time poster from Austin, Texas!  Thanks Dave for the amazing educational resources.

Short Version:

Help me identify a SMD part (service manual and/or schematic were not forthcoming in my searching), and maybe a "proper" tool for BNC terminal tightening (something that can tighten the nut despite the leads being in the way?

https://www.dropbox.com/sh/pnci1n08gicerbb/AAC-deHtmfQ74m0as6LPz9Tfa?dl=0&preview=ip2003_bad_part.JPG
and
https://www.dropbox.com/sh/pnci1n08gicerbb/AAC-deHtmfQ74m0as6LPz9Tfa?dl=0&preview=ip2003_bnc_structural.JPG

Long Version:

This is my first "real" repair attempt.  I've repaired a handful of electronics (turntables, DSLR, iPhones, iPads, Arduino dabbling, some kit assembly), but never attempted a true surface mount rework of any kind.  I scored this ImagePro in it's non working state off eBay for 70$.  It does power on but none of the logic seems to initiate and no response from the display/buttons.  My initial hope was it was just the front-panel daughter boards and I could still control it via RS232 or IP, but no luck. Some of the BNCs had been twisted to the point of breaking their solder to the posts, I repaired those already.  I also found evidence of prior repair attempts... a stray screw floating around in the chassis, rut-roh.

Have the manual, http://www.reflechi-son.fr/paris/manuel-utilisateur/Folsom-Image-Pro-instal-manual.pdf
... but not able to find a schematic or service guide (though I do have a callback and ticket from Barco pending, maybe they will toss me some docs).

The main board is littered with test points and obvious ground posts.  Many of them were marked with voltages, so I proceeded to test those... and lo, one of them (TP115 in photos, +3.3v reading at +1.5v) pointed me to find that part with the lid blown off, near the power intake section of the board.   I don't know much about power circuit/protection design but my hunch is that it is identical to it's buddy right next door, which without more knowledge is my only clue on identifying it.   Still not sure, not much personal experience searching for datasheets and interpreting layouts, my initial googling led me to believe it might be an N-Channel Mosfet?

Looking for some kind words and votes of confidence before I order a part.  Exactly what i'm looking for would be ideal?

Here are some more gear porn and photos under the hood, even though it is a generation or two old these universal video format converters folsom/barco made are still an industry standard.

https://www.dropbox.com/sh/pnci1n08gicerbb/AAC-deHtmfQ74m0as6LPz9Tfa?dl=0&preview=ip2003_lifting_the_skirt.JPG

https://www.dropbox.com/sh/pnci1n08gicerbb/AAC-deHtmfQ74m0as6LPz9Tfa?dl=0&preview=ip2003_in_like_flynn.JPG

https://www.dropbox.com/sh/pnci1n08gicerbb/AAC-deHtmfQ74m0as6LPz9Tfa?dl=0&preview=ip2003_problem_area.JPG

https://www.dropbox.com/sh/pnci1n08gicerbb/AAC-deHtmfQ74m0as6LPz9Tfa?dl=0&preview=ip2003_closer_in.JPG

With luck, maybe that is the only bad part and things down the line did not get fried/corrupted.  If the 3.3v rail is toast it certainly would make sense (to this relative novice) that none of the 3.3v logic fired up.  If it works out I can even upgrade it to the HD/DVI model (if I can find those upgrade mezzanine boards out there).

Thanks in advance!
Best from Austin,
Ryan

[poot36]

Check the resistance between the 3.3V test point and ground first.  If it is really low (usually under 5 ohms or so) then there is a most likely a short on the 3.3V rail.  Check the resistance between the drain and source pins on both Q1 and Q2.  If they read open then they are most likely ok.  If not they are dead.  You can also check the Q1 and Q2 with the diode mode on your meter and that will tell you if the body diode is ok or not.  If Q1 or Q2 is shorted remove them and test that there is voltage coming to the gate pin of both of them.  <- this test may not work due to the circuit not seeing any voltage on the output.  If the 3.3V rail is not shorted try powering it up with the 3.3V rail from a PC power supply and see what happens.  By the way all of my research on the somewhat legible markings of Q2 indicate that it is a transistor not a MOSFET but I could be wrong.  I am assuming the marking are First line: 4412 this is most likely a date code Second line: ST logo AL triangle logo Third line: L42C.  I have been looking at the various SMD code books online for this info.

[amc184]

What a beast!  It looks like a lot of design work went into this.

The component that's damaged, Q1, is a Vishay Siliconix SI4412 30V 8A N-channel MOSFET.  The top row is the abbreviated part number, the centre has the Siliconix logo on the left and the Vishay triangle logo on the right.  The bottom row is most likely a batch code.  You're lucky there was a pair of them like this, having an undamaged one made the identification much easier.

[clickykbd]

Thanks amc184 and poot36 for the pointers and part information.  I'll have some more time to poke at it after/during the holidays, will certainly have updates then.

Still seeking some advice on proper tool for re-tightening those bnc nuts around the leads.

Cheers!
Ryan

[cs.dk]

Can you get a crawfoot wrench in there?
They look like this;

[clickykbd]

Can you get a crawfoot wrench in there?
They look like this;

I was able to find some sockets that are like I was imagining might be required for the BNC jam nuts in this tight spot.  There are some 14mm flare nut sockets out there (usually for spark plugs or fuel lines it seems), the thinnest walled one I can seem to find is this, but still not sure it would fit, those shielding houses are really close to the nut.   The cutout might allow me enough rotation to check tension on the bottom row closest to the board and behind the posts from the upper row, of course the ratchet may not allow for alignment that close to the board.



http://www.lasertools.co.uk/item.aspx?code=4949

I'll leave that side of the project later if I can get it running again.

Turns out there are specific jam nut sockets too, but most of them seem to be designed for when you can get at it before you have it soldered in place.

[clickykbd]

Where does the time go!  I posted this repair project 2 years ago and JUST NOW got around to attempting the SMD rework replacing that mosfet.

For a first time doing surface work, on a pretty tiny part (SOIC-Narrow), it went smoothly!  I had some chip-quick I picked up a long time ago for this, but another pending project promoted the purchase of a hot-air rework station (Quick 9570w), which did the business.

For a newbie, my only hesitation was orientation-polarity of the mosfet.  I don't know what mark on the part typically indicates polarity.  But because it had a non-burnt buddy next to it, I just elected to put it on with the label oriented the same direction.

Did it all with a shitty helping hands magnifier, got nothing else for inspection yet!

But voila!  The face panel/input board (the former problem) on the 3.3v rail powered right up and everything is working!  Even my inexperienced repair to a damaged BNC post seemed to pass mustard.

I just wanted to post that it worked out and it felt great!  I owe most of my inspiration for the urge to tinker at this level to Dave (EEV blog), Mike (Mike's Electric Stuff), and Lady Ada (Adafruit).  Even my ability to reasonably navigate Digikey and Mouser I owe to the passive experience of watching everyone's videos.  Cheers!

PS - I fixed my broken dropbox links in the OP.