Maker Pro
Maker Pro

n-mosfet replacement on Qnix lcd monitor

Is it possible (or even probable) that if the drain tab isn't soldered on 100% right (I can't see how it wouldn't be, but there appears to be a micro gap about a fraction of a millimeter big between the bottom of the part and the pad on the PCB) that would cause it to go?
Yes, unfortunately that is possible. Soldering by hand is never going to be as effective as the reflow method used in the original assembly, which would have eliminated any gap. It's also possible that a faulty component elsewhere, or a design error, is causing the FET to pass excessive current.
Is it possible as well that the powerbrick they are supplying is too much for the board to handle when using NA designated electrical outlet?
Perhaps. But less likely; especially if the original board had quite a long life before failing.
It looks as though you have a fair bit of space around the FET. If it were me, I'd fashion a small finned heatsink from copper sheet, solder the FET drain pad to that, carefully bend up the gate and source pins if necessary, then solder the heatsink to the pcb pad and make the connections for the gate and source.
 
  • Like
Reactions: AMT
Thanks for the help. Is it also possible that the FET I used as a replacement is not spec'd well enough for the circuit?
I read something about high gate capacitance being an issue, but the monitor did work after the replacement until I pulled the plug to stop it from smoking, so would gate capacitance in this case be an issue?
I put in an order for some that have a slightly lower rDs ON (100 mOhms compared to 130) than the original, but with only a marginally higher gate capacitance of 18nC @ 5V whereas the original is at 16.
The replacement I just burned had a gate capacitance of 85nC @ 10V so would that be the issue?

I really have no idea at all, I am just using what I have found on other forums as well as I can in order to communicate in what is largely a foreign language (electrical) to me.

Here are the 3 different FETS I am considering and mentioned above. The first is the one that just burned out. The second is the one I just ordered and will see in a few days.

http://www.digikey.co.uk/product-detail/en/STD70N10F4/497-8806-1-ND/2043668
http://www.digikey.com/product-detail/en/FQD19N10LTM/FQD19N10LTMCT-ND/3042602
http://www.digikey.com/product-detail/en/AOD482/785-1224-1-ND/2353942
 
so would gate capacitance in this case be an issue?
It's impossible to say without knowing the conditions (e.g. frequency, driver impedance) under which the gate is being switched. Looking through various MOSFET datasheets it seems to me that, in general, the lower the gate capacitance/charge the higher the Rds(on) resistance. Which of those parameters is key for your particular application can hopefully be determined when you try the ordered FETs in circuit. Good luck. Keep us posted.
 
Mission Successful!

I used a heatgun to remove the burnt FET while isolating the other components with tinfoil so I could concentrate my heat in one area, after heating up the entire board up some to avoid cracking, I focused on the FET and had it off in about 15 seconds.

I added a small bit of flux and solder to the new FET and laid it on the board and then just used the heatgun to get it up to temp and it just popped right into place.

Been running it now for hours without issue, and as a happy side effect the power-brick no longer heats up to insanely hot temps like before.

Thanks for all the help you guys! :)
 
I ended up going with AMT's second option since that was the last thing they mentioned trying, MOSFET N-CH 100V 15.6A DPAK. So far this has worked, if it fails in a short time I will update this thread.

My board was pretty badly burnt from the previous mosfet melting down and this lifted most of the copper around the drain tab.
JgHZnli.jpg


I used a multi-meter to test the surrounding grid and most of it was connected so I used a copper plate to reach out to the pads that were still available. This should also help deal with heat.
U6VZQ4t.jpg

I had to scuff the plate up a bit to get the solder to stick and it still doesn't look pretty but it got the job done.



My Board(this will probably work with any similar board):
4amanX6.jpg
 
Imma year late to this party, but KUDOS to ur success, AMT & jawz. I too am now "enjoying" this issue with the QNIX monitor FET. Mine went bad in Feb of this year after a good 3 years of 27" 2k love. I found the same idea/option in another board to use ur choice of FET, also from DigiKey, sadly I've had to wait until they had it back in stock for about 2 months. Admittedly, the ur experience yields a bit of fear, but I'm at the point of no return: the DPL-2510 board is missing from the internet, and I'm in no financial way to buy another monitor, though, for the price I paid, I could get a 32" 144hz name brand (LG). I'm pulling the trigger in the morn. I've always had a bad experience at soldering, it seems a heat gun would work better for reflowing solder in the case, I'll pick one up from Harbour Chin.. er Freight as well. Wish me luck!
 
Imma year late to this party, but KUDOS to ur success, AMT & jawz. I too am now "enjoying" this issue with the QNIX monitor FET. Mine went bad in Feb of this year after a good 3 years of 27" 2k love. I found the same idea/option in another board to use ur choice of FET, also from DigiKey, sadly I've had to wait until they had it back in stock for about 2 months. Admittedly, the ur experience yields a bit of fear, but I'm at the point of no return: the DPL-2510 board is missing from the internet, and I'm in no financial way to buy another monitor, though, for the price I paid, I could get a 32" 144hz name brand (LG). I'm pulling the trigger in the morn. I've always had a bad experience at soldering, it seems a heat gun would work better for reflowing solder in the case, I'll pick one up from Harbour Chin.. er Freight as well. Wish me luck!
So I FINALLY got around to making repairs. I was blessed by Digi Key wh0 sent me two FETs due to their shipping error. But my soldering ex0erience in the SMD arena was sadly lacking. The Chip Quick approach was a joke. Id try to make the repair and get frustrated cause it wouldn't work right and put the monitor away... try it again in a few weeks or months. Well, its been a year., and found a dual soldering way of removing the FET, worked like a champ, and I cleaned to board and resoldered the FET.
Plugged in the monitor and the power lights worked! Ok, plugged into my video card and was greeted with the same high pitched whine that happened when it first died, The panel came on and displayed melted dark screen but it worked... for about 2 minutes then my room smelled of burnt hair again. Oh well. Tells me something else is wrong. What should I look for?
Mind u, I'm just a hobbyist, but I'm smart and resourceful and can navigate around electronics. But I'm not about to get a degree in EE just to figure this out. Also, I'm cheap and it annoys me to no end to know that the monitor is prolly and easy repair.. if I knew what was wrong.
Anyone got ideas or what to look for?
 
Top