SOLVED Unmarked Component Identification Help

[ajp8868]

Hi All,

Trying to fix an LED driver board with an unmarked component which I think has failed and is causing a short. I can't identify it and not sure what to Google.. my guess is a diode of some description?

In the picture it's the component to the left, paired with a MBR10100A which is a diode that is working fine. There is a tiny marking in the circle with L then underneath 3M (there are 4 identical components with different last letters)

Any help would be appreciated!

Thanks

[flolic]

Unknown mosfet.

[Per Hansson]

There is a loose leg on the inductor, did you notice that?

[ajp8868]

Per Hansson, I did not! Thanks for pointing that out.

Ah thanks Flolic, it's definitely shorted in that case. Will any N Channel MOSFET be ok to replace it as long as it's rated over the voltage? I.e https://uk.rs-online.com/mobile/p/mosfets/6711024/

[wraper]

There is a loose leg on the inductor, did you notice that?

Does not look loose to me. Soldering might be better, but still there is solder within the hole and on the other side.

[wraper]

Per Hansson, I did not! Thanks for pointing that out.

Ah thanks Flolic, it's definitely shorted in that case. Will any N Channel MOSFET be ok to replace it as long as it's rated over the voltage? I.e https://uk.rs-online.com/mobile/p/mosfets/6711024/

Did you forget about current? What is input voltage/current and total number of channels, output/voltage current? Considering diode used, you probably should use MOSFET rated for 20A, not 5.8A.

[ajp8868]

I did completely.. I'll remove one of the others later and see what I can measure to work it out.

Thanks for the help all, will update when I get a replacement.

[wraper]

I did completely.. I'll remove one of the others later and see what I can measure to work it out.

Thanks for the help all, will update when I get a replacement.

The best thing you could measure is RDS(on). Out of that you could figure out how beefy replacement you need.

[Syntax Error]

Your LED board component reminded me of a similar defective DPAK device, which was a constant current regulator for LED drivers, the NSI45090.
 
Similiar part: https://uk.farnell.com/on-semiconductor/nsi45090jdt4g/led-driver-to-252/dp/2535065

[wraper]

Your LED board component reminded me of a similar defective DPAK device, which was a constant current regulator for LED drivers, the NSI45090.
 
Similiar part: https://uk.farnell.com/on-semiconductor/nsi45090jdt4g/led-driver-to-252/dp/2535065

Certainly not that. It's a MOSFET in buck converter.

[Syntax Error]

Certainly not that. It's a MOSFET in buck converter.

You're right. Ignore the above.

[ajp8868]

Well I tried to measure resistance between source and drain while the led was operating but it didn't seem to measure anything at any point as if they're not connected.. is this normal?

[wraper]

Well I tried to measure resistance between source and drain while the led was operating but it didn't seem to measure anything at any point as if they're not connected.. is this normal?

How do you expect to measure resistance in buck converter while MOSFET is constantly switching?

[ajp8868]

I was doing it wrong, should have been measuring voltage drop.. but I've found a MOSFET with the same gate voltage threshold that is being supplied and a large current/voltage.

It's this one, I would guess it should work? https://uk.farnell.com/on-semiconductor/fqb19n20ltm/mosfet-n-ch-200v-21a-to-263ab/dp/2453431

[wraper]

I was doing it wrong, should have been measuring voltage drop.. but I've found a MOSFET with the same gate voltage threshold that is being supplied and a large current/voltage.

It's this one, I would guess it should work? https://uk.farnell.com/on-semiconductor/fqb19n20ltm/mosfet-n-ch-200v-21a-to-263ab/dp/2453431

Why would you use 200V mosfet with awful RDS(on)? If you go to high voltage, other specs suffer. You still did not say anything about the question about voltage and current (power) ratings I asked.

[ajp8868]

Ah yeah sorry I missed the other questions, it looks like it's around 12v which goes to the led first, then back through to the coil then to the MOSFET tab. The LED is this https://uk.rs-online.com/mobile/p/leds/9154705/

So the channel that has blown is the blue, which should only be around 4.5a according to Farnell's specs.

Does that answer the questions?

[ajp8868]

So in this case, is it a P Channel MOSFET?

Edit: Realised I was being stupid... So N channel MOSFET that will handle 12v at least 5a and with a 2v threshold is required. Is that correct? I'm not sure of any other factors that should be considered

[wraper]

Dunno where you got that 2V threshold figure. So input voltage is 12V? Is there single LED connected, or multiple in series? Does it have any sticker with specs?

[ajp8868]

When the LED is on, 2.5v is being applied to the gate. But there seems to be a lot of MOSFETs that start at 2v

It's a single LED, max current draw is 4.5a as stated on the attached spec sheet (page 10)

[fzabkar]

What are the marking on U1B?

I'm intrigued by the 3 dissimilar parallel current sense (?) resistors (R100, R150, R200). The equivalent resistance would be 46 mOhm. What's going on here??

[ajp8868]

I was wondering myself why there's 3 rather than one higher value, that chip is marked SQ9910B and is an LED driver. I've attached the datasheet.

The gate pin goes to the gate of the MOSFET, and the cs pin must be detecting if it's actually turned on as it's connected to the source pin?

[ajp8868]

So, I've been reading more and more into this and, if I understand correctly, the MOSFET is limiting the current to the LED based on the voltage applied to the gate.

Therefore I will need to replace it with one that limits the same amount with the same gate voltage applied otherwise, I risk blowing the LED.. am I correct in thinking this way?

If this IS how it's working, how do I work out how much current is being limited when the LED is on?

[wraper]

So, I've been reading more and more into this and, if I understand correctly, the MOSFET is limiting the current to the LED based on the voltage applied to the gate.

Heck no  . It's basically a buck converter as I already said. https://en.wikipedia.org/wiki/Buck_converter
The difference from usual buck converters is that switch is located on the low side and there is no output cap. Have you ever wondered why there is a big inductor? Also you provided controller datasheet but apparently never bothered to read even a first page of it. Now i suspect you "measured" so called "Vgs(th)" by simply measuring voltage on the gate by multimeter in DC mode.
Again you simply ignored my question about specifications (ratings) of this LED driver. At least simply write it's model, I will google myself.

[ajp8868]

Sorry I'm clearly mis-undertood the buck converter.

I did read the datasheet of the controller to work out what pins etc are what, and I did not know what specs you were after.. I've linked both the LED datasheet and the driver sheet.

  I missed the fact it says 10v MOSFET drive...

Yes I did measure the gate voltage being applied that way. I assumed that was how much MOSFET required to turn on.

The driver board itself hasn't got a model number, it's a custom board out of an Equinox Shard DJ Light.

What ratings of the driver do you mean?

[wraper]

I meant like input voltage, power rating. So to confirm, there is only a single led on the output, board is powered from 12V, and there are 8 channels/LED in total 4 channels / 1 led with 4 colors in total?