Buffer Board IC replacement.

[Glitcher]

I found a defective buffer board using the beep test with my multimeter. The shorts occur around the circled areas on the buffer board. But I can't figure out which chip or chip(s) are bad and which I need to replace? Can anyone help.

[tom66]

Re: Buffer Board IC replacement.

Samsung Q96HD right? The top chip drives odd lines, top of connector, second chip drives even lines, bottom of connector (or vice versa), and so on. I think replacing the second IC from the top would fix the problem, as it appears to drive the bottom most lines, and presumably those are the only ones shorted?

[Glitcher]

Re: Buffer Board IC replacement.

So the second from the right is the chip? I located one shorted pin from that one, just making sure there aren't any others.

This is the lower buffer board LJ92-01492A from a Samsung HPT5054

Found one shorted mosfet on the Y-sus board (88N30W) while the SMD mosfet on the board is still in question.

[Glitcher]

Re: Buffer Board IC replacement.

Heres a discovery I just made! This is the SMD Mosfet on the Y-sus board (Usually a common fault component)

My atlas DCA meter isn't able to get a reading off of it. EXCEPT when I heat up the Mosfet. Once it's heated up it gives me the following readings:

(Left to right, Gate, Drain, Source)
N-Channel MOSFET
Test current = 2.50 mA

Gate threshold Vgs = 3.7V (HOWEVER, as the component cools down the Vgs voltage slowly increases with each test. IE. 3.9V, 4.07V, 4.17V, and after the meter can't read it)

The Threshold voltage seems to increase as it cools. Does this mean the component is faulty? I'm not exactly sure of the mechanics of how MOSFETS work though.

Actually I just tried this test on several other FETS that were in a defective pile I kept, and the results are the same! The threshold voltage increases as the fet cools down. When I bought the replacement for the FET I always test it with my meter and it was able to take a reading so I know its not my meter.

So I guess the FETs may seem to have a good measurement with a multimeter but it won't work as well as its on the verge of failing or until it warms up through the circuit, but I'm guessing it would just get worst until it finally breaks down.

[tom66]

Re: Buffer Board IC replacement.

Atlas DCA55 manual page 55:


Peak test voltage across O/C (open circuit): +/-5.1V. A MOSFET gate is characterised as "open circuit".

It cannot test MOSFET Id/VgsON with a Vgs > 5.1V @ Id=2.5mA of which many high power MOSFETs have. It should still detect the device as a MOSFET, though.

You can check the datasheet of the B20NK50Z here: http://www.datasheetdir.com/B20NK50Z+download

In particular, page 4, bottom right chart. You can see that a test current of significantly more than 0A (and the Atlas looks for 2.5mA) would require a Vgs of at least 4V. The device will also vary with temperature and drain-source voltage (tested at 25V in the graph, but tested at 5V by the Atlas.) Page 5 bottom left shows a 30% drop in Vgs over 25C to 150C. Note on page 3 the variation in threshold voltage at Id=100uA to be from 3V to 4.5V, a tolerance of +/-25%; Vgs on a MOSFET is very rarely an exact parameter.

Therefore, the device appears to be functioning exactly as the datasheet characterises.

[Glitcher]

Re: Buffer Board IC replacement.

When the device passes that 5V threshold, the meter wouldn't read the device as it would cycle on a loop (measuring component... etc)

Its strange because one replacement I made in the past, comparing two of the same MOSFETs (one was brand new and the other was old, the one on the board) The new one had a threshold of 3V, while the old one was unreadable (threshold of > 5V). However once I replaced the mosfet the panel came back to life.

When mosfets age, can their increase in voltage threshold cause a problem?

[tom66]

Re: Buffer Board IC replacement.

It theoretically could, but that failure mode is really, really uncommon. The only way I could see it happening is if the gate-source became leaky, and the Vgs the DCA55 could put out fell. As threshold voltage increases, the DCA55 can't turn the device on properly, so it looks like an open circuit or faulty component.

I have a DCA55 too. It's a nice device, but it has some limitations.

[Glitcher]

Re: Buffer Board IC replacement.

Hm interesting, I always like the fact that even replacing them it would last a little longer, but I guess it isn't really necessary? I'll be able to save some costs in future repairs then.

That means I probably have a lot of good MOSFETs in that little pile. So glad I kept them (even though they were bad)!

[tom66]

Re: Buffer Board IC replacement.

I wouldn't usually reuse MOSFETs pulled from old devices because like capacitors they do have a service life and fail according to a probability function, getting increasingly probable as they age.

[Glitcher]

Re: Buffer Board IC replacement.

True, I would probably only use them for testing purposes though.