I think I fried a good ZSUS board on an LG plasma

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

to check the gate driver diodes do you take a ohm reading between the com pin and the others ?

[will62]

Re: I think I fried a good ZSUS board on an LG plasma

Do you have an oscilloscope, or does your meter have a frequency check on it?

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

no oscilloscope but my meter does have a frequency checker. Can i do this with it still on the board ?

[will62]

Re: I think I fried a good ZSUS board on an LG plasma

I don't think that you can check a gate driver completely without applying a pwm signal in and verifying that you get a pwm of higher amplitude out to the gate. Your meter probably tops at about 4kHz, so you may be able to confirm the in and out signals. I believe the scan rate is somewhere under 1kHz.

Looks like you have a few different groups of igbts. Why don't you go ahead and check for continuity between pin 7 and all of the gate resistors at the igbts that blew. Let's make sure this driver is going to those igbts. If it is, go ahead and check all the driver pins for obvious problems like shorts to ground or vcc.

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

getting late here. I will check that in the morning and report back. Thanks for the help.

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

Driver readings below

Pins to ground

1 (vcc) to ground 4.1k Ohms
2 (hin) to ground 4.6k Ohms
3 (NC) to ground OL/open
4 (com) to ground .2 Ohms
5 (nc) to ground OL/open Ohms
6 (vs) to ground 1.0m Ohms
7 (ho) to ground 1.0m Ohms
8 (vb) to ground 1.3 Ohms

vcc to Pin 2, 14k ohm
vcc to Pin 3, OL/open
vcc to Pin 4, 4k ohms
vcc to Pin 5, OL/open
vcc to Pin 6, 1.0m ohm
vcc to Pin 7, 1.0m ohm
vcc to Pin 8, 1.8m ohm


pin 7 to gate resistors supplying the four transistors that are blowing.

Three are reading 20K Ohms (they are the good gate resistors reading 5 ohms)
One is reading 1m Ohm (this is the bad gate resistors that reads open)

[will62]

Re: I think I fried a good ZSUS board on an LG plasma

Ok, I'll look those over. Can you take a full photo of the bottom of the board?

Also, I'm including a link where you can download the LG training manual for your set.
Please do the diode mode tests on pages 82 & 83. No need to list all results, just list
the variances to the tables in the manual.

https://www.manualslib.com/download/...Lg-50ps60.html

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

Attached are some pictures. I have not put put 4 new transisters in yet because my soldering iron just died. Also I did ohm out to the pins to ground and I was getting high reading of 3.5k ohms on the vs, open on the VA & 2.7m ohms on the M5V. Again this was with 4 transistors out.

Ok, I'll look those over. Can you take a full photo of the bottom of the board?

Also, I'm including a link where you can download the LG training manual for your set.
Please do the diode mode tests on pages 82 & 83. No need to list all results, just list
the variances to the tables in the manual.

https://www.manualslib.com/download/...Lg-50ps60.html

[will62]

Re: I think I fried a good ZSUS board on an LG plasma

A couple of problems:

1. Your different groups of igbts are doing different things (i.e. sustain/erase,bias waveforms), so they will have different driver ICs running the gates. But, the resistance that you measured from pin 7 to the gate resistors is far too high for both groups of igbts. IC306 may be running the 3 igbts to one side. Their gate resistors are OK, so we're not really concerned with that gate driver. We are concerned with the gate driver that goes to the bad gate resistors on the one blown igbt. Is the IC to the right and above in the pic, another gate driver? If so, does pin 7 go to the bad gate resistors. If you can't visually trace routes, you can use continuity checks with your dmm.

2. For the diode mode checks that are in the manual, You need to follow the directions carefully. Disconnect the connector, DMM in diode mode, black probe to ground, red probe to each pin. You are giving me results in ohms. If you were in diode mode, you would be reporting voltage values (i.e. 0.6V). There are 2 tests (p100,p101).

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

ok got it...in diode mode all pins on p101 are reading good as per chart. I think there is a problem with P100, pins 1-4 should be 1.9, I am only getting 1.1. Pin 5 is good. Pins 6,7,9,10,11,12 should be open. I am getting 1.5v

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

Is this a problem, I found these two resistors both shorted in each direction with no resistance in both diode check and in ohms. Diode read .0002v and ohms 2.3

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

Is this a problem, I found these two resistors both shorted in each direction with no resistance in both diode check and in ohms. Diode read .0002v and ohms 2.3

I checked my diode calculator and the ohm reading is correct but i am still getting a reading of 0.002v each way in diode mode.

[will62]

Re: I think I fried a good ZSUS board on an LG plasma

Good catch on the resistors. I think that is where the zbias voltage is measured for adjustments.

On the p100 variances, I found a summary sheet in the back of the manual that has completely different diode check values.
Which one is right? Are either of them right? Who knows.

[budwich]

Re: I think I fried a good ZSUS board on an LG plasma

Is this a problem, I found these two resistors both shorted in each direction with no resistance in both diode check and in ohms. Diode read .0002v and ohms 2.3

I don't think 2.3 ohms is "no resistance".... not every component has "vast amounts of resistance"... there are components that are in the "unit ohms". You need to check the actual value of the resistors either with the code or in the manual / board spec.

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

I have researched those two resisters, I don't think there is anything wrong. They both are rated for 2.2 and I am getting 2.3 (red red gold gold).

Getting back to the gate resistors. If there was a problem with the driver or drivers wouldn't they all blow ? Only two blew that were in parallel feeding one shorted tanducer. The other 3 transducers that were shorted the gate resistors were fine....Just thinking here.

Would the two blown gate resistors in parallel cause transducers to short ?

I replaced the four transducers and the two gate resistors so the board is ready to go again !! Nothing has really this time except I replaced the two gate resistors. Should I give it a shot ?

[will62]

Re: I think I fried a good ZSUS board on an LG plasma

Yes, I was thinking the same thing about having only 1 set of gate resistors blown. The gate drivers may be fine. Up to you on firing it up. How many of those igbts did you buy?

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

I have been robbing good ones from old boards but i am running out...Lol. I do have more coming from China but that takes 3 weeks to a month.

Yes, I was thinking the same thing about having only 1 set of gate resistors blown. The gate drivers may be fine. Up to you on firing it up. How many of those igbts did you buy?

[goldbird78]

Re: I think I fried a good ZSUS board on an LG plasma

Anyone have an idea if blown gate resistors would short out transistors ?

[will62]

Re: I think I fried a good ZSUS board on an LG plasma

From a T.I. paper on gate drivers:

"For completeness, the external series gate resistor and the MOSFET driver’s output impedance must be mentioned as determining factors in high performance gate drive designs as they have a profound effect on switching speeds and consequently on switching losses."

Those igbts are in parallel. If resistance of 1 set of gate resistors goes high like yours did, they will switch at different speeds = disaster.

That doesn't account for the other 3 that are on a separate circuit.

[will62]

Re: I think I fried a good ZSUS board on an LG plasma

I'm going to throw something else out there, and you tell me if it makes sense.
I wouldn't know without seeing a schematic, but let's suppose that one set of parallel igbts is responsible for generating the positive (high) side of the wave, and the other set of parallel igbts is responsible for generating the negative (low) side of the wave. If this is the case, then I know for a fact that the high and low igbts can not be on at the same time, without blowing them. There is usually some dead time between 1 set going off and the other coming on.

Now, if the 1 igbt with bad gate resistors (1 of 3 high), is not turning completely off before the other 3 igbts (low) turn on, then those 4 igbts would
blow. That is one way I could see those 4 igbts blowing together. In the attached pic, you can see just a bit of jitter at the positive to negative transition. That might be the dead time between switching that I mentioned above.