Panasonic TH-P50U30A SC Board TNPA5351 AF failure

[tw2005]

I picked this one up today, don't have the blink or LED status as I've decided to not power this up already knowing most likely shorted SC. It was described with either locked solid LED or may have been none but i was told no blink codes.

Just jumped straight into it and SC2 100% dead short, 4 screws loose on the SC, arcing and burn marks underside, burn marks under the screws and solder lands also burnt and eroded away.

very poor in comparison to my 2010 model and very poor of panasonic not to provided out of wty or recall. these are doomed from the day out of the factory.

Also these screws are not even dead flat, slight concave under the heads so only the outer rim contacts the board.

Looking like just GT30F131 and the 2 diodes although there's more than 4 30F131 showing short, hoping it's not all of these but I am going to sub these with FGD4536TM, from RS components so no risk of chinese copies, and about $1.50 each so cheap repair it should be. Also I'm going to use RF1501NS3S for b both as it's higher spec than the RFN10 and they are pretty common on panasonic sustains which I've scavanged.

I'll get some shots up showing the burns etc down the track

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Well packed the camera, no batteries or charger, poor cameraman but hopefully everyone gets the idea. Similar to all the other reports posted up.

What I'm thinking of doing is using solder tags and soldering them over the mount holes underside giving a larger contact pad and 360 degrees coverage.

If anyone thinks this could be an issue, sing out.

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Diodes(D401,421), gate resistors just fine(R402,403,422,423), just 4 x 30F131 (Q402,403,422,423), drivers and cph5524 transistors fine.

Initially gate resistors were off, R402,403 were 5 ohms, R422,423 were 7.2 but once all shorts removed all reading 7.5 ohms.

Also something I noticed. I've seen the driver ICs advertised by a UK seller as D86903, C0ZBZ0001822.

Well the full part is BD8693FVM-HVT, and the marking on the case is D8693. The "0" is in fact a round impression in the case, looks like an O, at a glance I guess 0.

[Glitcher]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

What are the specs for the 30F131? And was wondering if the 30F132 would be similar?

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

What are the specs for the 30F131? And was wondering if the 30F132 would be similar?

132 very close, just higher current, peak is 250A, not sure about constant current, would expect same or higher.

http://pdf.directindustry.com/pdf/to...79-278279.html

try a search

[tom66]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Constant current in a PDP driver application is a few amps, maximum, especially for these new energy efficient panels. It's peak current that makes all the difference. If you exceed the device's safe operating areas, then it will fail sooner or later...

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Constant current in a PDP driver application is a few amps, maximum, especially for these new energy efficient panels. It's peak current that makes all the difference. If you exceed the device's safe operating areas, then it will fail sooner or later...

So what can you tell me about this? Any ideas of what peaks they'd hit then?

Fairchild

PD: Maximum Power Dissipation @ TC = 25oC

125 W
-------------------------------------------
Toshiba

PC (W) @Tc = 25˚C

140W

[tom66]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Well, when you dissipate power in a transistor (or any device for that matter) its temperature will rise. One of the nice things about thermal dissipation in electronics is it's really easy. It's a simple linear relationship, and independent of the ambient temperature. It is simply additive and multiplicative.

A good FET/IGBT connected to a standing heat sink might achieve a "thermal resistance" of 5C/W, which essentially means for each watt dissipated in the device, the temperature will rise by 5 degrees.

If you assume an ambient temperature of 40C and say the acceptable temperature limit is 20 degrees below the rating of 125C (as you want to extend device lifespan) i.e. 105C max you can calculate the maximum temperature rise easily: it's 105C - 40C = 65C. This gives a maximum power dissipation of 65C / (5C W^-1) = 13 watts.

Oh, and note this is the junction temperature, i.e. the temperature right at the semiconductor itself. The device package always runs cooler, and the heatsink runs even cooler.

So, you might notice, 13 watts << 125 watts or 140 watts. If you try to dissipate 125W in an ordinary transistor on an ordinary heatsink, it'll be destroyed very quickly.

So, what does the rating mean? Sort of like horsepower ratings on a car. A bit theoretical, and a bit bullshit... The power rating assumes things like, if you could extract as much heat from the device as theoretically possible (say your heatsink had a 0C/W rating and was made of some advanced super alloy with infinite thermal conductivity and zero thermal capacitance and was infinite in size, but the silicon die and package still has thermal resistance) then YES you could dissipate that much power in the device. But in reality, the thermal resistance is dominated by the heatsink itself, rather than the losses in the package (which determines the maximum power rating)

Now, one of the things Panasonic did with this design is they ditched the heatsinks and used a thick board to dissipate the heat... This has a MUCH greater thermal resistance than a conventional design (using a proper heatsink.) I don't have any numbers for Panasonic's designs, but a typical thermal resistance on the order of about 20~60C/W can be achieved with a PCB heatsink.

And of course, this means the device power rating is even less important.

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Well, when you dissipate power in a transistor (or any device for that matter) its temperature will rise. One of the nice things about thermal dissipation in electronics is it's really easy. It's a simple linear relationship, and independent of the ambient temperature. It is simply additive and multiplicative.

A good FET/IGBT connected to a standing heat sink might achieve a "thermal resistance" of 5C/W, which essentially means for each watt dissipated in the device, the temperature will rise by 5 degrees.

If you assume an ambient temperature of 40C and say the acceptable temperature limit is 20 degrees below the rating of 125C (as you want to extend device lifespan) i.e. 105C max you can calculate the maximum temperature rise easily: it's 105C - 40C = 65C. This gives a maximum power dissipation of 65C / (5C W^-1) = 13 watts.

Oh, and note this is the junction temperature, i.e. the temperature right at the semiconductor itself. The device package always runs cooler, and the heatsink runs even cooler.

So, you might notice, 13 watts << 125 watts or 140 watts. If you try to dissipate 125W in an ordinary transistor on an ordinary heatsink, it'll be destroyed very quickly.

So, what does the rating mean? Sort of like horsepower ratings on a car. A bit theoretical, and a bit bullshit... The power rating assumes things like, if you could extract as much heat from the device as theoretically possible (say your heatsink had a 0C/W rating and was made of some advanced super alloy with infinite thermal conductivity and zero thermal capacitance and was infinite in size, but the silicon die and package still has thermal resistance) then YES you could dissipate that much power in the device. But in reality, the thermal resistance is dominated by the heatsink itself, rather than the losses in the package (which determines the maximum power rating)

Now, one of the things Panasonic did with this design is they ditched the heatsinks and used a thick board to dissipate the heat... This has a MUCH greater thermal resistance than a conventional design (using a proper heatsink.) I don't have any numbers for Panasonic's designs, but a typical thermal resistance on the order of about 20~60C/W can be achieved with a PCB heatsink.

And of course, this means the device power rating is even less important.

Ok, thanks for the mental aerobics so essentially if it were to operate at the max it would burn up because the heatsinking is inadequate, so i really don't need to worry about the shortfall.

Good news I don't have wait much longer , bits are in a courier van somewhere

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Ok, I think I have a problem. - FGD4536TM arrived and looks like I needed to pay closer attention.

Spec wise this should be fine but seriously so small is it really going to dissipate enough heat?

Also RS had these listed a P channel IGBT, I'm pretty confident that's wrong.

Any thought? Put them in and see or what?

[tom66]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

No, I wouldn't use them to fix it, the case resistance will be higher so it won't be able to dissipate enough heat into the board. Maybe as a quick test, but not permanent.

P channel IGBTs are theoretically possible, but I don't think anyone has ever made one - there's no need to.

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

In the interest of science, I forged ahead (LOL). Got them in just, problem was making suring 100% of the tab was soldered and getting the legs to reach. Just makes it .

They do say these are designed for PDP panels and so it fired up fine. Running full white, they appear not to be getting as hot as the remainder of the 30F131.

These heatsinkless boards are eerily quiet. Full white is not exactly like the old models, definitely washed out a bit but nothing abnormal at least. Has about 7500hrs

can also confirm the SOS history was 00:00:00:00

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

No, I wouldn't use them to fix it, the case resistance will be higher so it won't be able to dissipate enough heat into the board. Maybe as a quick test, but not permanent.

P channel IGBTs are theoretically possible, but I don't think anyone has ever made one - there's no need to.

Yeah, ok, well that's a disappointment. They do seem to run cooler than the 30F131 and the heat draws away quicker on shutdown than the 30F131 due to it's added bulk holds the heat.

I may make this one a guinea pig, only owes me $56 all up.

[tom66]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Just be aware, if one of those small IGBT/FETs dies around the ramp ckt area, it's night-night for the buffer board!

Turn off C.A.T.S. in the user menu to do a proper full bright test. (Or service menu full white works too.)

My experience with the U30B was it had a very faint buzz but that was all.

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Just be aware, if one of those small IGBT/FETs dies around the ramp ckt area, it's night-night for the buffer board!

Turn off C.A.T.S. in the user menu to do a proper full bright test. (Or service menu full white works too.)

My experience with the U30B was it had a very faint buzz but that was all.

I was running in serv tool. My ears have re-calibrated over the years so I could only just make it out.

Seems to have a clearer image than my 2010 FHD

[tom66]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Interesting, three U30Bs with nearly 8000 hours on them, all dead short Vsus. Wonder if it's just a coincidence, or something else is going on (like panel age algorithm transition.)

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

Interesting, three U30Bs with nearly 8000 hours on them, all dead short Vsus. Wonder if it's just a coincidence, or something else is going on (like panel age algorithm transition.)

I found with my 50V20 series, the hours were a bit random.

I think it's definitely related to the arcing of the SC, but then 8000hrs is probably about right these days.

I found 1 screw (very top ) on the SS burnt and loose. The SU top screw was also a little loose.

replaced all the SC / SS screws and the top screw at the SU.

This seems more internet friendly too. anyone know if a wi-fi dongle will work for these? Brand etc?

[tom66]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

My U30B had an ethernet port... but it was only for Freeview HD internet TV (you could watch some channels over the internet, ones even less worth watching like shopping channels.)

Mine didn't have arcing screws anywhere; I took them all off and examined them. Think it only affects some of the bigger models.

[jeffecor]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

HI, I had a question what is the ohms reading if all the mosfets remove from the same board (I have on the model TC-P46ST30) or what the ohm reads if the everything is good on the board on the main power going in on the SC board.
I have found that sc board is causing the short and last night I replaced the 4 mosfets Q402,403,422,423 thats was reading high on ohms.

Then I get 2.784 ohms on the main power going in and before I was reading .1-.0. I starting to wonder if i need to replace all the diodes and mosfets on the board.

[tw2005]

Re: Panasonic TH-P50U30A SC Board TNPA5351 AF failure

HI, I had a question what is the ohms reading if all the mosfets remove from the same board (I have on the model TC-P46ST30) or what the ohm reads if the everything is good on the board on the main power going in on the SC board.
I have found that sc board is causing the short and last night I replaced the 4 mosfets Q402,403,422,423 thats was reading high on ohms.

Then I get 2.784 ohms on the main power going in and before I was reading .1-.0. I starting to wonder if i need to replace all the diodes and mosfets on the board.

well something is not making sense. You replaced the 4 IGBTs that were reading high ohms? High ohms would mean not shorted but what value is high ohms?

I don't have acces to this Tv anymore so i can't get any values. the kits show replacing 2 diodes next to these D401,421 but mine were fine so too were the gate resistors.

So definitely 2.784 ohms and not 2.784Mohms at Sc2?

How are all the other IGBTs like q441 as an example?

What type of meter do you have?