Warm Speaker Magnets Phillips 26PF5321D/37

[ZnsaneRyder]

I am inspecting a TV that was donated to me, and trying to find faults on it before I decide to put the TV to use. All capacitors look good.
The donor claimed the TV shut off on him when watching it, but the TV works fine for me.

There are a couple suspect areas of the TV.

Mainboard - has Rectangle chip above the tuner that runs so hot, you can barely place a finger on it. I AM going to put a small Heatsink on this chip.

Speakers - They run WARM when they are plugged in! You can put your hand on the speaker magnets and they are warm. When I unplug the speakers inside the TV, the magnets get cool again. I KNOW this is not from heat conducted from inside the TV, the heat is being produced in the speaker coils, which has me very alarmed. I suspect CLASS D carrier wave is poorly filtered and going into speakers. I am tempted to add a ferrite bead or inductor to the speaker wires. Maybe a small capacitor across the speaker.

I have to ask, is the speakers getting warm a sign of the Class D amplifier going bad? Is this just a normal manufacturing "flaw" with this TV?

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Check the DCV between the two legs of the speaker terminals, it should show in mVDC range.

[ZnsaneRyder]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Thanks Budm for your assistance.

The DC across speakers reads approximately 3-4mV on each speaker. So no DC voltage in the speakers. You can also hear a "hum" in the speakers with your ear against them, forgot to mention in the first post.

I did a test with an inductor and a 0.1uf cap across the speaker, and all the heat goes away, so it is just a very poorly filtered class D amp. If I use JUST the 0.1uf cap only without the inductor, I actually get light sparking when connected, and the hum gets louder which verifies the class-D carrier is dissipating in the speakers!

I measured the frequency with my meter and get almost exactly 161khz and the AC setting reads 2.5V AC.

What I did to help lower the temp some, was to wrap the speaker wires four times inside a small ferrite toroid salvaged from a CRT. They are still warm, but not as much.

I didn't mean to ask such strange question, but I NEVER have come across this before. I suspected Class-D carrier, but didn't expect it to be THIS BAD. It is just a crappy design, unfortunately like a lot of LCD TV's anymore. This is an old one from 2005. BTW, the heatsink on the chip on the mainboard helped with heat, although unrelated to the speakers, I want it to run cooler before using the TV. I have a peeve with overly hot components.

Thanks again,
ZnsaneRyder

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

What is the P/N of this Class D amp chip?

[ZnsaneRyder]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

What is the P/N of this Class D amp chip?

OnSemi NCP5422AG It's a buck controller, not really an audio chip!

I had to remove the mainboard, the chip is on the bottom.

And I don't see any output filters for the speakers on the board.

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

That cannot be the chip for the Stereo amplifiers. Can we see the pictures of the board? I wonder if the VCC feeding the AMPLIFIERS are clean Voltage.

[ZnsaneRyder]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Here is the pics.

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Pictures are too blurry for me to see clearly.

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Service manual:

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

It does not use Class D amplifiers, it is linear amps.

[ZnsaneRyder]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Thanks, I checked out that service manual you attached. It will become VERY useful for repairing other phillips TV sets as well that are covered by that manual.

The Amps are Class D, the speakers are driven by little 8-pin MOSFETS.

The TV works fine, I have left it on continuously.

The sound is working, I'll just not worry about the speakers, adding that ferrite toroid helped a little bit.

Thanks for all your help and input.

[tom66]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Not necessarily a fault if the speakers run hot, but certainly not that energy efficient. I have seen a few class D speaker amplifiers that rely simply on the LC of the speaker to filter the audio, only requiring some ferrite beads or RC snubbers.

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Thanks, I checked out that service manual you attached. It will become VERY useful for repairing other phillips TV sets as well that are covered by that manual.

The Amps are Class D, the speakers are driven by little 8-pin MOSFETS.

The TV works fine, I have left it on continuously.

The sound is working, I'll just not worry about the speakers, adding that ferrite toroid helped a little bit.

Thanks for all your help and input.

Look at the diagram again, it is just simple OPamp (LM339) driving phase splitter output circuit, not class D.

[ZnsaneRyder]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Thanks for the more detailed schematic budm!

I admit I was totally incorrect about that Onsemi PWM chip being the audio source, because I have seen PWM chips used as drivers before for class D output stages. (TL494 for example used as an audio amp for class D)

I still say it appears to be class D and it even looks very similar to the "ultra simple discrete class D amp" on diyaudio forums.

LM339 is a comparator, not really an Op-Amp, but works similar, but without bias in the output transistors.

Also, if it was a class AB amp, it would have a bias on the NPN/PNP (phase splitter) transistors, not having the bases directly connected together. The NPN/PNP phase splitter input is to drive the N and P channel mosfets in opposite directions, but again, doesn't appear to apply any biasing to them either. Also, at the wattage spec this amp is, a class AB amp would have some sort of a heatsink.

Here is an example of a simple self oscillating class D amp driven by an opamp, only difference being it uses NPN/PNP totem poles to have more strength to drive the mosfets instead of just a constant current source transistor like the phillips TV does, other than that they are similar.

By the way, this topic is interesting, and thanks for the informative discussion.

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

LM339 can also be use as an OPAMP (see page 14 of TI AN-74), if you look at Philps diagarm, where is Triangle WAVE GEN that feeds the comparator I do not see it, see IR App-notes AN1071. Philips is too cheap to use real Quad Opamp.
You example circuits uses all 4 section of the 339.

[Alastair E]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Philips always did do weird stuff or use odd parts for different jobs....

One of the first to use Valve/Tube in a popular amp with no output transformers--and 600 ohm speakers...

Or to use PL36 horizontal-output tubes in cinema amps--even though much better valves were available...

Or pieces of string and springs to lace up video-recorders and shaded pole induction-motors with eddy-current brake-discs for head and capstan....

Philips = Quite Often Weird

[budm]

Re: Warm Speaker Magnets Phillips 26PF5321D/37

Their diagrams are sometimes hard to read also, especially the designator for the components using just numbers instead of I.E. Rxx, Cxxx.