TDA7295:

"mark g"

Look for cracked pins, bad soldering, cracked PCB tracks etc,

Or the IC was damaged during previous trouble or service work.

If all still looks OK, stick a new one in.

... Phil

as

TDA7265

working

are

the

What is the simple, no powering, DVM ohms measurement between , separately, output and each supply rail ?

I am in the process of repairing an audio amplifier that employs a TDA7295 as the woofer driver. The repair of the original fault, which was causing the TDA7295 and a TDA7265 to be held permenently muted, went fine, but having got the mute control level working again, I find that I have no output from the TDA7295. It seems to be shot - for reasons that are however unclear to me

- but I would appreciate a second (or more) opinion. The levels around the chip are as follows: Supplies: +/- 28V, Vin+: input signal of +/- 0.7V, Vin-: just a dc level of -9.6V, Vout: just a dc level of -24.8V, Bootstrap input: just a dc level of -25.6V. These levels remain the same regardless of the state of the MUTE input. STANDBY is always inactive, i.e. >+3.5V. Any suggestions on how this state could have arisen?

Is the PCB partly covered in a black goo? If so you will have to get all this off as it is often goes conductive and can wreak havok.

Gareth.

The word 'around' might be significant. You need to measure the levels actually on the pins of the chip (make sure you have a probe that will do it without risk of shorting) because it looks as though your +ve supply isn't getting through.

"Adrian Tuddenham"

Many scope probes have a simple accessory that shrouds the point so it will not possibly short adjacent pins on a DIL package.

BTW:

When troubleshooting, it pays to use a scope set to DC coupling first - rather than a DVM.

Cos the scope screen shows what is really there.

And I mean a REAL scope - ie an analogue one.

... Phil

responding to

The measurements I reported were made using the pointed tip of an oscilloscope probe and I thought I had been on pins of the chip, but I agree, the negative voltages do suggest the positive supply is missing.

What I forgot to mention was that the circuit actually worked for a few minutes. Prior to, and after that brief period, the symptoms were the same, i.e. as described earlier. It does rather suggest a bad joint, doesn't it, but I couldn't sort it out.

Anyway, I'll investigate again and let you know.

maybe you have latch up.?

Some chips have problems if you over drive the input past the rails, they latch up and remain that way until you reset them or some how over come it.

It appears that it has MOS technology in it, so that could be the case..

A good test would be to just power it up but don't inject anything in it. See how long it'll sit idle.

If it seems that is the case, you can put diodes from the rails to the input that is being driven to prevent the input from exceeding the rails.

Jamie

responding to

The measurements I reported were made using the pointed tip of an oscilloscope probe and I thought I had been on pins of the chip, but I agree, the negative voltages do suggest the positive supply is missing.

What I forgot to mention was that the circuit actually worked for a few minutes. Prior to, and after that brief period, the symptoms were the same, i.e. as described earlier. It does rather suggest a bad joint, doesn't it, but I couldn't sort it out.

Anyway, I'll investigate again and let you know.

Mark, I have seen lots of these TDA powered speakers, mostly studio monitors, with the "conductive glue" problem.

At the risk of repeating myself, have a good look at the PCB, particularly between the pins of the TDA ICs. If there is any kind of chemical coating, yellow or black, then this really is very likely to be your problem. It starts off as component adhesive, but over time melts and runs everywhere, eventually becoming brittle and conductive, causing erratic and illogical behaviour.

Gareth.

Mark, I have seen lots of these TDA powered speakers, mostly studio monitors, with the "conductive glue" problem.

Tannoy, Alesis, and KRK are 3 brands that immediately spring to mind, this is a generic Chinese manufacturing problem.

Gareth.

"Gareth Magennis"

Roland " Cube " series amps had problems with it and they were made in Japan.

VCRs had problems it in spades as do many SMPS too.

Class D amps used in high end line array boxes made by dB Technologies have it splashed about.

The stuff does NOT melt and move around, it simply goes dark with heat and time and becomes conductive and corrosive to copper and tin.

Long as it has no contact with metal, it hardly matters.

Now, the colossally stupid Chinese spread it about on PCBs like jam on toast.

They have the exact same philosophy on these matters as Alfred E. Neuman.

... Phil

I assumed that because it was SO spread around on PCB's where there was no need for it (on the pins of TDA IC's for example) that it must have migrated there. It beggars belief that anyone would do such a thing.

The stuff I see in modern equipment is usually black and goes brittle. Other stuff I have seen, which may be older (?), was more yellow/brown.

Cheers,

Gareth.

Been there, done that, didn't get the tee shirt. The picture on a 23" LCD TV was badly distorted. All voltage out of the power supply were spot on, as were the outputs of the LDO regulators on the main board EXCEPT for one in the area of the digital video output. The 3.3 volt output read 3.1. Checking it with a scope showed a clean input and nearly half a volt of noise on the output. In this case the fault was a pair of 100µF electrolytics on the output.

PlainBill

responding to

It's been a long time since I asked for your help. I've had a lot of other things to deal with.

Anyway, I tested for the presence of the +ve supply on the pins of the chip and I looked again for any sign of bridges, cracks etc., but couldn't find anything.

Tonight I finally replaced the 7295 *and it's running again*. It's current doing a burn-in test, i.e. I'm listening to a load of music with cables hanging everywhere, before I begin the onerous task of reassembling everything.

Incidentally, the board was indeed inundated with that disgusting browny yellow glue stuff, and it was in one of those area that the original fault occurred - a debris short between two pins of a dil connector.

Thanks everyone for your interest and your help.

Regards, Mark