TDA7293 Amplifier Design Blows Up Instantly

My excuses. It is not often the Germans make formal mistakes. They have all their DIN-norms etc. Skilled people.

I fully respect your personal opinion. My personal logic is that ST has sold this product for 15 years and ST is a rather professional company. I have two very simple and cheap TDA7293 boards and they "behave".

What would you recommend as replacement for that power level? LM3886?
 
Experience says you would be well advised pulling the fuses to the TDA chip, and replacing them with say 100 Ohm resistors. (A neat way is to solder 100 ohm resistors across blown fuses).

This will let you try powering the thing up (without a load) and checking you can get it to run OK unloaded.

If you have a serious issue like a short, or oscillation going bonkers, or outputs connected fighting one another, worst case you will see smoke from the resistors.

I dont guarantee 100 Ohms is ideal, but in my experience most power amplifiers will happily run with 100 ohms in each rail if not loaded.


The only other hint I offer (given all the other advice) is strip the thing back to the utter minimum that ought to work. Don't be shy, pull bits out till you can at least get the tiniest part working.

If the tiniest part doesn't work, well, drink some coffee and nut it out.

Then start adding bits back until those 100 Ohm resistors do something interesting.
 
A voice of reason, shortly to be drowned out again by the un-reasoned shouters ;-)


The guy wants help with his implementation. You don't help with just shouting personal opinions of what is good or wrong, which in itself may be totally wrong.
Please assist him, if you can't bugger off.

Jan

Agree.
So here is my suggestion: if the OP wants to build a parallel TDA7293 power amplifier, (fine with me), then he must buy a parallel TDA7293 power amplifier kit, with a ready made and tested PCB, an amp which has been professionally designed and tested and where bugs have been ironed out.

Any new project takes some development time and effort; if he adds to his workload also designing and making a PCB, creating a good, stable layout , etc. , he will probably bite more than he may chew.

Hey, I have been designing and building amplifiers for ages, lots of them, and STILL new projects show one problem or another which needs correcting, or at least adjusting.
And I am not ashamed in the least to recognize that, it´s part of the job reality.

As a side note, I *hate* paralleling chipamp outputs ... extend that to any "perfect voltage source" type amps, including discrete ones, and the reason should be self evident.
 
There are apparently many skilled people ready to help but I get the feeling the OP has left the thread. Problem solved or giving-in?
My approach would be to test the board connections without any TDA7293 inserted. When the board with all the passive components inserted has been approved, to insert the master TDA7293 and test without a load and at low supply voltage. Only when the master works, I would try to implement the slave.
But, without the OP it may be futile.
 
There are apparently many skilled people ready to help but I get the feeling the OP has left the thread. Problem solved or giving-in?
My approach would be to test the board connections without any TDA7293 inserted. When the board with all the passive components inserted has been approved, to insert the master TDA7293 and test without a load and at low supply voltage. Only when the master works, I would try to implement the slave.
But, without the OP it may be futile.

I have definitely not left the thread and neither solved my problems nor given in. I have not yet been able to find the time to make further progress and have only read all the input here with much interest. I will report back as soon as I have got new infos which will be very soon!
 
Thanks Marco for your confirmation. When you have time again, please let us know.
I believe you can get it to work but need a different approach than mounting everything and try to look for flaws on a board that has had a serious fault. You probably don't need us all to give you advises going in different directions so when you start again, choose a few as advisers.

I believe most of us were not aware that you try to implement a master/slave configuration that is rare but described by ST in the datasheet. Without knowing that particular option, your schematics appeared somewhat dodgy.
 
A thought: if the -supply is disconnected, how can it then blow up? Where is that 'blow up current' going to/coming from? Whats the path? Presumably from +supply but where to? Is there a ground connection at the chip, except for the standby pin?
Are we sure there are no other errors??

Edit: I see JMFahey has already pointed them out. Problem identified and diagnosed.

Jan

While testing, and assuming that the LS may be in the fault current path, it may be helpful to add a capacitor in series with the LS. You can start with a small one (maybe 10uF) and progress upwards. It also is then possible to measure the amplifiers offsets, and see possible DC-paths between the two amplifiers.

Note: with the amplifier outputs parallel (as here :)) use two capacitors.

P.s.: A resistor in series with the capacitor may also be helpful.
 
Last edited:
There are apparently many skilled people ready to help but I get the feeling the OP has left the thread. Problem solved or giving-in?
My approach would be to test the board connections without any TDA7293 inserted. When the board with all the passive components inserted has been approved, to insert the master TDA7293 and test without a load and at low supply voltage. Only when the master works, I would try to implement the slave.
But, without the OP it may be futile.

Using a simulator, to calculate voltage values, may be of help here, one does not need a model for the amplifier to do this.
 
The TDA7293 ICs have the tendency to blow up if the negative rail disconnects for any reason or at any instant in time when the chip is powered up is not the most negative connection to the IC. Also, if you install the bootstrap capacitor backwards (eg polarity of cap is backwards) this can happen.

See if you can check for these problems in your amplifier.

There is more info on the web about these kind of issues with the TDA7293 if you search for it.

I had a problem with the tda popping loudly in the speaker if the negative rail went first. I designed a hybrid valve/tda circuit and put the heaters from the +ve supply. Of course on power down the +ve supply went first and the tda would give a big thump through the speaker.
So I changed heater to be on -ve rail and this fixed the problem.

I also use 120R power resistors in power leads on first power up.

I also made a mistake mounting the tda right down on the pcb.
This caused the front pins to short on the rear pins pads on the pcb !
I desoldered the old tda and soldered in a new one about 2-3 mm above the pcb and everything was fine.