Hi!
I should've made this thread a bit earlier, but it's finally reached the point where I think I can't go on any further on my own. I received this amplifier as a hobby repair project. I'll try to be succint in my description to avoid making the post too long to read
Failure
It was accidentally switched over from 230V to 115V, as the switch is exposed. The fuse blew and the amp shrieked for a while and died.
Early analysis
I found a blown up TDA7294 and remains of two capacitors (C24 and C25). The capacitors have exploded their remains all over the board, which I've cleaned up. Later I learned that those were the caps on the MUTE and STBY lines. I've replaced the chip, plugged everything back, switched the amp to correct voltage, replaced the fuse and turned it on. The speaker shrieked and the TDA started to smoke immediately.
Repair attempts
There's no schematic or repair manual available. As such, at first i attempted to poke around the amp a bit. I've built a current limiter on a 60W incandescent bulb in series. I've measured the voltages with the chip unsoldered, but it didn't make much sense, so I tried to build a TDA reference schematic on a breadboard, but failed. Later on I've learned (also through this forum) that it's such a finnicky beast. I blew up three more TDAs while testing it with various power limiters (a pair of lab supplies, current limiter etc.).
Current state
I have written down about 95% of the schematic and the PCB to help with understanding the problem. I've looked at the MUTE and STBY lines but they have always properly reported as +7.5V from ground for me, with the delay capacitor charging over 1s, which I checked on the scope. That being said, every time I put a new chip in (speaker and input disconnected) my current limiter bulb is a bit too bright for my taste, and the supply voltages sag significantly. The last two chips I've fried trying to solder their connections one by one, starting with the negative supply, trying to figure out what could be causing the problem.
Vdd = 40V, Vss = -40V, GND is at 0, input at 0. With the chip in, output has shown DC bias. It's hard to measure voltages accurately as the chip is actively trying to fry itself anytime i put one in. I've also damaged the board pads slightly, as soldering and desoldering a 15-pin chip repeatedly takes its toll.
Files
I've attached a zip with my KiCAD project, with the schematics and PCB. I'm also attaching photos for reference.
EDIT: some P.S.
I should've made this thread a bit earlier, but it's finally reached the point where I think I can't go on any further on my own. I received this amplifier as a hobby repair project. I'll try to be succint in my description to avoid making the post too long to read
Failure
It was accidentally switched over from 230V to 115V, as the switch is exposed. The fuse blew and the amp shrieked for a while and died.
Early analysis
I found a blown up TDA7294 and remains of two capacitors (C24 and C25). The capacitors have exploded their remains all over the board, which I've cleaned up. Later I learned that those were the caps on the MUTE and STBY lines. I've replaced the chip, plugged everything back, switched the amp to correct voltage, replaced the fuse and turned it on. The speaker shrieked and the TDA started to smoke immediately.
Repair attempts
There's no schematic or repair manual available. As such, at first i attempted to poke around the amp a bit. I've built a current limiter on a 60W incandescent bulb in series. I've measured the voltages with the chip unsoldered, but it didn't make much sense, so I tried to build a TDA reference schematic on a breadboard, but failed. Later on I've learned (also through this forum) that it's such a finnicky beast. I blew up three more TDAs while testing it with various power limiters (a pair of lab supplies, current limiter etc.).
Current state
I have written down about 95% of the schematic and the PCB to help with understanding the problem. I've looked at the MUTE and STBY lines but they have always properly reported as +7.5V from ground for me, with the delay capacitor charging over 1s, which I checked on the scope. That being said, every time I put a new chip in (speaker and input disconnected) my current limiter bulb is a bit too bright for my taste, and the supply voltages sag significantly. The last two chips I've fried trying to solder their connections one by one, starting with the negative supply, trying to figure out what could be causing the problem.
Vdd = 40V, Vss = -40V, GND is at 0, input at 0. With the chip in, output has shown DC bias. It's hard to measure voltages accurately as the chip is actively trying to fry itself anytime i put one in. I've also damaged the board pads slightly, as soldering and desoldering a 15-pin chip repeatedly takes its toll.
Files
I've attached a zip with my KiCAD project, with the schematics and PCB. I'm also attaching photos for reference.
EDIT: some P.S.
- I can post images of the schematics if people don't generally use KiCAD - it's much easier to browse them in the software, especially considering that with the accompanying PCB you can quickly locate the components physically on the board.
- The schematics might have errors - especially around the component values (measured in situ) and the input section which I didn't really work on too much - just to clear things out.
- The PCB is made just for verifying connections, so it's missing large areas etc. - I am thinking about sending it to fab, but I'd improve it like the original first.
- The total count of dead TDA7294's (including the one in the initial catastrophe): 7
Last edited:
I got them from two different sources, both unreliable. I never had issue with fake circuits misbehaving like that. I expected it might sound worse or distort at max volume, not go into flames immediately. Is this something that fakes do?
I had several problems with fakes L4974, a Buck converter used industrially. I had to repair a 24/12V downconverter whose original part last 15 years. Lytics get dry and the IC blew. Once all recaped, 2 IC's appearently blown in the pcb without lasting a second. Having revised all, no found any problem: injecting 12V from the output wirhout the IC, no more drain as a small current from voltage sensing circuitry.
The third IC got in Argentina, prior to insert in the PCB was measued with a DMM: a powerfull shortcircuit between several pin of the IC just pulled from the plastic antistatic blister. We advised the seller and sent me 5 more units. All them in the +/- same condition.
A guy who was traveling got one in Europe. I told the converter's owner that wait some time this guy returns with the IC in the pocket. I put it in the PSU previous good DMM measuring and voilá: the Buck is nowadays still working since more than 8 years ago.
The third IC got in Argentina, prior to insert in the PCB was measued with a DMM: a powerfull shortcircuit between several pin of the IC just pulled from the plastic antistatic blister. We advised the seller and sent me 5 more units. All them in the +/- same condition.
A guy who was traveling got one in Europe. I told the converter's owner that wait some time this guy returns with the IC in the pocket. I put it in the PSU previous good DMM measuring and voilá: the Buck is nowadays still working since more than 8 years ago.
7294 is sold here for $2, claimed genuine.
Better build a transistor amp, safer, cheaper, more reliable.
Better build a transistor amp, safer, cheaper, more reliable.
Well, this is about repairing an existing amplifier, not building a new one. I know that it worked before, and worked well. No other recommendations other than trying to source reputable parts?
7294 is not available in good quality for you, or there is a fault which blows it up.
230 / 115 switching could have damaged other parts.
A transistor module costs like $9 here, and is easy to install, rather than repeatedly changing chips, I have no idea what you paid.
Your tracks are getting damaged, now the focus should be on a working amp, I think, rather than experimenting.
Your decision.
230 / 115 switching could have damaged other parts.
A transistor module costs like $9 here, and is easy to install, rather than repeatedly changing chips, I have no idea what you paid.
Your tracks are getting damaged, now the focus should be on a working amp, I think, rather than experimenting.
Your decision.
If the voltage was exceeded, the test should be started with the power supply - diodes (rectifier bridge) - electrolytic capacitors, stabilization circuits. Only after these checks can one reliable IC be established. Make sure everything works and then in order. Look at the markings on the board, especially under the blue wire. You may be able to find the manufacturer and schematic. Damaged tracks can be repaired with bare wire and glue.
I saw no indication of any problem on the power rails. The amp is powered by a bulky transformer that i suppose vaporized the problem before it became dangerous to the power section...