Class D Amplifier Troubleshooting

Status
Not open for further replies.
I have replaced the TDA7294(which was smoked), the 2 ceramic resistors(both cracked) and the pair of Power Transistors (1 of the matched pair was bad) and eliminated the dead short. Still blows the spec'd 2Amp fuse as soon power is applied. I have put a 5amp fuse in and the amplifier now works and sounds great, for about 2 minutes but overheats and clips and I shut it off before hurting anything. I am suspecting the transformer, any ideas on this? This unit has always had an issue with blowing the 2amp fuse, I suspect a faulty transformer, but have no schematics for this discontinued sub
 

Attachments

  • 20180119_222048.jpg
    20180119_222048.jpg
    404 KB · Views: 283
The TDA7294 isn't class D, that's an A/B amplifier. Could a moderator please move the thread and adjust the title?

The boards got their own rectifier and capacitors. Did you check them both?
 
My apologies, for the wrong designation. The capacitors have no sign of heat issues and hold a charge. Above that I haven't check'd above that. As for as the rectifier, I am going to have to look up how to check it. Before I did the replacement parts, the transformer had high voltage at the amp, and a low voltage to the signal board. Only after I changed the paired power transistors was I able to get output to the woofer, but at this point, I also am assuming I am pulling more current since it pops the 2amp fuse, but not the 5amp
 
Sounds like a TDA7294 with extra output power transistors.
You need to check output transistors and big resistors are ok.
If they have gone its possible the 7294 has gone too.
The 7294 is fun to de-solder !
You can try running the 7294 with no output transistors and see what happens.
If it still blows the 2 amp fuse the 7294 is likely gone.

The 7294 can pull a few amps certainly more than 2 amps.
 
Last edited:
The TDA7294 is new, and seems to work. It was shorted, and this fixed the short when I replaced it. Thats when I removed the power transistors and found the 2SA1493 bad. Replaced the pair and then I got signal output, but like I said, now it fries anything less than a 5amp fuse, Is it possible to check the rectifier in circuit? I can pop it out right quick if not.
 
I took out the bridge rectifier, it checks out ok and within range. I found a loose connection coming from the main board to the amp... this seems to have fixed the blowing the 2amp fuse issue. I can still play the sub when cool for about 3 minutes before the TDA7294 heatsink gets really hot and still clips, but now I have no shorting issue and can use the 2amp recommended fuse. I checked the transformer also, 34VAC on each leg going to the rectifier. The caps also are good. Any help is appreciated.
 
Voltage High...

34V+_ does not seem high if I am reading it correctly @4ohm speaker. What I am wondering is could it be possible to switch the legs of the transformer to cause this overheating issue? I have been studying other pcbs and it looks like it could cause an issue. Any thoughts?
 
I believe chermann has got a very valid point.
The way you write 34VAC it means 34V RMS (AC) at nominal loading. Then, you can add some 8-10 percent when unloaded. The rectifier will leave a drop of about 1.3V but the peak voltage (x1.4) will be just below 50 Volt! The rectified (DC) voltage will be almost the rectified peak voltage with little loading and that is simply too high a voltage for an operating TDA7294 chip.
Only the TDA7293 can handle 50 Volt.

You can check with a voltmeter but preferably without the amplifier connected.
 
Last edited:
I believe chermann has got a very valid point.
The way you write 34VAC it means 34V RMS (AC) at nominal loading. Then, you can add some 8-10 percent when unloaded. The rectifier will leave a drop of about 1.3V but the peak voltage (x1.4) will be just below 50 Volt! The rectified (DC) voltage will be almost the rectified peak voltage with little loading and that is simply too high a voltage for an operating TDA7294 chip.
Only the TDA7293 can handle 50 Volt.

You can check with a voltmeter but preferably without the amplifier connected.

Ahhh, so basically, I'm back to my original thougths that the transformer is my issue? If I am reading your post correct, I would be looking more at 24V for proper operation. The more I dive into this, am I correct in saying the TDA7294 amp circuit is a 100W circuit, and with the "bootstrap" power transistors is where they brought in the extra 50Watts?
 
My calculation:

A TDA7294 can handle 40V. Let's reduce the output voltage swing with 4V so the peak voltage the TDA7294 can deliver (with a 40V supply) is 40V-4V=36V. 36V peak means 36V/1.4= 25V RMS.
25V RMS from the TDA7294 leaves the following power in 4 Ohm speakers: (25V x 25V) / 4Ohm = 156 Watt! That would require a bit less than 7 Amp RMS but 10Amp peak! The TDA7294 has 10Amp (peak) as the absolute maximum rating (stay well below if you don't want to roast your chip!).

Hence, the TDA7294 in itself can easily deliver 100W in 4 Ohm with a supply voltage of 40V. It just doesn't like almost 50V!

Let's calculate backwards to estimate the transformer voltage you need (I will disregard the rectifier diode drops): 40V peak means 40V/1.4 = 28V (idle RMS). But, that is unloaded. So, to find the approximate transformer voltage when loaded (to nominal current) we divide by 1.10 (assuming a 10% drop in voltage) -> 28V/1.10 = 25V RMS.

A 24V-0-24V transformer is a good choice. You should have 150VA minimum capacity from the transformer for each channel.

Alternatively, you could use TDA7293. But as we do not know if there are other restrictions in the circuit, I would stay with the TDA7294.

Whenever they promise you XXX Watt, it depends a lot on the speaker impedance. Half the speaker impedance doubles the power (if the amplifier and power supply can handle the current). Your amplifier seems to have an extra current buffer stage such that the current margin is improved. Actually, as the TDA7294 heatsink is rather small, it leaves the impression that the TDA7294 circuit is only used as a driver circuit (but still with the limit in supply voltage).
 
Last edited:
...thx to Fauxfrench...you are an expert.😉

Please take in your consideration that your heatsink is too small!!!

its like this: 22K/W
V 5640B: Spezial-Kühlkörper, 33x25,4mm, 22,5K - W, lötbar bei reichelt elektronik


the recommended condition according to the datasheet (see post 5) on page 3 is 1.5 K/W. that means your actually heatsink force the chip to overheat.
that the size you need to get the datasheet 🙂
V PR173 - 100: Profilkühlkörper, 100x100x40mm, 1,3K - W bei reichelt elektronik



i have a similar chip 7297. i did some load test and the original heatsink is a "joke" - at 4 Watt both channels it get after 5min --> 72 °C.

DROK(R) Mini Verstarker Powered Stereo: Amazon.de: Elektronik


chris
 
Last edited:
Thanks guys, I really appreciate the help with the formulas and knowledge. I have ordered a 24v 0 24v transformer to see if this fixes the issue. I also agree this heatsink is way to small it seems, the designer felt like the bootstrap part of this design which is directly attached to the metal casing was more than enough to cool the internals... but doesn't address the tda heat at all IMO
 
Status
Not open for further replies.