Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

My IC getting hot !

Status
Not open for further replies.

learning

Member
After completion of components adding to my TDA2822M PCB when I power on IC and it is getting hotter very quickly as I can't understand its reason. As power supply(6v/500ma) with unregulated is applied and component values(approx.) are as par as datasheet recommendation. Can you tell me how I can solve it ?
 
Hi,

your amplifier circuit is most probably oscillating if you have connected the power source correctly.

Here is a PCB layout suggestion from ST.

Go with that and take good care of cooling, either by large copper areas around the ground pins or a clamp-on IC heat sink.

Boncuk
 

Attachments

  • TDA2822.gif
    TDA2822.gif
    30.5 KB · Views: 2,045
As we have told you many times, you should use the recommended pcb design in the datasheets of power amplifier ICs to avoid oscillation.

You forgot to say if your amplifier is a stereo amplifier with an output per channel at clipping of only 0.3W into 8 ohms, or is a bridged mono amplifier with an output at clipping of 1.1W into 8 ohms.
As the stereo amplifier, its heating when both channels are producing 0.3W is about 0.6W which is fairly warm because its max allowed dissipation when it is extremely hot is a little over 1W.
As the bridged mono amplifier, its heating when the output is 1.1W is 1W which is almost its max allowed dissipation so it will be extremely hot.
Most music is not played at full power all the time so the amplifier should be much cooler than above.
 
It's especially important that C3 be connected with short connections between pins 8, and pins 4, 5, 12, & 13 of the IC.
 
Hi learning,

here is the circuit according to ST suggestions. I placed all parts exactly as shown in the application and enlarged the ground area for sufficient cooling (if you don't use a heat sink).

If you make the PCB double sided you don't need any vias, but have the entire top layer as cooling area.

The application has a minor flaw. C6 is a dual part, once as non polarized capacitor of 100nF, connected to the left channel output and once as electrolytic cap, connected to signal ground and the the positive terminal connected to pin2 of the amplifier IC (which is not connected internally). I omitted that one.

Boncuk
 

Attachments

  • TDA2822-SCH.gif
    TDA2822-SCH.gif
    16.1 KB · Views: 2,923
  • TDA2822-GND.gif
    TDA2822-GND.gif
    27.7 KB · Views: 5,220
  • TDA2822-GND-TOP.gif
    TDA2822-GND-TOP.gif
    21.3 KB · Views: 752
  • TDA2822-SILK.gif
    TDA2822-SILK.gif
    16.4 KB · Views: 1,236
Last edited:
Hi learning,

here is the circuit according to ST suggestions. I placed all parts exactly as shown in the application and enlarged the ground area for sufficient cooling (if you don't use a heat sink).

If you make the PCB double sided you don't need any vias, but have the entire top layer as cooling area.

The application has a minor flaw. C6 is a dual part, once as non polarized capacitor of 100nF, connected to the left channel output and once as electrolytic cap, connected to signal ground and the the positive terminal connected to pin2 of the amplifier IC (which is not connected internally). I omitted that one.

Boncuk

Which software you use to design it ?
 
Learning said he is using the TDA2822M that is in an 8-pins case, not the TDA2822 that is in a 16-pins case.
 
Learning said he is using the TDA2822M that is in an 8-pins case, not the TDA2822 that is in a 16-pins case.

Sorry for the misunderstanding. I downloaded the data sheet of the TDA2822, which is manufactured in a 16-pin power DIP package, with pins 4, 5, 12 and 13 connected to ground for cooling purposes.

Of course, pins 2, 7, 9, 10 and 15 are unconnected.

An 8-pin DIP package with no extra ground pins will of course run hot quickly if no heat sink is used.

I'll make another design.

Boncuk
 
Last edited:
Even if it is oscillating, it shouldn't be that hot...post a high resolution picture of your actually assembled pcb.You might have done something really stupid. We can track it out on image. This is such a simple circuit that you can assemble it neatly within seconds on a general purpose pcb provided bit of concentration during work:)
 
Learning didn't say the impedance of his speaker and didn't say if the amplifier is stereo or bridged mono.
With a 6V supply and an 8 ohm speaker, a bridged TDA2822M amplifier has an output power at clipping of 1.1W and its heating dissipation is 0.95W. A stereo amplifier has the same total output power and the same heating when it drives 4 ohm speakers.
It is allowed to dissipate 1W when its ambient temperature is 50 degrees C.

Maybe the ambient in India is 50 degrees C so the chip is pretty darn hot if the ampifier is playing at full blast all the time.
 
He told his amp is getting too hot within seconds after switched on. That means there's a serious assembling problem.

That is in theory. There's a stereo computer speaker what I've seen has TDA2822M running at 9V or 12V(don't remember). It never gets that hot at normal volume levels. Just warm thats all and I don't think this IC ever needs a heatsink as it is not designed for that.

BTW, temperature in India(0-40C and 25C average) is pretty good for normal working of all electronic appliances and circuits. Below 0C is the real problem suppose the behaviour of rechargeable batteries, you charge it today and tomorrow no charge :p
 
Here is the alternative "boiler" circuit.

Board dimensions are 1.76X1.425".

Transistor495, an oscillating circuit can cause a chip to get frying hot.

Boncuk
 

Attachments

  • STEREO-AMPL-I-SCH.gif
    STEREO-AMPL-I-SCH.gif
    12.6 KB · Views: 657
  • STEREO-AMPL-I-BRD.gif
    STEREO-AMPL-I-BRD.gif
    25.6 KB · Views: 1,239
Last edited:
A couple of days ago it was -19C in my city in Canada. Yesterday and today it is 0 degrees which is close to normal in winter.
Ni-MH and Li-po batteries work fine in the cold temperature. My 5 years old car battery cranked my car's engine a little slower than normal so maybe I should replace it next year.
 
Nice work Boncuk, but should C3 be closer to the chip? Andy

Hi Andy,

you're welcome to improve the board layout. I just used the one published by ST (SGS-Thomson).

+UB is forced to flow via C3 towards pin2 of the amplifier (the one and only current consumer within the circuit) and is surrounded by ground from start to end.

The ground trace forms a closed loop around the entire board and using the ground fill the amplifier should work without any problem from the board design - good enough for government work anyway. :)

Here are some screenshots.

Boncuk
 

Attachments

  • UB-trace.gif
    UB-trace.gif
    26 KB · Views: 537
  • GROUND-TRACE.gif
    GROUND-TRACE.gif
    26.3 KB · Views: 803
  • GROUND-FILL.gif
    GROUND-FILL.gif
    27.5 KB · Views: 549
Last edited:
you're welcome to improve the board layout.
Na it should work fine, and I have no need for it. I just thought a shorter power path and supresson cap would be better but I doubt you would ever hear the difference.
Andy
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top