TDA2822M 'custom' 1W audio amp problems

Thread Starter

-Excalibur-

Joined Jul 20, 2008
19
Hello everyone, this is my first time on this forum.:)

I'm building my first circuit, a 1W audio amp with the TDA2822M IC. The only modification I've made is swapping which pin the audio input connects to, i.e. 8 and 5 instead of 7 and 6, but I'm having some problems:

Note - DC in is 4 x 1.2V 1600 mA double A batteries in series.

1.) The pots are being used as variable resistors, though not in place of the normal resistors in the original TDA2822M circuit. The problem is that when I increase the resistance, the IC heats up beyond what seems appropriate, i.e. too hot to touch in around 10 s. When there is less resistance, all is fine.

2.) The amp seems to be in overdrive, I can't even turn it all the way up without distortion when the signal is as small as the leakage audio from my clock radio.

The speakers connected are 2 x 8 Ohm 27 mm.
Also, please ignore the red parts in the circuit diagram, they have been excluded from the circuit.

Any help greatly appreciated, and if you need more info, please ask and I will try and find it for you.

http://www.st.com/stonline/books/pdf/docs/1464.pdf - Original
http://www.e-dan.co.uk/electronics/TDA2822.html - Different version of original
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
Make sure that the output capacitors are good. If either are conducting, you could get some pretty high currents in the output of the amp.
 

Audioguru

Joined Dec 20, 2007
11,248
The amplifier will work perfectly if you connect its inputs properly as shown on the datasheet. The amplifier is not an opamp. Even an opamp won't work in your circuit.

The inputs pins 5 and 8 are part of the AC and DC negative feedback and must not be shorted to ground through your pots.
The real inputs must have a resistor to ground.

With your 4.8V supply then the output power will be next to nothing. But since your "speakers" are tiny then they will sound like little earphones.

The "Dan" guy that you copied his circuit doesn't have a clue about electronic circuits.
 

Thread Starter

-Excalibur-

Joined Jul 20, 2008
19
Thanks for that, I understand about the pots now, but I still don't get the thing about swapping the input - it didn't seem to make any difference when I swapped them back - is the amp meant to amplify such a small signal this much?
Do you think there is some sort of feedback occuring so that the amplified signal somehow gets back through the amp and goes round and round?
 

SgtWookie

Joined Jul 17, 2007
22,230
The amplifier IC you are using has internal feedback to control the gain of the amp.
It's set at 39dB gain. See the 2nd attachment - the internal feedback resistors are in the green boxes. Note that they are in the signal path from the output to the '-' (inverting) inputs, also green-boxed.
The '-' (inverting) inputs must be connected to ground via 100uF caps as shown in the datasheet, and your audio input must be to the '+' (non-inverting) inputs.

If it's still doing the same thing, you did not switch the inputs at the IC properly.
You will have to cut the traces on the PCB and solder jumpers, or re-make the PCB.

See the attached for the modifications you need to make around pins 5,6,7, and 8. I inverted it because it made more sense that way ;)

If there is a current path to ground from the inverting inputs (-) to ground, the gain of the amplifier will increase. Originally, when you decreased your volume setting, you were also lowering the resistance of the current path to ground of the inverting inputs. The lower the resistance to ground, the higher the gain of the amplifier until it was in "saturation" - it was amplifying as hard as it could, and it began overheating. This is very hard on the IC, and it may now be damaged.

My initial reply was hasty. AudioGuru tried to explain it to you.

For the amplifier to work properly, it must be connected as shown in the datasheet. Dan's circuit will do nothing but cause you to burn up amplifier ICs, and perhaps blow out your speakers.

[eta]
Wait a minute - I just went to Dan's page, and Dan's diagram had the IC wired correctly; he even had the layout from the datasheet! Apparently, it was you who arbitrarily decided to swap the inverting and noninverting inputs. Sorry, you can't do that ... as has already been explained.

[eta]
Also, you do not have the far side of the log pots connected to ground; instead you have R3 and R4, fixed resistors, connected to ground via the wiper of the pot. This will stress the pot. You should remove R3 and R4, and jumper the unconnected side of the log pots to ground.
 

Attachments

Last edited:

Thread Starter

-Excalibur-

Joined Jul 20, 2008
19
Sorry about the confusion - what I meant was that "Dan's" version just looked different, which is why I called the circuit in the attachment "custom", to signify that I was the one who changed it.

Anyway, I think I may just put the pots on the outputs - I tried this, and nothing heated up when on full resistance - as for the input, I might just put fixed resistors on all 4 inpits (experiment) just to save making to much of a mess or remaking the PCB. If it doesn't work, I will have to remake the PCB :( following your advice.

I haven't actually done electronics before - this is my first time, so please excuse me if I seem like a noob. (I've allready remade the PCB once :eek: )



EDIT: Actually, mabye the cutting and rerouting may not be so messy - I should view attachments before posting, as for the internal IC shematic, it will be a while before I totally get that, but no need to explain yet.

And thanks again for your help. :)
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
Don't feel too bad - everyone makes big mistakes when they start in electronics. I've made worse mistakes. :rolleyes: ;)

Really, the cuts and jumpers necessary aren't too bad. If you follow the cuts/jumpers, remove R3 and R4, and ground the floating end of the pots, you should be golden.

It's a learning experience. I'm sure that your next board will be a lot better. Make the most out of this one though - fixing a board that has problems is a great learning experience.

If you run into problems, let us know. If you get it working properly, let us know too. Just add to the thread ;)
 

Audioguru

Joined Dec 20, 2007
11,248
Don't put pots in series with output of a power amplifier. The speakers need to have their resonances damped by the extremely low output inpedance of the amplifier.

The volume control belongs at the input of the power amplifier as shown by its inventors on the datasheet.
 

Thread Starter

-Excalibur-

Joined Jul 20, 2008
19
Ok, thanks - I think I will just follow your instructions - I was feeling sort of crap yesterday, but good today...:)

EDIT: Oh, and the board looks good, just some of the things are in the wrong spots :)
 
Last edited:

Thread Starter

-Excalibur-

Joined Jul 20, 2008
19
I've got it working:), now all I need to do is finish off the enclosure it's going to be in. And I'll put one 1.2K r's (4, one each for L/R, one for each ground, since there are 2 due to the pots) on each input, so I don't have to listen to the carrier wave of the input...

Thanks for all your help.
 

SgtWookie

Joined Jul 17, 2007
22,230
I've got it working:)
Good deal! Thanks should go mostly to AudioGuru, who hit the nail on the head with the initial diagnosis. I just explained what needed to be cut & jumpered. ;)

now all I need to do is finish off the enclosure it's going to be in.
All successful projects need a suitable enclosure :)

And I'll put one 1.2K r's (4, one each for L/R, one for each ground, since there are 2 due to the pots) on each input, so I don't have to listen to the carrier wave of the input...
Now, this part I don't understand.
If you basically followed Dan's schematic and used a 50k audio taper pot on each input, that should be all you need.

I've already indicated that your R3 and R4 were a mistake to add; apparently you were trying to compensate for the missing ground on the low side of the pots.

On your board, you have audio in routed to one end of the pot. The wiper of the pot (the middle connection) should be going to the non-inverting input of the audio amplifier - if you've cut & jumpered it correctly.

But when you made the board, you didn't provide a path to ground for the low side of the pot. This will cause the volume controls to not work very well. You must ground the low side of the pot in order for the circuit to work properly.

There is still the possibility that your volume controls will seem terribly sensitive; for a very small change in position, the volume increases a great deal, and only near one "end" of the travel. If this happens, it means that you have installed the pot with the ends reversed; ie - the taper is backwards. The only proper fix for this is to unsolder the connections to the pot's 1 and 3 terminal connections (the ends) and swap them.
 
hello! i got here because i had a computer speaker set, and i wanted to lookup the IC number. then i found this thread. so i took some pictures of the circuit (sorry for .3 megapixel quality, darn you nintendo) and its acually a pretty simple circuit. if you need to see what one looks like, here they are: http://www.mediafire.com/?2jf957zzs6w2ca7 http://www.mediafire.com/?y47td4yidijthve sorry i had to use mediafire... but my power suppy is like 9-10 volts, your 4.5 really isnt gonna cut it
 
Top