Tda2822m

[Dragon Tamer]

i tried making a bridge amplifier using a TDA2822M ic and the results were a little disappointing. My LM386 amplifier circuit had a higher output than the bridge amp. i can't figure out what i did wrong, i set up the circuit just like it showed in the data sheet. my problem is there is a lot of oscillation and the chip was getting really hot. it was running off of 12V DC, so i bumped it down to 9V DC, still a lot of heat, bumped it down to 5V not as much heat, but it's still a lot. I would like to know what i am doing wrong. Thanks

Vince

TDA2822M

 

[audioguru]

The datasheet for the TDA2822M shows a max supply of only 6V when it is bridged and has an 8 ohm load. Then its output is 1.35W into 8 ohms but an LM386 has an output of only 0.325W.
With an output of 1.35W into 8 ohms the chip temperature is about 95 degrees C which is very hot.

It will probably oscillate if made on a breadboard. It should be built using the pcb design shown in the datasheet.

 

[Dragon Tamer]

i decreased the voltage to 5 volts and the problem with the heat went away, i replaced the 10k resistor in the schematic with a 10k pot connected similarly to the 10k gain adjust pot in the 386 data sheet, and the problem with not enough gain went away. i also found the source of all the noise in the circuit, it was not caused by RF or feedback, it was caused by the 2 fans sharing part of the circuit (they did the same thing to the 386 amplifier i made). the problem is that when i build my circuits, i usual build them i bulk so there tend to be some power issues. witch is why i want to make my own high current power supplies. the problem is that everyone left me hanging on my post. they said that a 4.7mF capacitor was too small for the power supply, and they never told me how large i should make it, or how to calculate the size.

 

[Dragon Tamer]

now i have another question; the circuit that goes with this will run off of 4 AA sized batteries, with an additional 2 to 4 stacked onto it to give the circuit the ability to charge an iPod. to solve the problem of constantly wasting batteries, i intend to put a MAX712 battery charger in to charge these batteries (or cells if you want to get technical). my concern is that with the sensitivity of the TDA2822M, i am concerned that the MAX712 may destroy the chip with over voltage if i use a total of 6 cells (that would be the 2 additional ones). my concern isn't as serious with a total of 8 cells because there will be a total of 6V at the point where the TDA is getting it's power. should i be concerned with this iminent threat to cardboard-box-radio security or not?

 

[audioguru]

You forgot to say which type of AA battery cells you are using. I assume they are Ni-MH cells that are 1.5V when fully charged.
Then six in series produce 9V which is too high for a little TDA2822M bridged and driving an 8 ohm speaker. The amplifier IC will be fine with the 9V supply if you keep the volume turned down.

 

[Dragon Tamer]

the TDA2822M is getting it's power from 4 AA cells and the rest of the cells are for the sole purpose of charging my iPod. Here:

 

[audioguru]

Isn't 12V much too high for an ipod?

 

[mneary]

If you charge cells in series with a different load on some of them you will destroy them in a hurry. Some will over charge while the rest won't charge enough.

 

[Dragon Tamer]

12V is too much for an iPod, the 12V will be regulated to 5V using this circuit:

 

[Dragon Tamer]

If you charge cells in series with a different load on some of them you will destroy them in a hurry. Some will over charge while the rest won't charge enough.

so what you are saying is that if i try to charge the cells with the load attached, then i will end up destroying the load from over voltage? what if i had a voltage regulator before the regulator to prevent something like this from happening? that would still be hard to do though... what i need is a circuit that can sense when the charging has started and activates the regulator automatically somehow maybe some kind of transistorized sensor.

 

[Dragon Tamer]

i think that i've figured out how to get this to work. what if i used a circuit like this, then it should be able to overcome the sharp increase in voltage when the MAX712 is activated... right?

 

[audioguru]

What is the 8 ohm resistor for?
Aren'y you using the TDA2822M as a bridged amplifier that has two outputs to feed a signal to both ends of a speaker for much higher output power?

 

[Dragon Tamer]

yes, the switch is a DPDT switch. on one pole the switch changes the supply voltage to 9V, while the other pole with the 8Ω resistor is to increase the load to 16Ω so that there can be an increase in output power. i'm calling it the "public/private" function.

 

[audioguru]

Adding an 8 ohm resistor in series with the speaker reduces the power to the speaker and wastes a lot of battery power.
The TDA2822m IC will over-heat if it has a 9V supply, is bridged with an 8 ohm speaker and is playing loudly. Use 6V.

 

[Dragon Tamer]

according to the data-sheet this idea will work

 

[audioguru]

Where will you find a 16 ohm speaker? Use two 8 ohm speakers in series?
If you simply connect an 8 ohm resistor in series with an 8 ohm speaker then each get only 1W at a horrible 10% distortion or about only 0.8W each at clipping.

 

[Dragon Tamer]

well then what about 2 speakers in series? or using decoupling capacitors?

 

[audioguru]

The TDA2822M can be bridged and feed two 8 ohm speakers in series, but the TDA2822M can also be in stereo and each channel feeds one of the two speakers for exactly the same total output power. Then it is simple to switch from mono speakers to stereo headphones.

 

[Dragon Tamer]

I'm not 100% sure i understand what you mean. so don't take this as an offense, but I'm gonging to try my ideas regardless of what was discussed here (i want to see or hear rather, what the 10% distortion sounds like), fail, then inevitably come back here crying like a little girl because nothing worked. again it's not your fault, i'm just boneheaded, and curious, i hope we can still be friends.

 

[audioguru]

Semiconductor manufacturers turn the volume up too high so that the output power rating number is higher. The distortion is also higher but they want the power number to be very high. 10% distortion gives a power number that is 20% to 30% higher than when a sine-wave is barely clipping.

When the volume control is turned up way too much so that the output is square-waves then the output power is double the amount that has a sine-wave that is barely clipping. They advertise the phoney doubled number as "peak power or maximum power".