Problems using a TDA7265 amplifier

[screwdriver]

Hello, I'm trying to make a simple stereo amplifier for my computer, I decided to use a TDA7265 because of availability but after having tried with 3 different chips (I mean 3 TDA7265) the thing doesn't work. I can't imagine what's happening anyone could help?

When I connect the TDA7265 gets hot very soon even if there's no load and makes a humm in the loudspeaker but nothing like sound.

The circuit I'm building is the one that comes in the datasheet (**broken link removed**) I include here an image of the circuit built and the board:

Thank you for your help.

 

[Optikon]

Hello, I'm trying to make a simple stereo amplifier for my computer, I decided to use a TDA7265 because of availability but after having tried with 3 different chips (I mean 3 TDA7265) the thing doesn't work. I can't imagine what's happening anyone could help?

When I connect the TDA7265 gets hot very soon even if there's no load and makes a humm in the loudspeaker but nothing like sound.

The circuit I'm building is the one that comes in the datasheet (**broken link removed**) I include here an image of the circuit built and the board:

Thank you for your help.

Can you provide a schematic of what you actually built? Is it identical to the one shown in the datasheet other than your mute modification and power supply bulk capacitance?

Are you using + and - 20V supplies? or something else?

Measure the voltage on pin #5, what is it?
IF it is between -3 and -6 volts,(with+/- supplies) it would explain why the part is hot and you get no sound output. Under this condition, Iq is 70 mA and you have 70dB of attenuation.. according to the datasheet.

A similar range exists using a single supply.. Take your supply voltage (+Vs) and look up in the datasheet the Mute / standby thresholds relative to +Vs... make sure you are in neither range. If you are, it is doing exactly what it is supposed to...

 

[screwdriver]

The schematics is the same as the one shown in page 7 of the datasheet, the differences in values are these:

Mute circuitry substituted by a 10k resistor to ground (on pin 5)
R5, R8 20k
R6, R9 520 ohm
This shoud give a gain of 32 dB

the rest of components are the same values as stated on page 8.

I measured the tension at pin 5 and it was 0V, acording to page 6 that should be an "on".

Tha power supply is a +/- 12 V, I started with a +/- 15 but as it was idle it went to +/- 26V so I reduced tension not to damage the IC.

 

[Optikon]

The schematics is the same as the one shown in page 7 of the datasheet, the differences in values are these:

Mute circuitry substituted by a 10k resistor to ground (on pin 5)
R5, R8 20k
R6, R9 520 ohm
This shoud give a gain of 32 dB

the rest of components are the same values as stated on page 8.

I measured the tension at pin 5 and it was 0V, acording to page 6 that should be an "on".

Tha power supply is a +/- 12 V, I started with a +/- 15 but as it was idle it went to +/- 26V so I reduced tension not to damage the IC.

Personally, I think you need a bigger heatsink for this part. Hav eyou calculated the maximum power dissipation for this design running on +/-12V? The app note shows location for amuch bigger heat sink albeit running at +/-20V but you should check to make sure you are ok with the one you have.

Also, I noticed in your layout, you placed the smaller bypass ceramic capacitors all the way up at the input where the power comes in. These should be placed right at the pins of the amplifier or they will be ineffective. With that said, you ought to install the larger capacitors and also just skywire on some more of the smaller bypass caps right at the amplifiers power pin. Do this to help eliminate the possibility that you are oscillating through the power supply pins.

Have you verified that your circuit is not oscillating with a scope?
If you have no way to verify it, you could try reducing the gain from 32dB to say 20dB and see if that changes anything if you suspect oscillation.

Also, because the datasheet shows pin 5 voltage of 0V has 115dB attentuation and it doesnt show it with respect to +vs.

Try pulling the voltage more towards your -12 V. Make a voltage divider with ground and -12V that puts your pin 5 voltage at say around -11 or -10 volts with respect to ground.. Better yet, clip that pin and hook it up to a floating supply and vary it over the range and see if you can get a different output.

 

[screwdriver]

Thanks for your suggestions I'll try tomorrow, Ive been trying to play with pin 5 tensions and at least "something happened": with a voltage divider I punt pin 5 to +Vs and got "low hum" with these tensions: pin 5 11 V, +Vs 11 V and attention -Vs -4,6 V.

When changed pin 5 to half way between +Vs and ground (1k on each side) I got these tensions: pin 5 1,2 V; +Vs 2,6 V; -Vs -1,7 V and the hum much louder.

There's also an intermediate point between the 2 states in the 5 with intermediate hum, I didn't measure the tensions.

I have no oscilloscope so I can't check the circuit oscillating.

Concerning the heatsink, this is a test assembly but anyway I don't intend to use this amplifier at high power so I don't expect it to heat very much, anyway I'll check later.

 

[audioguru]

Of course the amplifiers are oscillating at a very high frequency, without any power supply decoupling capacitors right at the IC. Those power supply connecting wires have significant inductance at very high frequencies. Voltage regulator ICs have a similar problem.
Also, that little 3.5VA transformer can barely supply the IC's idle current, let alone any more current for any output power.

 

[chemelec]

Having built Numerous power amps using these types of chip (But Not this particular one), I have found PCB Layout is important, So are Bypass Caps.

I would suggest you use the Layout they gave!

And DEFINATELY a BIGGER heat Sink.

Gary

 

[Optikon]

Thanks for your suggestions I'll try tomorrow, Ive been trying to play with pin 5 tensions and at least "something happened": with a voltage divider I punt pin 5 to +Vs and got "low hum" with these tensions: pin 5 11 V, +Vs 11 V and attention -Vs -4,6 V.

When changed pin 5 to half way between +Vs and ground (1k on each side) I got these tensions: pin 5 1,2 V; +Vs 2,6 V; -Vs -1,7 V and the hum much louder.

There's also an intermediate point between the 2 states in the 5 with intermediate hum, I didn't measure the tensions.

I have no oscilloscope so I can't check the circuit oscillating.

Concerning the heatsink, this is a test assembly but anyway I don't intend to use this amplifier at high power so I don't expect it to heat very much, anyway I'll check later.

Well, while you are debugging the thing, slap a big heatsink on it just so you wont have to think about over-heating damage until you get it running the way you intended... its one screw..go ahead and splurge...

 

[screwdriver]

After reading your posts and rethinking my project I decided not use the TDA7265, I want to do decent amplifier for a computer, I'd like to avoid the power supply but then I'd have to feed it with +12V.

My tests with TDA7265 have all gone wrong and I understand nothing about its behaviour I can only think the chip its damaged it seems to be in some sort of short circuit (or perhaps it's oscillating even with the capacitors soldered to chips legs).

So I changed to TDA 2030, I made exactly the circuit in the datasheet http://www.st.com/stonline/books/pdf/docs/1458.pdf and it worked beautifully well from the start. With a portable CD player and my big loudspeakers I get quite good sound.

Now the problem: when I put a connecting wire to the computer it starts to hum loudly, in fact it gets all the noise when a wire is connected to the + input, no matter which power supply I use (transformer or switching power supply from the computer).

I imagine this problem is very easy to solve but have no idea how to: input cable disconnected, or connected to the computer: big hum, input cable connected to the portable CD player: absolute silence or music when playing.

I show you a pic of the new amplifier connected, yes the cabinet in the pic is the computer, you can see it here: **broken link removed**, I made that small loudspeaker for the comp, it sound much better computer loudspeakers do.

Thank you for your help

 

[audioguru]

Hi Screwdriver,
Are you powering the TDA2030 with only a +12V supply? That's its minimum voltage rating at which it produces very little output power.

Its big hum problem is probably caused by the circuit's high voltage gain of 33 times, and the high input impedance of the circuit. You can't reduce the circuit's voltage gain much, because the TDA2030 is spec'd for a minimum gain of 24dB, which is about 15 times. You need a voltage gain of only about 6 to 10 times if fed from the low impedance "headphones" output of your portable CD player.
So why don't you make a low-impedance voltage-divider for the amplifier's input, 100 ohms in series with about 15 ohms, and the 15 ohm resistor grounded at the amplifier and input cable's shield. The amplifier's input is from the junction of the two resistors and the CD player's "headphones" output feeds the 100 ohm resistor. Then you should hear absolute silence when the amplifier's input cable is disconnected.

 

[screwdriver]

After reducing gain to ~24 dB, shielding the wire and changing the input divider from 10k to 1k (I didin't have a smaller one) the improvement has been very big but there's still noise some coming out, for example, every time I move the mouse I hear a noise. I'll keep trying.

thank's a lot

 

[grrr_arrghh]

there is a common issue where power supplies make noise when (for example) the mouse is moved. Try using a completly independant power supply and see if that helps.

 

[screwdriver]

grrr_arrghh, using an independent power supply is what I'm trying to avoid

 

[grrr_arrghh]

fair enough, but it would eliminate another possible problem. If it worked fine with a separate power supply, you know it is definately a power supply problem.

 

[screwdriver]

Sorry grrr_arrghh I had not understood your intention, I did the test supplying with a moto battery and the results were the same. The pity was that I misconnected the wires and I broke a lot of things (computer power supply, amplifier...). Finnally I got everything fixed and I made an amplifier based on TDA2030 with 2 circuits just like the one in the datasheet. It's not working very well, it still has the noise from the mouse and some other hum. The cable to connect it to the computer has the shield earthed at the amplifier end only. In the tests I did it didn't seem to changed anything if I earthed the shield on on end, the other, or both.