1
\$\begingroup\$

I found this 1W guitar amp design on the Internet:

http://www.diale.org/punch.html

schematic

And I am trying to build it on the protoboard. I could not find the IC TDA7052, instead I've got the a TDA7052B. According to the datasheet, the difference is that it has one extra pin for volume control (pin 4), which I connected to an 1M potentiometer, like this (and I am hoping that wire below is ground):

volume

I am using a 12V DC input.

Now, I can't see the difference of what I have built and the schematics, but clearly there is something wrong. The speaker output on open circuit is about 12V when volume potentiometer is at maximum, with no input. First that I believe that the mean DC voltage should be around 0, otherwise it will have a constant current drain. Second that, according to my calculations, to have a 1W power output at the 8 Ohm speaker, the peak tension must be at most 4 / sqrt(2) = 2.83 V. At the measured 12V, the power would be much greater. Indeed, if I turn it on, the IC will quickly get too hot. If place 270 Ohm resistance between the speaker outputs, it will get too hot, instead.

Here it is: protoboard

For someone with more experience in analog electronics, it should be obvious what is wrong. Can you please help me?

\$\endgroup\$
  • 1
    \$\begingroup\$ Maybe the IC does this, but I've often seen a decoupling cap between the speaker and driver. Constant DC output to the speaker will likely destroy the coil. Then again, the example circuit in the datasheet does not show this capacitor. \$\endgroup\$ – sherrellbc Jun 27 '15 at 19:19
  • 3
    \$\begingroup\$ The schematic looks like a bridge output. If so, no coupling cap required. \$\endgroup\$ – Brian Drummond Jun 27 '15 at 19:49
  • 1
    \$\begingroup\$ I have checked your breadboard, looks fine to me. What do you mean with "no input"? Can you disconnect C1 and see if the problem remains? It looks to me you are just getting a ton of interferences and/or the amp is auto oscillating. \$\endgroup\$ – Vladimir Cravero Jun 27 '15 at 19:54
  • 1
    \$\begingroup\$ No input means nothing pluged in the guitar jack. What about the 12V on the output and the heating? \$\endgroup\$ – lvella Jun 27 '15 at 20:12
  • 1
    \$\begingroup\$ Try grounding the input or removing C1 as suggested and see if the problem disappears. \$\endgroup\$ – Vladimir Cravero Jun 27 '15 at 20:23
4
\$\begingroup\$

You have a number of problems that I can see.

First, your grounds. You must learn that solderless breadboards, while convenient, are very bad for audio work. Using as short a jumper as possible (like, about 1/2 inch) connect pins 3 and 6. Then use a somewhat longer jumper to connect pin 6 to ground. Better yet, use 2 in parallel. Also, tie your buffer grounds to pin 3, not to the ground strip on your breadboard.

Next, decoupling. Connect your 0.1 and 220 \$ \mu {F} \$ capacitors directly from pin 1 to pin 6. Do not use these long wires, especially for the 0.1 \$ \mu {F} \$.

Third, your input AC coupling, C1. Do not use an electrolytic for this. Get a non-polarized 1 \$ \mu {F} \$.

Fourth. With no signal in, measure between pins 2 and 3. It should be zero. If not, figure out why your input buffer is messed up.

Fifth. Be aware that this is not a single-ended amplifier. That is, it does not have one side of the speaker grounded while the other varies. BOTH outputs will change with signal changes. In general, with zero input and a 12 volt supply, both outputs should be at 6 volts.

Sixth. Your voltage calculation is only partly correct. It calculates the RMS voltage to produce 1 watt RMS into 8 ohms. However, the actual swing must be multiplied by 1.414 to find the peak voltage, since for a sine wave of amplitude A, the RMS is \$ \frac{A}{\sqrt{2}} \$.

Finally, be aware that, since you are running 8 ohms/12 volts, your chip will run hot.

| improve this answer | |
\$\endgroup\$
  • 1
    \$\begingroup\$ What will happen, sound-wise, if I connect a 8.2 ohm resistor in series with the speaker? \$\endgroup\$ – lvella Jun 27 '15 at 20:48
  • 1
    \$\begingroup\$ You'll cut down on the high-frequency response. \$\endgroup\$ – WhatRoughBeast Jun 27 '15 at 20:59
  • 1
    \$\begingroup\$ OK, so you know your input is bad. My first guess is leakage in C1. If not that, you amp is bad. \$\endgroup\$ – WhatRoughBeast Jun 28 '15 at 1:19
  • 1
    \$\begingroup\$ @ivella - Oh, wait, I see what might be your problem. Get rid of R3. See farnell.com/datasheets/809105.pdf, figure 13 for a standard connection. In this case, don't bother measuring the input voltage. Just check the difference between the two outputs-it should be zero. \$\endgroup\$ – WhatRoughBeast Jun 28 '15 at 1:25
  • 1
    \$\begingroup\$ C1 is unplugged, as well as R3, pin 2 is completely free. But I'll try to measure the output difference without R3 to see what happens. \$\endgroup\$ – lvella Jun 28 '15 at 1:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.