0
\$\begingroup\$

I am new here and I realy need your help. For diplome at the end of my studying I am making the audio amplifier module (power amplifier, preamplifier, tone control, power supply, linear regulator) all in one. A have almost everythink work yust can solve noisy hum in preamplifier (TL082) that is caused by DC power supply ripple. I put pi filter CRCRC (1k Ohm, 20 uF) before linear regulator to filter DC voltage from power amplifier and get pretty good symetrical voltage, but with 100 Hz noise at both voltages. How can I get rid of that 100 Hz ripple?

I am using LM337 and LM317 to voltage drop from 40 V to 18 V and that ripple also copy to the output of the linear regulator and causes hum in sound. Just for information, the power amplifier work great alone and also preamplifier work great with external DC power supply from professional power supply device.

enter image description here

enter image description here

\$\endgroup\$
  • 5
    \$\begingroup\$ "noisy hum in preamplifier (TL082) that is caused by dc power supply ripple" What brings you to this conclusion? Also, show schematic. \$\endgroup\$ – winny May 30 '18 at 12:42
  • 4
    \$\begingroup\$ 1 megohm series resistors before the regulator? I doubt that very much. A schematic is really needed. \$\endgroup\$ – Elliot Alderson May 30 '18 at 13:04
  • 2
    \$\begingroup\$ Show your schematic. Have you followed the datasheet recommendations on the LM337 output caps in particular? \$\endgroup\$ – Spehro Pefhany May 30 '18 at 14:24
  • 1
    \$\begingroup\$ 1 ohm and 10,000 uF have 0.01 second tau, or 16 Hertz F3dB, giving about 2 octaves of filtering of 60Hz ripple, or only 12dB attenuation. If you can have 10 ohms in the preamplifier VDD path, you'll get 20dB (10X) more ripple reduction, or 32 dB total. \$\endgroup\$ – analogsystemsrf May 31 '18 at 4:02
  • 1
    \$\begingroup\$ It sounds like you might have an earth loop. This is typically caused by mains earth connections on each device followed by an audio GND connection between the pre-amp and the power amp. See if you can disconnect the screen on the input to the power amp (if there is GND connection via the mains earths). \$\endgroup\$ – Transistor May 31 '18 at 16:46
0
\$\begingroup\$

The ground may well be the issue. But since that's layout dependent I have a couple of other suggestions.

Parallel each of the rectifier diodes with 10nF. And isolate the supplies from the power rails with diodes (one for the positive, one for the negative). And make sure there is only one ground wire from the preamp supply to the rest of the circuit.

\$\endgroup\$
  • \$\begingroup\$ I am curious about that unfinished line from Gnd to preamp. Is it really not there? And if it is there then I am pretty sure there will be another one from preamp to power amp somewhere down the line, since they are drawn as independent modules. \$\endgroup\$ – Maple May 31 '18 at 18:08
0
\$\begingroup\$

I solved the problem. Hum from preamplifier has been partly removed by RC filter on the linear regulators input (1000uF and 1000kOhm) on both voltages and the other part of hum was removed by removing grounding loop. Problem with grounding was because i took ground just after power supply circuit before power amplifier, but I should take it after power amplifier like i did now and work perfect. Thank you for all advices.

\$\endgroup\$

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.