1
\$\begingroup\$

I'm working on breadboarding this amplifier circuit before transferring to a perf board.

enter image description here

When we power it from an adjustable DC power supply set to 12V, the audio output sounds clear. When we switch to a 12V 5Amp AC-DC adapter, we are getting almost nothing but noise.

We've already switched from one adapter which was providing an uneven output between 10V and 12V, to an adapter which we've measured at a constant 12.6V.

We're using the negative power terminal as our ground.

I would appreciate some suggestions for ways to troubleshoot this issue or any ideas as to why we're only getting clear sound from the adjustable power supply.

\$\endgroup\$
  • 3
    \$\begingroup\$ Can you describe the noise or (even better) show a 'scope plot? \$\endgroup\$ – brhans Mar 13 '17 at 16:41
  • 2
    \$\begingroup\$ Your +12/-12 supply is 24V which exceeds the chip's maximum 18V power supply rating. \$\endgroup\$ – peufeu Mar 13 '17 at 16:50
  • 2
    \$\begingroup\$ Is adjustable supply grounded? Does grounding source help? \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Mar 13 '17 at 16:51
  • 1
    \$\begingroup\$ May not be related but that input cap seems VERY large and should not be electrolytic. \$\endgroup\$ – Trevor_G Mar 13 '17 at 21:34
  • \$\begingroup\$ There is something wrong with this schematic. It shows -12V connected to ground. it's either one or the other (or a blown fuse/power supply). I guess you mean 0V not -12V, from your other words. \$\endgroup\$ – danmcb Dec 30 '18 at 8:17
0
\$\begingroup\$

Based on your description, I am assuming your schematic should be reading +12v and 0v, rather than +/-12v which, as mentioned above, would yield a total of 24v.

I am also going to assume that the noise is a buzz or a hum, rather than a hissing, or white noise.

If that is the case, the first thing I would suspect is that the wall adapters are switching power supplies, and your adjustable supply is linear. From my understanding, certain amplifier circuits can be more sensitive to the noise generated by the switching action, whereas linear supplies do not produce such noise. So, I believe you may also be amplifying the noise generated by your adapter.

\$\endgroup\$
  • \$\begingroup\$ Sorry, didn't realize that diagram had +/- 12. We're using just +12V in our actual circuit. We get noise from the adjustable supply in the form of a light buzz. Our 12V wall adapter just gives us a constant loud buzz and nothing else. \$\endgroup\$ – RockHopper Mar 13 '17 at 21:09
  • 1
    \$\begingroup\$ OK. I am not as well versed as many others on this site, but it might be a combination of factors. The difference between the two supplies tells me that it's worse because of a switching supply, but there may be a grounding issue as mentioned above. I've used a TDA amp in the past and they are noisy by nature. I scrapped it. Can you make a 12v battery to test? If the light buzz remains, I'd say it's the amp. Last thought. Is there fluorescent lighting on the same mains circuit? An audio engineer once told me to keep audio away from fluorescent lighting. I never tested the theory. \$\endgroup\$ – Jay Mar 13 '17 at 22:09
0
\$\begingroup\$

looking at the datasheet, your dc power supply is not sufficient.enter image description here

I would size it about twice the non repetitive peak Amperage for power headroom, but twice its repetitive peak output would be my recommended minimum.

If you go by typical practices, the maximum current draw should be 80% of the power supply so the rectifier filters in the power supply can recover.

But that is only the 1/3 of the issue, the main issue it has is its design. Unbalanced = 100% of power supply noise injected onto the signal.

Now if you create isolated power, or create a ground reference that has the inverse of the power supply noise on the ground, the noise would get subtracted by the output circuit.

\$\endgroup\$
  • \$\begingroup\$ All of the values on that sheet are maximums, not typicals. Just because the amp can cope with that much current doesn't mean it's drawing it in this application. \$\endgroup\$ – Harry Beadle Sep 2 '19 at 14:11
  • \$\begingroup\$ There are sever factors in play, but in designing, your maximum possible currently draw should be 80% or less than what the power supply can deliver. Mainly because the effects it has with the DC filtering in the power supply, But to makes matters worse is the system is unbalanced which is the major cause of the noise. \$\endgroup\$ – drtechno Sep 18 '19 at 15:10
0
\$\begingroup\$

Could be several things:

1) The wall adapter is junk. Older SMPS are known to be noisy. Newer SMSP like Mean Well LED supplies are known to be very good and can be very quiet.

2) Even if the SMPS is the modern good variety, they may not "like" large capacitors directly on the output. The ideal scenario is to use a common mode choke that is as large as you can find for the current required and then follow that with whatever size cap you need to filter down below 20kHz. For example, you should be able to find at least a 1mH choke which in combination with 10uF cap will filter out everything above ~1.5kHz.

3) Even if the SMPS is the modern good variety, they may throttle switching which causes moduation at low frequencies if the output is not loaded with at least 10-20% of the load capacity. So you might need to just add a smallish resistor of appropriate wattage. But unless the SMPS is very large, my guess is that the quiescent current of your amp is enough to load properly.

So my guess would be that the issue is 1 - the adapter is some junk pulled out of a bin. There is no recovering from that. You can get a proper 24V (or 12V) Mean Well constant current LED supply for under $20 USD on Mouser.

\$\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.