0
\$\begingroup\$

I have a homebuilt subwoofer for my surround sound audio system. 400W constant, 900W peak. However, there appears to be an AC hum on the input line, and it becomes especially noticeable when I connect a preamp inline at the subwoofer end of the line. As in, house shaking 60Hz hum. That hum's gotta go.

Is it possible to remove the effect of a ground loop by feeding an equal amplitude, ~180º out of phase copy of the loop frequency? I've been thinking about this and it seems like there would be some feedback or something.

Would it work to use this schematic? If not, please correct it.

schematic

simulate this circuit – Schematic created using CircuitLab

And I'd love to have a simpler solution for active ground loop/AC hum elimination if there exists one.

\$\endgroup\$
7
  • 3
    \$\begingroup\$ Better to eliminate the source of the ground loop instead of trying to suppress it. Could you post a schematic for the subwoofer amplifier? And a diagram of how your audio system is wired together? \$\endgroup\$
    – uint128_t
    Commented Feb 13, 2016 at 19:58
  • \$\begingroup\$ The amp for the sub is a Bryston 3B (from like the 80s or 90s, so it's old). It's a discreet components amp that doesn't have a hum on its own. Would you recommend snipping the ground prong on it to test? \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 20:39
  • \$\begingroup\$ i.sstatic.net/9zTlc.jpg \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 20:41
  • \$\begingroup\$ The 'Powered Subwoofer' is just a Bryston 3B in bridged mode with a 15" speaker across the output. The hum is there even when the receiver is off. \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 20:42
  • \$\begingroup\$ When I put a preamp inline (powered from a 12V battery) the hum goes up drastically according to the gain of the preamp. If I unplug the input line, then the hum leaves. I've tried snap-on chokes, as well as wrapping the cable around a toroid. Neither worked. \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 20:43

5 Answers 5

2
\$\begingroup\$

You would need a very selective (read complicated) band pass filter or you would distort a decent chunk of your bass range. Feedback would not be a real problem since there are no real feedback paths in that circuit.

It would be a lot simpler to take a mains transformer that outputs a low AC voltage, add a potentiometer and a capacitor in series with the secondary winding and mix the output with the line audio.

Even better would be to dismantle the ground loop altogether with an audio isolation transformer, or a TOSLINK S/PDIF optical digital audio cable.

\$\endgroup\$
5
  • \$\begingroup\$ Sorry, but this Bryston was built when TOSLINK was still a twinkle in his father's eye. \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 20:44
  • \$\begingroup\$ Would clipping the AC ground on the Bryston amp fix it? Don't tell me about safety hazards, I know plenty about those. \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 20:47
  • \$\begingroup\$ Electrically isolate all connections from the amp to other grounded devices (with audio isolation transformers) and you have fixed your problem. That is unless the problem is actually in your amplifier (e.g. dried out electrolytic capacitors) \$\endgroup\$
    – jms
    Commented Feb 13, 2016 at 20:55
  • \$\begingroup\$ I'll see if I can find an isolation transformer. If not I may build one. \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 21:34
  • \$\begingroup\$ Whoever voted my anwer as "not useful", could you please enlighten me to what went wrong so I can actually try to improve? \$\endgroup\$
    – jms
    Commented Feb 13, 2016 at 21:55
3
\$\begingroup\$

Trying to address power line hum with a notch filter in the signal path is a kludge at best, and is unlikely to yield the results you expect anyway.

The way to address this is to deal with the real problem of how the power line noise gets into your signal in the first place. Surely there is much out there on this issue already. Go study it.

This isn't the only possible cause, but one of the common ones is bad grounding between equipment. You probably have a ground loop somewhere, so that the signal ground in one or more places is carrying some power return current too. If all the equipment is in the same room, plug it all in to the same outlet. If it's dispersed, then you may need to look into isolation transformers and the like.

\$\endgroup\$
2
  • \$\begingroup\$ Beat me to it. The correct way to fix this is to remove the ground loop. Rane Note 165 is the canonical group loop article. You may have to mod a piece of gear, but doing anything else is a bandaid on a gunshot wound. \$\endgroup\$
    – Matt Young
    Commented Feb 13, 2016 at 21:21
  • \$\begingroup\$ Well, it is all plugged into the same outlet (in fact, I have far too much on that outlet... Note to self - get more breakers). However, the receiver has an 80-foot extension cord, and the Bryston has a 10-foot. Would 70-foot make a big difference? If so, would it help to ground the two directly to each other with a wire? \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 21:37
1
\$\begingroup\$

All you've done is found a way to turn a bandpass filter into a band-stop filter. And it would be a lot more readable if you drew it with the signal flow from left-to-right:

schematic

simulate this circuit – Schematic created using CircuitLab

Also, that isn't exactly how opamps work; you'd need additional feedback paths around each one to implement the functionality that you're suggesting.

But in any case, this will only work to the extent that your interference is a pure 60-Hz sinewave. It's far more likely that your problem includes harmonics (120 Hz, 180 Hz, etc.). It would be far better to find and eliminate the source of the hum.

\$\endgroup\$
2
  • \$\begingroup\$ Well obviously this is a simplified schematic. I need power to the op-amps too 😛 \$\endgroup\$
    – Daniel
    Commented Feb 13, 2016 at 20:46
  • \$\begingroup\$ At best you would turn the hum into a raspy buzz due to the harmonics which are more plentiful than in bygone years. \$\endgroup\$
    – Autistic
    Commented Feb 13, 2016 at 21:00
0
\$\begingroup\$

I just happened to have similar problem with my headphone amp. The hum was -72db at 50hz mains fundamental when usable sine signal was at -15db. Solution was to create separate bias(the voltage level around which the signal wiggles) with another opamp and then with that opamp configuration connect -(minus) power pin(330p) and plus power pin(1u) both to input signal created from 8.3V plus terminal and ground. Now the signal can be -3db and hum is at -96db and rest of the signal is better than 105db. The interface used for measuring is Behringer Ultramatch connected to AES/EBU from RME Raydat.

\$\endgroup\$
-1
\$\begingroup\$

The hum you are hearing is the odd numbered harmonics of the 60 Hz usually called THF. Pure 60 Hz just rattles the windows with little perceptible sound. About the only way to get rid of it/them is to use a balanced audio feed and effectively remote the "ground" of the input amp. Multiple twin T notch filters result in sharp phase shifts as does any filter with sharp cutoffs. Same artifact occurs if trying to use digital tracking filters to remove "hum" in a recording. An incurable problem is that the AC mains wiring has impedance on both sides and thus there are differing amounts of common to "ground" noise as you go back to the distribution transformer. Run the + side of the input op amp to the remote source "ground" reference.

\$\endgroup\$
2
  • 3
    \$\begingroup\$ 60 Hz is only about 2 octaves below "middle C", and should be quite reproducible by pretty much any high fidelity audio system. Also, why only odd harmonics? \$\endgroup\$
    – Dave Tweed
    Commented Feb 13, 2016 at 20:26
  • 1
    \$\begingroup\$ Third harmonic is the dominant harmonic produced by the symmetric clipping (magnetic saturation) in AC distribution transformers. In a delta connection it is a circulating current and in Y connections shows up on the common. Symmetric clipping of a sine wave yields odd order harmonics 3,5,7,9 etc but they drop off at 1/(f^2) (Taylor's expansion). As for hearing 60 Hz, try a good sine wave generator and headphones. For me, it is a sensation and not so much auditory hearing. \$\endgroup\$
    – wa9vez
    Commented Feb 15, 2016 at 0:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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