1
\$\begingroup\$

I have an A/V system with two active sub-woofers.

One is close to the amplifier connected with a 1m RCA cable. Its mains connection is into the same power strip as the amplifier. There will be an earth loop but it is short and it does not add a significant amount of noise. Update, I now know that the amp is not directly earthed but it may be through one of the other devices.

The second sub is further away and is connected with a 10m RCA cable. It is not actually that far but the cable needs to go around the wall and not across the floor. 7m would be sufficient but that is not a commonly available length. I tried a 5m cable with a 2m extension but it was not noticeably better. The mains for this sub is close to the sub and hence not close to amp. This earth loop is much larger and picks up a significant hum which I would like to eliminate.

Further evidence that the problem is an earth loop:

  1. The hum is stopped if the signal cable is disconnected. I tested by disconnecting the at the sub end. My initial concern was that there was a fault within the sub. I have tried disconnecting the signal cable at the amp end. Initially, I got a worse hum but when I shorted out the loose end, this went away.

  2. If I feed a signal from a battery powered iPod, there is no hum. The iPod was connected with a short cable. I'll test with a longer cable along a similar route to the usual signal cable and update later.

  3. Temporarily breaking the supply earth while connected to the amp also stops the hum.

All plugs are three pin and connected to grounded sockets. It is possible that the earth is not used but test 3 suggests that the sub is certainly grounded. I could repeat test 3 but break the earth at the amp end as an additional test. Most of the plugs are sealed and not easy to inspect.

So, a simple and effective but maybe not safe and legal solution is to break the supply earth. I would like to find a better solution.

I found this item on Amazon: AV Link Ground Loop Isolator. Cheap and easy to fit. I feared that it might attenuate the signal and cause harmonic distortion but a quick test using test tones from my iPod and my ear to judge suggested that it was not too bad. However, when I connected it with the amp, it actually made the problem far worse rather than better. I guess that it contains a transformer which was picking up more hum than the earth loop.

I found other items, e.g. PAC SNI-1 Noise Isolator which did not seem to be transformer based, but the reviews were terrible so I did not even try them.

So, I am still looking for a solution and preferably not a very expensive one. The devices above are passive but I would consider a powered device.

It is common advice to keep the signal cables away from the mains and that is already the case here but I was considering the exact opposite: power the sub from the same power strip as the amp and the other sub. Hence, it would need an equally long power cable as the signal cable which I would run together. The idea is that both the signal and power cable would pick up the same hum but out of phase and hence they would cancel over. Is this worth trying or have I missed something?

I am in the UK so the supply is 230VAC 50Hz. Since this is the UK, I expect that everything in the house is on the same phase. In this case, both sockets are on the same breaker in the consumer unit.

The AVR is a Denon X3200W - manual.

The subwoofers are BK Gemini IIs. I cannot find an online manual but even the supplied paper manuals are very simple.

Other devices connected to the amp are: the TV, a Tivo (digital video recorder), two disc players, a digital iPod dock, and the other sub. The sub uses an RCA cable, the iPod dock uses an optical cable, and all others are HDMI.

All devices except the Tivo and the distant sub are powered by the same power strip. The Tivo goes to another power strip but both strips go to the same double socket. This is partly just a matter of sockets on the strips but also one strip contains devices that can be off when no one is present and the other is for devices that should remain on: Tivo, internet router, phone base station, etc.

I have now checked the amp end of its power cable and it has a two pin connector. So, it is not earthed and my loop must be through one of the other devices, possibly even the other sub.

Thanks to the comments, I have more ideas to try. Annoyingly, repeats of tests that I have made before don't seem to entirely consistent. Possible explanations of the inconsistent results are that my hearing varies and other noises vary. E.g. sometimes I hear the hum from a fridge two rooms away more clearly than the hum from the subwoofer. More research is required.

\$\endgroup\$
20
  • \$\begingroup\$ Is the hum stopped if the cable is disconnected at the sending end? You didn't say which end you disconnected. The iPod signal - where along the cable was this injected? Start or end? \$\endgroup\$ – Andy aka Jun 6 '20 at 10:09
  • \$\begingroup\$ Does the amplifier have grounded or ungrounded plug? Does the subwoofer have grounded or ungrounded plug? What other devices connect to the system and do they have grounded or ungrounded plugs? Are all devices with grounded plugs connected to grounded outlets? Are the devices connected to same mains phase or different mains phase? Which subwoower and amplifier model? Link to manuals we can read? \$\endgroup\$ – Justme Jun 6 '20 at 10:42
  • \$\begingroup\$ Transformer isolator ought to be in a (mu-metal) screened can. But try the one you've got at both ends of the cable, it may pick up less hum in the other position. \$\endgroup\$ – user_1818839 Jun 6 '20 at 11:20
  • \$\begingroup\$ yeah, look for one with a steel-looking case if it's not metal under that black plastic \$\endgroup\$ – Jasen Jun 6 '20 at 11:25
  • 1
    \$\begingroup\$ In general consumer AV receivers don't use ground. Good to add that mains is single phase. Yes, disconnecting one wire or device at a time will help determine what causes the hum. However, do as the manuals say and never connect or disconnect two devices when their mains cords are powered. For example, if you have a BD player, and only connections to it are mains input and HDMI to amplifier, disconnect BD player mains plug first before HDMI. You don't want to connect and disconnect a grounded device with ungrounded devices, or two ungrounded devices. \$\endgroup\$ – Justme Jun 6 '20 at 15:33
1
\$\begingroup\$

The best sollution to remove the hum is to use all optical cables for the audio. You can buy cheap analog (RCA) to optical SPDIF and replace all audio with this (but this is only an alternative path to the grounding problem)

\$\endgroup\$
9
  • \$\begingroup\$ An interesting idea and I have half of what I need already (left over from another project): the optical to RCA part. However, would the cheap items have a suitable frequency range for a subwoofer? \$\endgroup\$ – badjohn Jun 7 '20 at 11:06
  • \$\begingroup\$ yes, no problem. If you look into this converters there are some simple chip like Texas Instruments PCM2902 (used in sound cards like Behringer UCA 202) which does everything. Depending on how this chip is powered you can ad new noise to the system or not, so if you found a new noise try to add some capacitor to the power supply or change the power supply. \$\endgroup\$ – danirebollo Jun 7 '20 at 11:17
  • \$\begingroup\$ This is cheap and promising enough that I'll try it out. I'll need a long optical cable but I can test the idea with a short one that I already have. \$\endgroup\$ – badjohn Jun 7 '20 at 11:35
  • 1
    \$\begingroup\$ You are right, the coaxial interconnect did not solve the problem: neither the ADC nor the DAC broke the earth. However, with an optical connection, it was perfect. It was even better than RCA with a broken supply earth to the sub. I hit one problem: there was a new noise but I traced that: I had powered the ADC from the same USB supply as a nearby Raspberry Pi; switching to it own supply solved this. Now, I just need a longer optical cable to achieve a wife approved solution. A perfect solution: cheap, low priced, and effective. I could not detect any attenuation at low frequencies. \$\endgroup\$ – badjohn Jun 13 '20 at 9:56
  • 1
    \$\begingroup\$ yes, the power supply needs to be clean enough. You can try with different power supplies. If some noise remains: try with some LC filter, or go for a linear power supply \$\endgroup\$ – danirebollo Jun 13 '20 at 16:19
2
\$\begingroup\$

you may reduce the hum, using power cables that have rapid-rotation twists of the hot/return pair. This fast rotation of the magnetic fields will reduce the induced voltage in all nearby wiring loop.

The density of twists will increase the self-cancellation of magnetic fields, even for small short audio cables or small regions of your PCB boards.

To evaluate this, get apiece of sheet steel, tie it to GROUND with a flexible wire (to ensure it will not become an electric_field radiator) and experiment.

  • Hold the sheet next to aggressor wiring --- hidden power wiring inside the walls, or near power cords running to your AV setup or near power transformers inside your AV boxes.

  • Hold the sheet near your victim wiring (coax cables?) or audio Preamplifier boxes.

Ohhh ensure all your audio cables have grounds running end-to-end.

\$\endgroup\$
1
  • \$\begingroup\$ I don't recall seeing that, at least not for a very long time, but I'll look out for some. However, I won't be able to change the wiring in the house which will be part of the problem. The signal cable might still pick up interference from the fixed wiring. \$\endgroup\$ – badjohn Jun 6 '20 at 17:10
0
\$\begingroup\$

Hum in AV systems is often related to antenna or (old coaxial) network, cable TV or other signal cables connected to the system. Those cables bring a 2nd or 3rd ground connection to the amplifiers' ground.

\$\endgroup\$
3
  • \$\begingroup\$ I am familiar with those issues but I don't think that they apply here. I could test by disconnecting all devices except the amp and the sub but that breaking the earth for the sub solves the problem (test 3) seems to confirm that the issue is caused by the loop between the amp and the sub. \$\endgroup\$ – badjohn Jun 6 '20 at 12:19
  • \$\begingroup\$ Update. The loop will not be between the amp and the sub as I now know that the amp is not earthed. The loop may be between the sub and another device connected to the amp, maybe even the other amp. \$\endgroup\$ – badjohn Jun 6 '20 at 13:17
  • \$\begingroup\$ A mistake in my last comment. It should end: "maybe even the other sub". There is only one amp and I now know that it is not earthed, \$\endgroup\$ – badjohn Jun 6 '20 at 17:13

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.