• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

How dangerous is lifting the ground on a source component?

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,023
Likes
4,423
Location
Berlin, Germany
@tmtomh,

Use a balanced cable from the EQ to the power amp which has XLR's as I may assume, this is an unbalanced RCA to balanced XLR adapter cable which must look like this:
RCA Center goes to inner conductor (+) of the balanced cable,
RCA Shell goes to inner conductor (-) of the balanced cable as well as the shield of the balanced cable.
By this, the offending mains balancing current is diverted away from the signal path wires, causing way less -- if not totally removed -- hum/buzz.

Combine this with moving the EQ Unit (probably a class-II -- 2-prong -- mains device) right on top of the Oppo (which is Class-I) and connect them with the shortest RCA cables you can find, preferably ones with a thick and solid shield construction.

That should suffice to fix that little issue. If not, please report back for further assistance which then might require more details to be checked. Do you have a good multimeter than can display true RMS AC voltages and measure resistances, and some soldering skills?
 

brimble

Active Member
Forum Donor
Joined
Oct 18, 2020
Messages
159
Likes
198
Location
Australia
A possibility is that the mains cable goes directly into the transformer in a way that ensures mains voltages cannot come into contact with the metal enclosure. In this case it is important that the metal enclosure is NOT making contact with the common (audio ground) either otherwise the metal enclosure can still have a voltage on it coming from connected gear.
Ah, yes, that makes sense. Thank you.
 

tonycollinet

Major Contributor
Forum Donor
Joined
Sep 4, 2021
Messages
2,075
Likes
2,939
Location
UK/Cheshire
A possibility is that the mains cable goes directly into the transformer in a way that ensures mains voltages cannot come into contact with the metal enclosure. In this case it is important that the metal enclosure is NOT making contact with the common (audio ground) either otherwise the metal enclosure can still have a voltage on it coming from connected gear.

Double insulation (class 2 insulation system) means that there are effectively two layers of certified basic insulation between dangerous voltages and any touchable metal parts, or enclosure openings.

One of these "layers" can also be a clearance distance of at least 5.5mm, and a clearance of 8mm is considered double insulation (Note these distances must be maintained under worst case tolerances - so for example if one of the layers is the insulation around a wire, then it must not be possible for any part of that wire to get closer than 5.5mm to the chassis - even when "pulled".

It is not necessary to worry about audio ground connections between equipment on a class 2 device, because if another connected device is class 1 - then its earth connection will prevent any dangerous voltages appearing on the interconnect.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
11,432
Likes
25,363
Location
The Neitherlands
It is not necessary to worry about audio ground connections between equipment on a class 2 device, because if another connected device is class 1 - then its earth connection will prevent any dangerous voltages appearing on the interconnect.

But what if the audio ground is connected to the double insulated amp and someone connects another device that is supposed to be connected to safety ground but some user decided to use it in a wall socket or extension cord that is not grounded but doesn't matter for that device and can still electrify the enclosure of the amp ?
Or someone uses a DAC with a 5V input and decided to use a cheap wallwart that fails catastrophically to the 5V ?

All what ifs and slim chances but it could (theoretically) make a presumed safe situation unsafe anyway.

In any case... using ground lift should only be done when one knows the risks involved, what the ground lift does (should be disconnect the audio ground from safety ground but leave safety ground connected).
Some devices, however, disconnect the safety ground from the enclosure which could be dangerous.

Ground lifts can be problem solvers, especially in PA and studio conditions, where this is found.
 

tonycollinet

Major Contributor
Forum Donor
Joined
Sep 4, 2021
Messages
2,075
Likes
2,939
Location
UK/Cheshire
But what if the audio ground is connected to the double insulated amp and someone connects another device that is supposed to be connected to safety ground but some user decided to use it in a wall socket or extension cord that is not grounded but doesn't matter for that device and can still electrify the enclosure of the amp ?
Or someone uses a DAC with a 5V input and decided to use a cheap wallwart that fails catastrophically to the 5V ?

All what ifs and slim chances but it could (theoretically) make a presumed safe situation unsafe anyway.

In any case... using ground lift should only be done when one knows the risks involved, what the ground lift does (should be disconnect the audio ground from safety ground but leave safety ground connected).
Some devices, however, disconnect the safety ground from the enclosure which could be dangerous.

Ground lifts can be problem solvers, especially in PA and studio conditions, where this is found.
All those examples are failures (or bypassing via ground lift) of the safety earth system.

I absolutely agree that if people are ground lifting then that is a serious violation of safety standards, and quite apart from the horror of injuring or killing someone as a result there may even be legal liability to worry about.

The cheap wallwart example is interesting. With a properly designed double insulated wallwart, it should be almost impossible even for catastrophic failure to cause a safety problem (although even the standards define safety as a probability of injury - so it is not 100% impossible). However there have been sufficient examples of cheap imports (whether they are wallwart PSUs, or USB Chargers, or anything else plugging into mains power) being marked with all the safety standards, yet when opened up don't even pay lip service to any sort of safe insulation.

Only ever buy your kit that plugs into the mains from reputable manufacturers with a name to protect - and from places where risk of clones/copies is low.
 

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
If you do lift the ground you could at least check that the circuit is protected by a GFCI (ground-fault circuit interrupter), as that at least gives you some protection in case of a fault.

I also had to lift the ground of my source (computer connected to receiver with HDMI cable) to get rid of some nasty humming, even though all analogue connections are balanced XLR (Nvidia GPU to Marantz AV7706 to Hypex NC502MP based amps was my hum chain). I’m not overly worried about lifting the PC’s ground though since I have modern GFCI breakers, and the PC is still grounded through the HDMI cable.

In your case the Oppo would likely still be grounded through the RCA cables, or through its HDMI cable if you’re using that. You could always measure the resistance from the Oppo to a ground prong to be sure.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
6,128
Likes
11,475
Location
Monument, CO
Article on ground loops: https://www.audiosciencereview.com/forum/index.php?threads/ground-loops-101.7162/

I have pulled the safety ground with a cheater plug as a test, always making sure there was a valid ground path someplace before plugging things in and powering them on. I do not consider that a solution, just a test case to be performed by a knowledgeable person. I am perhaps more sensitized to the safety issue because I have seen a number of cases of harm, and at least one death of a friend's father, when a safety ground failed or was removed and a fault someplace caused the chassis to rise to line potential and shock (electrocute) someone (sometimes me).
 
OP
tmtomh

tmtomh

Major Contributor
Joined
Aug 14, 2018
Messages
1,375
Likes
4,042
@tmtomh,

Use a balanced cable from the EQ to the power amp which has XLR's as I may assume, this is an unbalanced RCA to balanced XLR adapter cable which must look like this:
RCA Center goes to inner conductor (+) of the balanced cable,
RCA Shell goes to inner conductor (-) of the balanced cable as well as the shield of the balanced cable.
By this, the offending mains balancing current is diverted away from the signal path wires, causing way less -- if not totally removed -- hum/buzz.

Combine this with moving the EQ Unit (probably a class-II -- 2-prong -- mains device) right on top of the Oppo (which is Class-I) and connect them with the shortest RCA cables you can find, preferably ones with a thick and solid shield construction.

That should suffice to fix that little issue. If not, please report back for further assistance which then might require more details to be checked. Do you have a good multimeter than can display true RMS AC voltages and measure resistances, and some soldering skills?

Thanks very much for this! I wondered if an RCA-XLR cable somewhere in the chain could also remedy this, but I ordered the Hum-X because I wasn't sure where in the chain the cable should go (Oppo XLR-EQ RCA or, as you recommend, EQ RCA to amp XLR) - and I also was unsure about ensuring that the cable would be internally wired correctly as you specify. I just ordered a pair. It and the Hum-X should both arrive at the same time, so I'll try each, and both, and see what happens.

You are absolutely correct that the EQ is a 2-prong device, and I am also wondering if that is part of the issue - I am wondering if taking the EQ out of the chain for testing purposes and connecting the Oppo directly to the amp, even with the unbalanced cables, would achieve the same solution to the hum. And similarly, I am wondering if any source component connected in my current setup would exhibit the same hum as the Oppo.

As for the RCA cable lengths, the Oppo is on a shelf just above the EQ, which is just above the amp. Both connections are via very short RCA cable runs, 18 inches/0.5M. (I also tried physically moving the units' power cords closer to and then away from where the RCA cables are, but I detected no variance in the hum, which reinforced my feeling that it's a ground loop and not EMI.)

On a related note, the main reason I have my setup this way - aside from it just being a partial legacy of how my system started out 30+ years ago - is to enable the playback of physical SACDs from the Oppo without an extra A/D/A conversion step and without a DSD-PCM conversion step. Given that I own fewer than 40 SACDs, only about 10 of which are either single-layer or have a different mastering on the SACD layer, I suppose it's time to consider listening to those few discs' content exclusively via rips of them played through my computer music server. My current setup uses iTunes and so I have to use PCM conversions of the DSD rips, but even then I can't imagine there are any detectable sonic differences between those PCM files and the original SACD disc. Right now I think it's just a psychological/emotional attachment to a few of those discs. And it's certainly not worth maintaining this signal path in the long run just for that.

My dream would be to ditch the EQ and amp and just run the Oppo, via SPDIF, directly into a pair of Genelec 8361s.
 
Last edited:
OP
tmtomh

tmtomh

Major Contributor
Joined
Aug 14, 2018
Messages
1,375
Likes
4,042
Article on ground loops: https://www.audiosciencereview.com/forum/index.php?threads/ground-loops-101.7162/

I have pulled the safety ground with a cheater plug as a test, always making sure there was a valid ground path someplace before plugging things in and powering them on. I do not consider that a solution, just a test case to be performed by a knowledgeable person. I am perhaps more sensitized to the safety issue because I have seen a number of cases of harm, and at least one death of a friend's father, when a safety ground failed or was removed and a fault someplace caused the chassis to rise to line potential and shock (electrocute) someone (sometimes me).

Thanks for this, Don! I am not sure I fully understand - for some reason my brain can grasp digital sampling theory more easily than basic electrical principles :) - but I find your initial examples quite instructive. In the first example there's no ground loop because only one of three components (the source) is grounded; the preamp and amp are floating. In the second example all three are grounded. If I am understanding the basic idea correctly - and again, I might not be! - my current setup has 2 grounded devices (Oppo and power amp), which can create a loop. And when I temporarily lifted the Oppo's ground for testing purposes, the chain then had only one grounded device - the power amp rather than the source as in your example, but still....

So at the risk of asking a stupidly basic question, is that what it means to break the ground loop - I lifted the ground of one device, leaving only one grounded device in the chain, hence no loop?
 

tonycollinet

Major Contributor
Forum Donor
Joined
Sep 4, 2021
Messages
2,075
Likes
2,939
Location
UK/Cheshire
For gods sake people. Don't think you can rely on analogue grounds on signal cables (whether they are HDMI, RCA, XLR or any other) to protect you in the case of electrical fault.

The whole point of safety earth connections is they are specified to take high fault current until the fuse or circuit breaker clears. Your signal cables may or may not be able to do that - probably in most cases not. They do not make a safe earth - even if they measure low resistance.

It is also not safe to rely on ELCB devices. The current trip level of which is normally higher than can be dangerous.
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
6,128
Likes
11,475
Location
Monument, CO
Thanks for this, Don! I am not sure I fully understand - for some reason my brain can grasp digital sampling theory more easily than basic electrical principles :) - but I find your initial examples quite instructive. In the first example there's no ground loop because only one of three components (the source) is grounded; the preamp and amp are floating. In the second example all three are grounded. If I am understanding the basic idea correctly - and again, I might not be! - my current setup has 2 grounded devices (Oppo and power amp), which can create a loop. And when I temporarily lifted the Oppo's ground for testing purposes, the chain then had only one grounded device - the power amp rather than the source as in your example, but still....

So at the risk of asking a stupidly basic question, is that what it means to break the ground loop - I lifted the ground of one device, leaving only one grounded device in the chain, hence no loop?

Yup!
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
2,712
Likes
6,795
Location
Suffolk UK
For gods sake people. Don't think you can rely on analogue grounds on signal cables to protect you in the case of electrical fault.

The whole point of safety earth connections is they are specified to take high fault current until the fuse or circuit breaker clears. Your signal cables may or may not be able to do that - probably in most cases not. They do not make a safe earth - even if they measure low resistance.

It is also not safe to rely on ELCB devices. The current trip level of which is normally higher than can be dangerous.
Absolutely this a thousand times !!!! Using the audio interconnect as a safety earth path is very very dangerous. One valve pre-power combination I used once had the HT and heaters to the pre-amp going via a multiway between the power amp and preamp, with the audio and ground connection in the same multiway. There was no independent earth connection to the preamp, just through the multiway. The earth return in the multiway went open circuit, so the case of the preamp was sitting at 400v which bought a few tears to my eyes when I touched the metal volume control.

Every item that's not double insulated should be independently earthed so it would take several simultaneous failures to create a dangerous condition. With vintage equipment, even if originally double insulated, I prefer to run a separate earth connection as I'm not happy to rely on 50 year old insulation and technical standards to maintain safety.

S.
 

stubaggs

Member
Forum Donor
Joined
Mar 6, 2020
Messages
13
Likes
14
My current signal chain is not ideal but for now it's what I have: source component (Oppo UDP-205) into graphic EQ, into power amp (Purifi-based unit). I have recently been hearing some low-level hum through the speakers when I go near them and no music is playing. (And no, I have no idea why the hum has appeared or else increased in level recently.)

I am using unbalanced interconnects because the EQ unit does not have balanced inputs or outputs. I have reasons for having the EQ in the chain, and while I am not permanently attached to this setup, I'd rather not get into that right now - I understand that moving to balanced interconnects could solve this problem in the long term.

But here is my question: After unplugging things, switching things around in different receptacles, swapping AC cords, power strips etc, and fiddling with lights and dimmers around the house, I have narrowed down the source of the hum to the Oppo in the following way:

  • Amp turned on, other components off, RCA cables disconnected from the amp = no more hum. So it's not coming from the power amp. It must be one of the two upstream components.
  • Amp on, other components off, RCA cables connected from amp to EQ, but disconnected from Oppo to EQ = no more hum. So it's not the EQ (or at least it's not the EQ interacting with the amp).
  • Amp on, other components off, RCA cables connected to all components, but Oppo power cord unplugged from wall = no more hum. As soon as I plug the Oppo back in, the hum returns.
  • And finally, amp on, other components off, RCA cables all connected, Oppo plugged into AC BUT the Oppo's ground lifted (via a heavy-duty extension cord with a three-prong receptacle for the Oppo's cord but only a two-prong plug going into the wall socket) = no more hum. The setup is literally just as quiet with the ground lifted as it is if I unplug the Oppo from AC entirely.

So while I freely admit I do not fully understand how hum can be caused through the interactions among components, this process of elimination seems to show that when I lift the ground of the Oppo, the problem goes away.

For this reason, I have a Hum-X on order. Even though it does not lift ground (and therefore is safer), it apparently achieves the same result if the problem is a ground loop (and not EMI, for example). Only time will tell of course - I'll have to try it out and see.

In the meantime, I am wondering if it is actually unsafe to operate the Oppo with a lifted ground. I would not even consider, even as a temporary measure, lifting the ground on a power amplifier. But I am wondering about lifting the ground on a 2V line-level source component (which I am using for audio only - there is no TV hooked up to it or even in the room). Is it just as dangerous because we're still talking about mains power going through the unit's internal transformers? I would definitely want to prioritize safety, and it's a simple matter to wait a few days for the Hum-X to arrive and then try it. I'm just curious about the real-world implications of using a source component with a lifted ground.

Thanks for any advice or guidance folks can provide - and again, I understand that moving to a setup with balanced interconnects would likely be a better long-term solution here.
I successfully eliminated hum from a projector connected to my home theatre using a 'Hum Eliminator' (used to be called a HumX), details from the product page:
  • Exterminate Ground Hum Right at The Source!
  • Designed to solve GROUND RELATED Hum and Buzz
  • Filters the Ground Line of Excess Low Voltage and Current that Cause Audio Noise
  • Maintains Proper Ground
  • NOT a Ground Lift
  • Simply Connect Hum Exterminator™ to Power Plug of Offending Gear
  • Maximum recommended Current per Hum Exterminator™ Is 6 Amp
 
OP
tmtomh

tmtomh

Major Contributor
Joined
Aug 14, 2018
Messages
1,375
Likes
4,042
I successfully eliminated hum from a projector connected to my home theatre using a 'Hum Eliminator' (used to be called a HumX), details from the product page:
  • Exterminate Ground Hum Right at The Source!
  • Designed to solve GROUND RELATED Hum and Buzz
  • Filters the Ground Line of Excess Low Voltage and Current that Cause Audio Noise
  • Maintains Proper Ground
  • NOT a Ground Lift
  • Simply Connect Hum Exterminator™ to Power Plug of Offending Gear
  • Maximum recommended Current per Hum Exterminator™ Is 6 Amp

Thanks! A Hum-X is one of two potential solutions I have on order.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
6,128
Likes
11,475
Location
Monument, CO
Thanks! A Hum-X is one of two potential solutions I have on order.

I have used and recommended the Hum-X a lot over the years. It is pricey but fills a niche. One thing to be aware is that, unless they've changed, it has a current rating of about 6 A, so better used on lower-level stuff than power amps.
 

raindance

Addicted to Fun and Learning
Joined
Sep 25, 2019
Messages
725
Likes
625
I'd wager that the Oppo is not the cause of the issue and that it's either the equalizer or the amp. There are some products that use poor grounding practice where the RCA connectors are directly connected to the chassis. When I build amps, I use a grounding scheme where the signal grounds are connected in a star to chassis through a fairly low value, high power resistor, say 10 ohms, 25 watts, bypassed by a 0.1uF high voltage cap to provide a low impedance signal path to earth. Incoming mains is hard grounded to the chassis at all times and I use a fused IEC connector. I don't get ground loop issues with this scheme. The resistor / cap combo simply needs to handle enough power for the fuse to blow when a fault occurs.
 

Cbdb2

Addicted to Fun and Learning
Joined
Sep 8, 2019
Messages
657
Likes
527
Location
Vancouver
Just thinkin out loud. Seems like lifting the ground would be safe if you use a GFCI. And cheap.

"The GFCI monitors the electric current in the hot line and the neutral line (see Figure 1). When these two currents are equal operation is normal and there is no ground fault. However when these two values become unequal by as little as 4-5 mA another path to the ground (ground fault) may be present, in this case the GFCI will open (with a response time of 1/30th of a second) thereby preventing electric shock to a person. A GFCI has a reset button to restore it to original conditions after it has opened the circuit, as well as a test button to ensure it is functioning properly.[3]"

Comments?

An opportunity to get rich selling audiophool GFCIs, if people will spend $100 on a regular outlet (http://www.enjoythemusic.com/magazine/equipment/0114/audiophile_ac_outlets.htm) think how much you can soak them for a an outlet that "gets rid of ground loops" etc.
 

Cbdb2

Addicted to Fun and Learning
Joined
Sep 8, 2019
Messages
657
Likes
527
Location
Vancouver
I'd wager that the Oppo is not the cause of the issue and that it's either the equalizer or the amp. There are some products that use poor grounding practice where the RCA connectors are directly connected to the chassis. When I build amps, I use a grounding scheme where the signal grounds are connected in a star to chassis through a fairly low value, high power resistor, say 10 ohms, 25 watts, bypassed by a 0.1uF high voltage cap to provide a low impedance signal path to earth. Incoming mains is hard grounded to the chassis at all times and I use a fused IEC connector. I don't get ground loop issues with this scheme. The resistor / cap combo simply needs to handle enough power for the fuse to blow when a fault occurs.
98% of unbalanced commercial gear has the RCA neutral connected to ground with no problems.
 

Matias

Major Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
3,192
Likes
5,771
Location
São Paulo, Brazil
I always say that ground loops are cause by excessive grounding paths, so disconnecting one device still has it grounded by the cables to the other. If the other device was not grounded, than you wouldn't have a loop and would not need to lift the first, right?
 
OP
tmtomh

tmtomh

Major Contributor
Joined
Aug 14, 2018
Messages
1,375
Likes
4,042
I always say that ground loops are cause by excessive grounding paths, so disconnecting one device still has it grounded by the cables to the other. If the other device was not grounded, than you wouldn't have a loop and would not need to lift the first, right?

That totally makes sense - but others have said that such a situation leaves the lifted device connected to ground only by the interconnects, and they feel that's not a sufficient or safe condition. Are you disagreeing with that?

Also, what do you think of @Cbdb2 's notion that one could safely lift a device's ground if it were plugged into a GFCI receptacle?


I have used and recommended the Hum-X a lot over the years. It is pricey but fills a niche. One thing to be aware is that, unless they've changed, it has a current rating of about 6 A, so better used on lower-level stuff than power amps.

Glad to hear another endorsement of the Hum-X! For $80 it might be expensive/overpriced for what it is, but in the scheme of our hobby, it's a pittance, and if it solves this issue until I get the money and time together to move to my next (and hopefully last) revamp of my system, it will be more than worth it.
 
Top Bottom