AC Polarity and the reduction of noise and hum

marty said:

There was an outstanding article by Greg Weaver in February's TAS on how to reduce noise and hum by minimizing AC voltage from chassis gear to ground by altering AC polarity. Unfortunately the article is not available as a pdf now. However, I checked previous articles/blog posts by Greg Weaver and he basically reported the results of his findings and recommendation in a virtually verbatim article here: See post #4:

https://www.thegearpage.net/board/i...x-polarity-tester-to-reduce-hum-noise.882742/


As he points out, this issue was first identified by Enid Lumley in 1981, so its an old issue. However many may not be aware of it, or the possibility to obtain best performance from his/Enid's advice. Weaver's article is the best and clearest I have seen on the subject. His recommendations are easily implemented and might just help you reduce or banish the noise and hum in your system if it's an audible problem.

Click to expand...

As you say, it is an old issue - I learned about it in TAS long ago. Since then I have a cheater plug and even before being connected every piece of equipment entering my room is checked for polarity and receives a red dot close to the power connector. Unfortunately many pieces of equipment are not wired in accordance to the IEC code.

The technical explanation is however much simpler than what referred in your link - the correct position creates less leakage current through the ground wire, reducing ground noise. Just Ohm's law - the VOM behaves like an high value resistor!

I did this recently with my DIY Linear Regulated PSU and had quite a big difference in SQ, so that now I have actually marked one leg of the two-prong jack so that can always plug this one in the longer slot (Canada AC).

You can check it with the Multi-Meter method or simply try A/B'ing the two configurations by listening.

If device-side there's also a plug of the reversible type, then you may want to mark both one cord side and the corresponding plug side.

Yes, this reduces Leakage Current, and it is very audible, the extent of which may vary with your gear.

It is a necessary step in a proper audiophile initial set-up IMO.

Your transformer could already be the correct orientation. That's possible too.

Folsom said:

Your transformer could already be the correct orientation. That's possible too.

Click to expand...

A rather pointless comment, as this must be determined for better sound anyway.

I just mean to point out that reversing the AC hot/neutral relation to transformer primary/primaries isn't a guarantee (it's 50/50), since you're not [actually] reversing polarity.

marty said:

Interesting. If this is the case, couldn't you effect a solution by grounding every chassis. For example if I measure 0.20mV AC between chassis and then place a ground wire between chassis and the ground screw of the duplex wall outlet, the measurement goes to 0.00. Same thing?

Click to expand...

Probably wouldn't work very well. AC takes the path of least impedance, so the only way to be sure this works is to try and place the safety ground wire close to where the parasitic qualities from the transformer are leaking. And you'll need to get compression on it for good contact. Even if you do, it'll split the voltage so it may be hard to read depending on your meter.

The easiest thing to do is just swap the transformer wires and listen, but that's not easy for everyone. This is the ultimate solution because you still want the Hot to be on the power switch and fuse.

Folsom said:

I just mean to point out that reversing the AC hot/neutral relation to transformer primary/primaries isn't a guarantee (it's 50/50), since you're not [actually] reversing polarity.

Click to expand...

This is nonsensical, just as your previous comment.

No it's not. We're dealing with AC here that is potentially deadly, and one should be aware that you can bypass the safety by just swapping hot/neutral and not the connection to the transformer where it has no orientation except the winding that has the highest parasitic qualities to chassis; which is not a requirement or this "mod" wouldn't be worth the time because it would already be done.

AC doesn't have "polarity" as it the polarity changes in cycles. Think of it as a single ended interconnect, the 120v 60hz cycle is only on the hot line, where as neutral is dead unless it's making a complete circuit with hot. For that reason you cannot change the polarity as the potential is already changing (+/- polarity) but only on the one leg (hot) at 60hz cycles.

YashN said:

This is nonsensical, just as your previous comment.

Click to expand...

Folsom said:

No it's not. You need to review this subject before you talk someone into damaging their equipment. I'm not sure you understand many of the factors involved. We're dealing with AC here that is potentially deadly, and you should be aware that you can bypass the safety by just swapping hot/neutral and not the connection to the transformer where it has no orientation except the winding that has the highest parasitic qualities to chassis; which is not a requirement or this "mod" wouldn't be worth the time because it would already be done.

AC doesn't have "polarity" as it the polarity changes in cycles. Think of it as a single ended interconnect, the 120v 60hz cycle is only on the hot line, where as neutral is dead unless it's making a complete circuit with hot. For that reason you cannot change the polarity as the potential is already changing (+/- polarity) but only on the one leg (hot) at 60hz cycles.

Click to expand...

Take seriously what the man wrote, rather than insulting him in the name of great sound. AC shock can be quite serious. Don't mess with the AC with cheater plugs and whatnots... We are talking about safety here: Life and deaths issues. Not trivial sonic pursuits.

marty said:

Interesting. If this is the case, couldn't you effect a solution by grounding every chassis. For example if I measure 0.20mV AC between chassis and then place a ground wire between chassis and the ground screw of the duplex wall outlet, the measurement goes to 0.00. Same thing?

Click to expand...

Surely having a thick ultra low recessivity strap between all the chassis's attenuates grounding effects. In laboratories sometimes we have a thick copper bar along the wall and ground every chassis to it. I remember seeing pictures of a great system and room with several turntables using this approach in the old Audiogon "systems" section. But you then have to ground EVERY equipment to it.

The main question is that manufacturers follow different priorities in equipment grounding techniques - they have to consider signal ground, power supply ground and electrical mains ground, required for safety and having to obey different local codes. And then we mix equipment!

Electrical codes do not allow physical separation the grounds in equipment - the old discontinued practice of having separate grounds strapped together was of great help, but could have disastrous consequences if misused. I witnessed a big explosion due to improper use of this feature in an ARC SP10 - OTL system!

Folsom said:

Probably wouldn't work very well. AC takes the path of least resistance, so the only way to be sure this works is to try and place the safety ground wire close to where the parasitic qualities from the transformer are leaking. And you'll need to get compression on it for good contact. Even if you do, it'll split the voltage so it may be hard to read depending on your meter.

The easiest thing to do is just swap the transformer wires and listen, but that's not easy for everyone. This is the ultimate solution because you still want the Hot to be on the power switch and fuse.

Click to expand...

Not exactly true, at less if the manufacturer exports to Europe. Schuko mains plugs are not oriented and manufacturers must consider that any of wires can be hot or neutral. Just rewiring the power cable (or in our case, rotating the power plug) is enough.

BTW, mains power systems and codes are different around the world - no universal rules on this subject, except minimizing leakage current.

In apparatus that have more than one transformer a perfectionist could think about following your suggestion - re-wiring each of them separately for low leakage. But than probably he would find that changing the magnetic coupling of them was causing hum, manufacturers usually optimize for this aspect!

microstrip said:

Not exactly true, at less if the manufacturer exports to Europe. Schuko mains plugs are not oriented and manufacturers must consider that any of wires can be hot or neutral. Just rewiring the power cable (or in our case, rotating the power plug) is enough.

BTW, mains power systems and codes are different around the world - no universal rules on this subject, except minimizing leakage current.

In apparatus that have more than one transformer a perfectionist could think about following your suggestion - re-wiring each of them separately for low leakage. But than probably he would find that changing the magnetic coupling of them was causing hum, manufacturers usually optimize for this aspect!

Click to expand...

I know that sound can change when we change Schuko plug orientations, but is there any risk to it, of a component blowing up?

marty said:

There was an outstanding article by Greg Weaver in February's TAS on how to reduce noise and hum by minimizing AC voltage from chassis gear to ground by altering AC polarity. Unfortunately the article is not available as a pdf now. However, I checked previous articles/blog posts by Greg Weaver and he basically reported the results of his findings and recommendation in a virtually verbatim article here: See post #4:

https://www.thegearpage.net/board/i...x-polarity-tester-to-reduce-hum-noise.882742/

As he points out, this issue was first identified by Enid Lumley in 1981, so its an old issue. However many may not be aware of it, or the possibility to obtain best performance from his/Enid's advice. Weaver's article is the best and clearest I have seen on the subject. His recommendations are easily implemented and might just help you reduce or banish the noise and hum in your system if it's an audible problem.

Click to expand...

I actually don't think it is very clear . I had to read it a few times to figure out what he is saying and what is trying to be accomplished.

Turns out what he is saying may make things worse, not better. Let me explain.

What he is trying to do as Folsom correctly explained, is to take advantage of one of the leads in the input transformer leaking more into the chassis than the other, say through proximity or construction. That by itself can be a good thing. As Folsom again correctly says, if there is such kind of "polarity" then it is best resolved by changing the wires internally. I doubt that high-end equipment makers would leave this stone unturned but maybe some do.

The reason this can be beneficial is that if the source equipment has higher ground potential than your target (say between pre and power amps) then that current flows from source to destination in unbalanced (RCA) connection causing 50/60 Hz originated hum. So it seems that lowering the leakage in every gear would be a good thing but it is not. Read again what I said: what flows is the differential between equipment which will never be zero. So what you want to do is to get that differential to be the lowest, not the individual leakage voltage. If you measured the first gear and it is at say, 50 microvolts and the second at 60, and then you swapped leads and got one to 30 and the other to 10, you are worse off. In the first case the differential was 10 microvolts, the second, 20. So you doubled the voltage which then doubles the leakage current and doubles the hum.

The biggest issue here is that almost all of our equipment comes with three prong cables. Swapping the hot and neutral with a cheater plug requires eliminating the ground which takes away the safety redundancy in your equipment. Should one in a million chance thing happen with the hot lead coming loose and touching the chassis, and said gear is connected to other gear that is grounded, lots of things could short out and cause fire, etc. And if all of them are floating, now you could electrocute yourself.

A better solution than any of this -- assuming there is a problem -- is to use balanced connections. Then it doesn't matter what the chassis voltage is.

+1 Amir, Folsom.

bonzo75 said:

I know that sound can change when we change Schuko plug orientations, but is there any risk to it, of a component blowing up?

Click to expand...

No. When I introduce I a Shucko plug I am playing a 50% probability game of having the "correct" orientation that results in the proper phase and neutral in the IEC plug. This can be problem for some particular equipment - my house central heating system controller only works in a defined position!!!

We now also have this problem in laboratories - in the old days, before Schuko, the plugs were asymmetrical. Water heating systems also used these asymmetrical plugs for other reasons - for increased safety they wanted to be sure that the thermostatic switch has on the phase wire.

Fuse/breaker should be on the hot side, not the neutral side. In the USA it's not possible to do this anyways, and plugs are polarized for a reason. I think this is a bad idea, but maybe EU gear is different as Shuko can be inserted either way afaik?

If your system is setup properly this shouldn't make an audible difference. If resistance between component ground planes is far too high to begin with than sure, but that just points to poor setup to begin with.

amirm said:

I actually don't think it is very clear . I had to read it a few times to figure out what he is saying and what is trying to be accomplished.(...)

Click to expand...

Yes, it is so unclear that people will probably misinterpret as it seems to me you and others did.
The cheater is only used to determine the lower leakage position.

IMHO people should use originals to comment in depth on techniques, not just use a confusing reference.

BTW, we will eventually disagree on this issue, as you will just consider the 50/60 Hz hum, not the non scientifically proved wide band grounding issues.

DaveC said:

(...) but maybe EU gear is different as Shuko can be inserted either way afaik? (...)

Click to expand...

Not in all EU countries (Shuko's also have an asymmetrical ground connection for some specific countries), but mostly yes!