I saved your note to digest. Thank you.
Isolation Transformers Used For Audio Systems and Grounding
- Thread starter wmhart
- Start date
- Status
- Location
- New Jersey
- Occupation
- Journeyman Electrician
A separate service from the utility wouldn't be the same as a feeder from one building to the other. You simply cannot isolate one building grounding system from the other when the feeder originates in one of those buildings.
I'm surprised a 60/120V isolation transformer has not been mention yet. Perhaps it has without knowing it...? It is a supply configuration permitted for audio signal processing equipment. I believe it is also used in hospital operating rooms and in other situations utilizing sensitive electronic equipment. In short, with ground connected to the neutral point of the transformer secondary, the power supply system is "balanced". Being an audiophile, you should know the theory behind balanced vs. unbalanced signal transmission. Essentially the same principle applies to this power supply configuration.
Two NEC articles, in particular, will apply here:
The 120V branch circuits will all be connected line-to-line, so each circuit requires a two-pole common trip breaker. All 15A and 20A receptacles are required to be GFCI protected and marked [see 647.7(A)(2)] Not required, but suggest running as isolated ground circuits and receptacles.
With that said, I have never set up such a system. My statements here are based my interpretation of Code, experience, and knowledge. As to Code requirements, the only section I'm having a problem with is 647.7(A)(4) regarding receptacles, which states "All 125-volt receptacles used for 60/120-volt technical power shall have a unique configuration and be identified for use with this class of system." Exactly what is meant by unique configuration? Do isolated ground GFCI receptacles qualify? I ask because...
Two NEC articles, in particular, will apply here:
- ARTICLE 640— AUDIO SIGNAL PROCESSING, AMPLIFICATION, AND REPRODUCTION EQUIPMENT
- ARTICLE 647— SENSITIVE ELECTRONIC EQUIPMENT
The 120V branch circuits will all be connected line-to-line, so each circuit requires a two-pole common trip breaker. All 15A and 20A receptacles are required to be GFCI protected and marked [see 647.7(A)(2)] Not required, but suggest running as isolated ground circuits and receptacles.
With that said, I have never set up such a system. My statements here are based my interpretation of Code, experience, and knowledge. As to Code requirements, the only section I'm having a problem with is 647.7(A)(4) regarding receptacles, which states "All 125-volt receptacles used for 60/120-volt technical power shall have a unique configuration and be identified for use with this class of system." Exactly what is meant by unique configuration? Do isolated ground GFCI receptacles qualify? I ask because...
647.7(A) said:
The Equi=Tech is one such system, but believe it must be grounded. (I thought the hospital transformers floated ground at the secondary). I own a 10kVa version of the Equi=Tech wall cabinet that supplies balanced power and integrates a breaker panel, all in a pretty large 400 lb plus box that must be installed indoors. I was holding that in reserve for my separate outbuilding. I also wasn't sure it would be permitted within a residence, even if the room I am now using is otherwise dedicated--which it is. I think we are talking about the same thing or similar.
Would it not be better to hire a local electrical or audio engineer since you are willing to pay?
Austin TX should have a slew of them being music industry intensive
they could take a few measurements and see if an issue exists
it may be as simple as reconditioning/augmenting your existing grounding system
Hi end hifi equipment have very good dc power supplies
ground noise has little effect since signal path is isolated from frame
could be as simple as:
measure v from frame to ground
reverse plug if possible and remeasure
use the lowest orientation
U of TX has a good engineering school and does alot of defense electronics work
most likely have experts in emi/rfi noise mitigation
Last edited:
junkhound
Senior Member
- Location
- Renton, WA
- Occupation
- EE, power electronics specialty
U of TX has a good engineering school and does alot of defense electronics work
most likely have experts in emi/rfi noise mitigation
Great suggestion. I have worked with U Texas power electronics engineers in the past and discussed various topics with them at various conferences (e.g APEC, Applied Power Electronics Conference) and they are very knowledgeable.
Some of these non-code discussions are of good general interest, I never even knew there were such things as $19,000 air bearing turntables and tone arms! (looked up some on ebay for the prices).
I actually found the OP's response to me interesting and enlightening, and own opinion is that there is a place here for that type of occasional discussion as it gives insights into what drives some of the technical specifications for various products.
There are likely more than a few grad students and UT that are familiar with both audio and EMI suppression mitigation techniques that would love to earn a few extra $$ from a wealthy benefactor.
PS: my poor redneck upbringing leads me to think anyone who can spend thousands on a turntable (v.s a $1.50 used unit from Goodwill) as wealthy. :lol:
He's got an 'in' too, he's a law prof at U of TX
they probably have an ee audio engineering concentration too
how about 2 mil $ speakers?
https://www.whathifi.com/features/10-worlds-most-expensive-loudspeakers
Last edited:
hbiss
EC, Westchester, New York NEC: 2014
- Location
- Hawthorne, New York NEC: 2014
- Occupation
- EC
I may have missed it in all the OP's verbosity, but other than the transient from the compressor turning on (that seems to have been taken care of) is there an actual problem that needs to be addressed? Or is this another "I want to replace my underground aluminum service wire with copper because it will make my system sound better".
-Hal
-Hal
Problem solving steps
is there a verifed and validated problem? Yes next, no stop
develop solutions next
select best solutions based on expected improvement, cost, etc next
does it make sense to implement? Yes next, no stop
implement
verify problem solved/mitigated
result satisfactory? Yes stop, no restart
A quick visit to equitech's website shows they have several models of the wall cabinet style (http://www.equitech.com/products/wall/wall.html). and some are 230/115V while others are 120/60V (FWIW, IEEE convention for split single phase systems is to state the lower voltage first). Which model do you have? They also offer smaller units.
As stated by Rob (infinity), you cannot have a truly isolated ground and be compliant with Code. Per 250.20(B), a system is required be grounded if in doing so the ungrounded conductors' voltage to ground does not exceed 150 volts.
I thought perhaps you'd grasp what I implied regarding balanced vs. unbalanced (think 3-pin XLR vs. 1/4" ring tip). With the transformer secondary center tap (neutral point) connected to ground, both ungrounded conductors "float" at 60V referenced to the ground at equal magnitude but opposite polarity. Whether this offers any significant benefit is debatable, but audio enthusiasts believe it does. In many cases, transients on the power system are the source of audio signal "noise". Whether a system is grounded or ungrounded, or the method of grounding, will not negate the effects of these transients. Power line filters can reduce if not eliminate some effects. The AC/DC/AC system mentioned by gar is an extreme form of power line filtering.
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
170701-1728
A so called typical professional on this site would have no idea on how to solve the problem. wmhart probably knows more about the problem and how to solve it than 90 % of the electricians in the country.
wmhart isn't asking questions so he can do the job without hiring an electrician, but rather questions so he can teach the so called professional how to correctly do the job to get the intended result.
This not a very usual task, but a discussion that can probably benefit anyone with an interest.
.
A so called typical professional on this site would have no idea on how to solve the problem. wmhart probably knows more about the problem and how to solve it than 90 % of the electricians in the country.
wmhart isn't asking questions so he can do the job without hiring an electrician, but rather questions so he can teach the so called professional how to correctly do the job to get the intended result.
This not a very usual task, but a discussion that can probably benefit anyone with an interest.
.
Thanks, Gar. I work with professionals in all kinds of fields-- I've been down the road on the making of vinyl pellets, on ultra sonic cleaning of vinyl, on purifying water to lab grade. I like to push beyond my limits without necessarily pretending expertise. I'm retired so I have more time to do this now. I write occasionally about such topics, but it is largely a matter of self-interest. And I understand some of the skepticism-- believe me, even with the audio world there are many divides. You guys have given me some stuff to chew on. No disrespect intended or sought.
bill hart
there are entire papers and texts written on the subject by engineers and physicists
just read a few while research martin glasband mfg of equi-power products
if I had a desire to learn the basics of this subject I don't think an internet post is where I would start
not with schools like Pitt, CMU and PSU close by
in the stuff I just read
no math analysis
no measurements
I have no doubt bal pwr attenuates common mode noise
but is the difference audible?
I have a large commercial/residential electrical company that has been working with me-- they are top notch, and I love the guys, but they do this stuff for institutions, not necessarily studios. I had them open a permit for the work. I have also had a couple discussions early on--before I bought this house, with the city electrical inspector's office about isolation transformers, separate service and grounding. I teach at UT, but don't resume til Fall so your point about tapping into some expertise there is a good one. You almost need someone who is multi-disciplinary, or at least willing to take the leap from their specialty to this sort of gray area that is partly science, and partly art.
The music industry guys, as you put it earlier, are all over the place; there's a huge talent pool here, but very little infrastructure compared, say to Nashville. There are some well respected acousticians who will work on site, do measurements, sweeps, and based on their knowledge and ears, help tune a system. They aren't here in Austin, but in Dallas or elsewhere. I have those guys in queue too-- part of it is scheduling, part of it is how far I am willing to go, e.g. tearing out walls, etc.
You'd be surprised- a lot of studios, while wired correctly, don't go to crazy lengths- perhaps in the mastering suite, but a lot of mastering these days is done by a handful of big names who run 3rd party mastering studios and those guys are crazier than me (in a cool way) and are typically on the East or West Coasts. Some of the best stuff isn't even coming out of old school recording studios of the type from the golden era- big Neve boards and 24 track 2 inch tape or whatever-- sometimes it's just a two track, some really good mikes and people who can play. I had a guy here the other night with a 24 track Studer- what a nightmare that thing is--and there aren't a lot of folks who know how to maintain them.
For what it's worth, the electrical subsystem for the hi-fi, with the isolation transformer installed, is as quiet as I could ask for, the room is very quiet, even though we are just a few blocks from South Congress, an extremely dense area where a few of the old time venues are located. My main question was answered- if the ground isn't isolated, what benefit am I getting from the transformer? There have been some pretty interesting and helpful responses.
junkhound
Senior Member
- Location
- Renton, WA
- Occupation
- EE, power electronics specialty
RE: no measurements
Wmhart: as serious an audiophile as you are, you definitely need to invest in a good spectrum analyzer for 5 Hz up to a few hundred kHz.
Of the used stuff on ebay, an HP 141T with low frequency upper plug-in section would be a good low cost high resolution choice.
141T is one of my old favorites, back in 1971 we got the 5th production unit of the 141 series off the HP line as had priority on a defense program. Think it cost $31,000 with full complement of plug-ins (5 Hz to 100 MHz) back then (in 1971 dollars yet, could buy a really nice house for that back then!) and now a working one can be had for a few hundred. Literally worth more than it's weight in gold in 1971.
Having a spectrum analyzer will give you a quantitative assessment of any changes you make to filtering, grounding, shielding, etc. Will also be able to show you an accurate frequency response of any of your systems, and show the PWM frequency of any digital signal also. Looks cool in an equipment rack also!
Wmhart: as serious an audiophile as you are, you definitely need to invest in a good spectrum analyzer for 5 Hz up to a few hundred kHz.
Of the used stuff on ebay, an HP 141T with low frequency upper plug-in section would be a good low cost high resolution choice.
141T is one of my old favorites, back in 1971 we got the 5th production unit of the 141 series off the HP line as had priority on a defense program. Think it cost $31,000 with full complement of plug-ins (5 Hz to 100 MHz) back then (in 1971 dollars yet, could buy a really nice house for that back then!) and now a working one can be had for a few hundred. Literally worth more than it's weight in gold in 1971.
Having a spectrum analyzer will give you a quantitative assessment of any changes you make to filtering, grounding, shielding, etc. Will also be able to show you an accurate frequency response of any of your systems, and show the PWM frequency of any digital signal also. Looks cool in an equipment rack also!
Test equipment is useful
but one must know how to use it
interpret the measurements
use them to make improvements
but we know that
I have recent grad work in signals, after all that is all we are talking about
we have no influence over the ear to brain decoder lol
in his system the most susceptable noise source is the phone pickup or cartridge
2 types, moving magnet or coil
he probably has an mc based on his level of interest
very low output 0.x mv/ma range
does not take much noise to distort
once amplified to line level 2v or so, emi/rfi has much less influence
inconsequential compared to nonlinearity in the electronic stages
the other 3 major kinks in the chain
the source material
the speakers
the room
emi/rfi has no bearing on these
I thought about a dedicated listening space but interest wained
a good ground grid, centerpoise interconnected with 4/0
a 5 or 7.5 kva sola xfmr fed from a 60 in the mdp
60 A panel with 3 x 20
1 each: audio line/signal, audi amp, video use
but one must know how to use it
interpret the measurements
use them to make improvements
but we know that
I have recent grad work in signals, after all that is all we are talking about
we have no influence over the ear to brain decoder lol
in his system the most susceptable noise source is the phone pickup or cartridge
2 types, moving magnet or coil
he probably has an mc based on his level of interest
very low output 0.x mv/ma range
does not take much noise to distort
once amplified to line level 2v or so, emi/rfi has much less influence
inconsequential compared to nonlinearity in the electronic stages
the other 3 major kinks in the chain
the source material
the speakers
the room
emi/rfi has no bearing on these
I thought about a dedicated listening space but interest wained
a good ground grid, centerpoise interconnected with 4/0
a 5 or 7.5 kva sola xfmr fed from a 60 in the mdp
60 A panel with 3 x 20
1 each: audio line/signal, audi amp, video use
Junk Hound: I have a digital one for an iPad that I don't trust, partly because I need a calibrated mike (I have some recording mikes). If I can find what you are describing at reasonable cost, it's invaluable. I'll need how to learn to read the outputs and correlate it with the physical room, but it's good to learn-that's a large part, but not the only part, of bringing in an acoustician. Many thanks,
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
170701-2149 EDT
wmhart:
I believe your original post was mostly about noise or interference getting into your sound equipment, and listening room. Some basic electrical principles apply.
1. Magnetic coupling. To induce a current or voltage in something requires a changing magnetic field coupled to that something. This can occur in several ways.
Move a fixed magnetic field in space relative to the something, or vice versa.
Change the magnetic field intensity produced by the source so that field intensity changes at the something. Typically from an AC current in a conductor.
A single wire with current is big problem (a large loop for the current path). Two wires with exactly the same current but opposite in direction tightly positioned together produce a much lower stray magnetic field when measured a short distance away compared to the single wire.
Romex is fairly good. A twisted pair with a short twist pitch is probably better. A steady DC current is better, but how do you make it steady.
An electrostatic shield has little effect on low frequency magnetic fields.
Magnetic materials can provide magnetic shielding.
Transformers generate leakage flux.
You can probably assume magnetic field intensity drops off at least as fast as 1/r, where r is the radial distance from the source.
2. An electrostatic shield operates by providing a conductive path physically between an electric field source and the said something being shielded.
Two conductive plates separated by an insulator creates a capacitor. A capacitor conducts an AC current.
If a transformer is built without an electrostatic shield between primary and secondary, then a capacitance exists between primary and secondary that can couple high frequency noise between the pri and sec.
Insert a shield between pri and sec and connect the shield back to the source, and little noise is coupled to the secondary
.
wmhart:
I believe your original post was mostly about noise or interference getting into your sound equipment, and listening room. Some basic electrical principles apply.
1. Magnetic coupling. To induce a current or voltage in something requires a changing magnetic field coupled to that something. This can occur in several ways.
Move a fixed magnetic field in space relative to the something, or vice versa.
Change the magnetic field intensity produced by the source so that field intensity changes at the something. Typically from an AC current in a conductor.
A single wire with current is big problem (a large loop for the current path). Two wires with exactly the same current but opposite in direction tightly positioned together produce a much lower stray magnetic field when measured a short distance away compared to the single wire.
Romex is fairly good. A twisted pair with a short twist pitch is probably better. A steady DC current is better, but how do you make it steady.
An electrostatic shield has little effect on low frequency magnetic fields.
Magnetic materials can provide magnetic shielding.
Transformers generate leakage flux.
You can probably assume magnetic field intensity drops off at least as fast as 1/r, where r is the radial distance from the source.
2. An electrostatic shield operates by providing a conductive path physically between an electric field source and the said something being shielded.
Two conductive plates separated by an insulator creates a capacitor. A capacitor conducts an AC current.
If a transformer is built without an electrostatic shield between primary and secondary, then a capacitance exists between primary and secondary that can couple high frequency noise between the pri and sec.
Insert a shield between pri and sec and connect the shield back to the source, and little noise is coupled to the secondary
.
junkhound
Senior Member
- Location
- Renton, WA
- Occupation
- EE, power electronics specialty
steady DC current is better, but how do you make it steady. etc
Just for kicks (a fun thread to me) will throw out a couple of actual extremes:
Once put RG-11A inside 3/4" rigid steel conduit with couplings circumferentially welded to keep a missile silo protected from elec and magnetic fields hypothesized from nuke bursts close enough to tilt the silo.
Worked on a star-wars type 3 MW radar years ago that used 126 each phase shifted delta-wye transformers to feed the rectifiers - followed by a low ripple switching regulator(s) and then a large quantity of parallel 2 stage 120 dB per stage, 600 kHz bandwidth linear regulators to get dc ripple below 6 microvolts to feed the antenna mimics.
Really don't think that level of noise reduction is needed for audio :huh:
Just for kicks (a fun thread to me) will throw out a couple of actual extremes:
Once put RG-11A inside 3/4" rigid steel conduit with couplings circumferentially welded to keep a missile silo protected from elec and magnetic fields hypothesized from nuke bursts close enough to tilt the silo.
Worked on a star-wars type 3 MW radar years ago that used 126 each phase shifted delta-wye transformers to feed the rectifiers - followed by a low ripple switching regulator(s) and then a large quantity of parallel 2 stage 120 dB per stage, 600 kHz bandwidth linear regulators to get dc ripple below 6 microvolts to feed the antenna mimics.
Really don't think that level of noise reduction is needed for audio :huh:
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
1270702-0853 EDT
junkhound:
My comment on how do you get a constant DC current was meant to discuss current and not voltage. A variation in a DC current will induce a voltage or current in a conductor somewhere. If one uses a DC power distribution system for audio, then there will be a varying magnetic field from that distribution system from audio signal level changes.
Thus, in a DC distribution system it is important to keep the source and return conductors close together.
.
junkhound:
My comment on how do you get a constant DC current was meant to discuss current and not voltage. A variation in a DC current will induce a voltage or current in a conductor somewhere. If one uses a DC power distribution system for audio, then there will be a varying magnetic field from that distribution system from audio signal level changes.
Thus, in a DC distribution system it is important to keep the source and return conductors close together.
.
- Status