Discussion of AC mains isolation transformers

[Speedskater]

I hadn't noticed any of this in the old thread.

So random thoughts on Isolation Transformers.

a] There are larger permanently installed Iso Trans that power the entire audio system and there are small portable Iso Trans that power a few components. These are two very different tasks. We must keep thoughts about them separate.

b] Like John S, I too have a 5kVA Topaz transformer. It's near the audio equipment, but in the basement so a small amount of buzz isn't a problem.

c] The biggest virtue of a large Iso Trans is that it can be wired as a Separately Derived System near the audio equipment. It doesn't even need to have a lot of isolation! It's the Separately Derived System part that's important.

d] Instructions for installing a Separately Derived System can be found on page of this Middle Atlantic paper.

"Power Distribution and Grounding of Audio, Video and Telecommunications Equipment - White Paper"

Integrating Electronic Equipment and Power into Rack Enclosures

Optimized Power Distribution and Grounding for Audio, Video and Electronic Systems

White Papers

[Speedskater]

In situations like this we can treat Watts & Volt/Amps as about the same (even though they are not really).

But power transformers are rated for resistive loads like heating units or old fashioned light bulbs. For big amplifiers leave a large safety factor. Use 60% or 50% of the rating.

[Speedskater]

has anybody measured noise on their AC line?

It takes some engineering knowledge, some hands-on skills and the correct test equipment. Doing it incorrectly could cause equipment damage (hi-fi and/or test), human injury or death. And experience to understand what the readings really means.

[Speedskater]

measuring noise on the AC line makes sense before spending many thousands of dollars to reduce noise on the AC line

Power line noise is ever changing. Most of it comes from the hi-fi system components, other near-by entertainment equipment, appliances and HVAC.

 

Above John S. has instructions for measuring noise voltage. But of more concern are noise, interference and leakage currents.

 

It's Ham radio operators that are concerned about the noise voltages in the MHz range.

[Speedskater]

A proper way to install an isolation transformer, wired as a Separately Derived System.

From the Middle Atlantic " Power Distribution and Grounding of Audio, Video and Telecommunications Equipment" White Paper

http://www.middleatlantic.com/resources/white-papers.aspx

[Speedskater]

Well a DC supply may have to deal with noise well up into the megahertz range at the circuit end. As semi-conducts turn on & off they generate this noise. But the higher the noise frequency, the closer to the semi-conductor the filter needs to be.  Meanwhile a filter at the DC supply end can only deal with lower frequencies.

[Speedskater]

  Obviously DC doesn't pass through any transformer.  What is called DC off-set is asymmetry in the AC waveform. It's just harmonic distortion caused mostly by SMPS's. Well in the old days there were heating units (mostly hair-driers) that would go to low power by placing a diode in series with the AC line, so the heater would see half-wave AC. But it's been decades since these were popular. 

   Anyway the harmonic distortions add together to create this off-set. Audio editing programs have a function to measure this DC off-set in audio files.

[Speedskater]

What is an isolation device?

What is an isolation circuit?

By that I mean, how are the Hot, Neutral and Safety Ground/Protective Earth connected?

In the case of an isolation transformer, is it:

A portable corded transformer?

Or a permanent hard wired transformer?

[Speedskater]

How many power amplifiers on one Topaz transformer, depends on the size of the amps and of  the transformer.

It's best not to load a transformer anywhere near it's rating, but most audiophiles don't push their big amps hard.

Home theater buffs on the other hand push their amps to the limit.

[Speedskater]

11 hours ago, Daudio said:

Until someone more knowledgable comes along, I have a few throughts for you.

I think you would be Ok using 2 isolation xformers, 1 for the front end, and 1 for the amps.  What I would try and do is to have the AC lines about the same length and gauge between the Iso tranny and the back of your gear (IEC input).

The point is to try and minimize leakage loop currents, by making the AC wire impedences as equal as possible between each component. So, similar Iso trans, similar cable runs, power strips, and even power cords (if possible). You still might have a little leakage between the front end and amps, but probably too small to worry about.

Could you hook up everything to the one Iso xformr, piled on the floor temporarly, just to see if it makes a difference for you ?

I have read the idea of having all the AC Safety Grounds equal lengths.  But different components won't have the same leakage currents (not in frequency, phase or amplitude) so you can nothing by this extra work.  Even identical amps, unless they are playing a mono signal won't have the same leakage currents.

 

I would try to reduce the lengths of the Safety Grounds from component to component.

[Speedskater]

6 hours ago, Daudio said:

 

Still waiting for someone more knowledgeable to come along. Till then could you refrain from muddying the waters with your idle speculation ?  TIA

 

True, I'm not an expert. For expert knowledge, you might read the papers and books of the real experts.

Experts like:

Ralph Morrison

Henry Ott

Neil Muncy (RIP)

Jim Brown

Bill Whitlock

[Speedskater]

With a good XLR balanced interconnect system, AC power system noise is greatly reduced.

Note that not all components have correctly wired XLR connectors.

[Speedskater]

31 minutes ago, Jud said:

 

Yep, though to be absolutely precise, I’ve got two power strips (8 receptacles per strip and >8 plugs) and my amp, each of which is plugged into the same wall receptacle on the same circuit.  So no isolation among components,

....................................................

Yes, that's a very good way to connect components to AC power.

[Speedskater]

2 minutes ago, Jud said:

 

This says to me that you will (also) have to determine whether it’s more important to have shorter power cables or shorter speaker cables.

That's a tough call. Short power cords (from component to component) is always good. Depending on the loudspeaker impedance curve, short speaker cables can also be good.

Balanced or unbalanced interconnects work their way into the equation.

[Speedskater]

Note that NEC & UL rules don't permit running cords thru walls or floors.

The home theater folk get around the rule by using products the have wall mounted plugs and sockets. so they can use normal in-wall wires.

 

Also there is almost zero chance of hearing a difference in these different (good engineering practice) AC wiring layouts. Not likely to measure a different either.

[Speedskater]

Solar panels are a known source of noise & interference. Jim Brown a pro-audio EMI/RFI expert writes:

 

Solar Power Systems can be very strong noise sources as a result of poor design, poor installation,
or both. The best charge current regulators are DC-DC converters, and most are noisy. DC – AC
inverters that provide 120VAC are also often noise sources.  Both charge and discharge circuits
carry large pulsed currents with strong harmonics; those harmonics will radiate if the current flows
through large area magnetic loops. A large system with all wiring in steel conduit has the best
chance of being quiet, provided that the conduit is continuous and bonded to all equipment enclosures at both ends.

 

But on the other hand he also writes:

 

.  Solar power systems should always be wired with twisted pair. Genesun, a relatively new company makes a line of MPPT (the most efficient type) solar charger regulators for small systems that is quiet enough for nearly all installations

 

There has been a trend in recent years to solar panels with self-contained regulators and inverters;
properly built and installed, these can greatly reduce the loop area and thus the noise. Because the
power leaves the panel as 120VAC (or even 240V), wiring from the panel is usually in conduit. And
because the output of the panel is at the higher line voltage, the current is much less than if it were
at battery voltage. RF trash produced is directly related to current, so all of these factors can combine
 to result in less radiated noise if the units are well designed.

 

 

 

[Speedskater]

It's probably not a true DC current. It's a DC offset in the AC waveform. Many audio editing programs can test for this in a music file. The different harmonic distortions on the power line can sum together to make an asymmetrical waveform.

[Speedskater]

13 hours ago, jamesg11 said:

Hmmm ... is the 'explosive' risk exacerbated or ameliorated in 240V circuits?

Higher voltages bring higher dangers.

But in the US type system, the 240V is line to line. Either line is still 120V to ground.

[Speedskater]

That paper was written over three decades ago. Things do change, like rules about capacitors and ground rods.

[Speedskater]

On 7/27/2017 at 9:42 AM, Speedskater said:

That paper was written over three decades ago. Things do change, like rules about capacitors and ground rods.

Capacitor Rules:

Capacitors connected to the AC power line must follow the safety rules. (this includes isolation transformer secondaries) Across the power line to line it's type "X" and for line to ground it's type "Y".

 

I'm not getting into DC blockers.

Be back later.

[Speedskater]

After further thought about the capacitors in a DC Blocker.  Because the electrolytic capacitors in a DC Blocker are neither connected line-to-line nor line-to- ground, the "X" & "Y" type safety rules don't apply.

However that doesn't mean that connecting electrolytic capacitors to an AC line is completely safe! Over voltage or reverse voltage can cause an electrolytic capacitor to explode!  So when trying out a DIY DC Blocker, stand well clear of the unit.

[Speedskater]

Yep, bad things can happen in a DC Blocker.

Heck, bad things can happen when there are any DIY add-ons to the AC power system.

[Speedskater]

Most good transformers in most situations don't need a DC blocker.

If there is a DC offset problem, it is caused by things like high-tech lighting systems, appliances, heating & air conditioning systems and motor drives. Most of those items start & stop so the problem will come and go.

[Speedskater]

Thinking about it further, why would one think that a good grounding system, wouldn't have a measurable improvement  in the signal-to-noise ratio?

===================

But on the other hand, two different just audible background noises, that measure about the same, could have different subjective ratings.

[Speedskater]

Isolated Ground receptacles/systems are only useful in buildings that have metal conduit/tubing or metal building framing. They prevent ground currents from other circuits from using your interconnects as a path back to their voltage source.