Question about how to wire an Isolation Transformer

[gpapag]

Hi all

I would like your advise on the proper way to wire the secondary of an isolation transformer, so as not to violate the electrical safety of the equipment to be connected to it ( audio gear) and to retain - as much as possible - the isolation from the noise which is superimposed on mains wiring.

I spent a few hours reading books and some more hours on the internet, but the opinions are really conflicting.

Attached is a sketching of 3 different connections.

Please commend

Best Regards
George

[richie00boy]

1
2 is bad
3 is dangerous

[zigzagflux]

Why are 2 and 3 not equivalent ? They are both labelled E, yes? Or are you saying the output end of E is not bonded to the panel earthing conductor? It must be, no exceptions.

With very rare exceptions, 2 is the only legal method (in the USA) if you are feeding multiple equipments with this power supply. If it is only being used inside a single enclosure, then 1 can work. However, it is typically a good idea to earth the equipment at some point, even if through a resistor. A completely floating supply has its own set of problems.

What do you intend on this supply feeding, specifically?

[gpapag]

Zigzagflux
I intend to use such an isolation transformer (IT) to feed the pre-amp, main amp, tuner and CD drive .

Richie00boy
In fact the wiring per sketch No 1 is the way that any step-up or step-down transformer is wired in any linear PSU

I have an isolation transformer (IT) in my lab, which is wired as in the No 1 sketch. I use it during repairing equipment for reducing the likelihood of becoming a spirit. So far so good.
But today morning during some mains noise monitoring, I thought of measuring the noise at the output of this IT.
Then to my surprise, I noticed that the voltages of the thus wired secondary were behaving quite strangely (strangely to me).
See attached photo.

Is it normal? Is it safe?

When the secondary is wired as per No 2 and No 3 sketches, voltages are as on the primary. No surprise.

Regards
George

[Richard Ellis]

I would bet the wiring of the output of your IT is currently correct. Did you notice your measurements on the secondary don't add up to your 229Vac?
Think how the circuit works as in electron flow. The measurements from Live or Nuetral to Earth are capacitatively(?) coupled...I think that your DVOM puts such a light load on the circuit...that you can get any reading. I would bet if you loaded the circuit just a little you won't read any voltage at all. Wouldn't be baffling if you had a cheap meter & you tested both 'L' & 'N' ...and it gave you no readable voltage?
You should be able to pull down that reading with microamps & the voltage would drop.to nothing.
As has been said over & over, rectifying AC from the wall is dangerous because the live side to earth AND/OR Nuetral is hot.
Thats my take on it....I'm probably wrong in my assumptions, but....
___________________________________________Rick... ..........

[zigzagflux]

Originally posted by gpapag
I intend to use such an isolation transformer (IT) to feed the pre-amp, main amp, tuner and CD drive.

Well, not to be a party pooper, but at least in America, you are required to ground the neutral, by NEC 250-20(b)(1). The only exception that I have been able to identify would be balanced systems, which have the center tap of the transformer secondary grounded, so you have 60-0-60V. Note: this exception is still a grounded system. Since your isolation transformer does not appear to have a secondary center tap, I won't bother discussing this option.

I am not sure where the idea came from that a completely isolated system is the ultimate goal, but it's not. Ungrounded systems are prone to transient overvoltages and nondesirable capacitive coupling effects (which is why the Code limits their use).

If the transformer was exclusively used inside one single component, such as your amplifier, then you are able to leave it isolated (or resistance grounded as mentioned above). This is found in section 450, IIRC. Section 640 addresses some specific allowances for audio equipment, but there is no allowance for isolated systems feeding a power strip to multiple devices.

Others can debate the pros and cons of grounding the secondary from an audible standpoint; but sorry to say, legally it must be grounded. I am open to someone showing me a Code exception that I have missed (they are very easy to miss).

[Iain McNeill]

Yes, earth must be earth. /Do not defeat your safety ground!

The strange readings you get are as Richard Ellis alluded. The uA leakage currents flowing through your voltmeter are causing an AC voltage to be read. It's an artifact because the secondary if floating.

The requirement to tie neutral to earth is only at the power panel. Never do this anywhere else.

The key reason to use an isolation transformer is to protect yourself from live chassis. If you are holding a grounded (earthed) pipe for instance and touch a live chassis, the isolation transformer will prevent electric shock. You'll still get a tingle from the leakage current but it won;t stop you heart.

Remember power engineering 101 - keep one hand in pocket!

To filter noise from the mains, you're better off with filters to target the band of interest: common mode chokes, small properly rated caps. Toroids have high bandwidth and couple a lot of cr@p. E-I iron laminated cores are pretty narrow bandwidth and help.

The interwinding screen (faraday shield) only stops common mode capacitively coupled noise, not the predominant dirty sine + RF that usually comes out the wall

[VictoriaGuy]

Originally posted by Iain McNeill

The key reason to use an isolation transformer is to protect yourself from live chassis. If you are holding a grounded (earthed) pipe for instance and touch a live chassis, the isolation transformer will prevent electric shock. You'll still get a tingle from the leakage current but it won;t stop you heart.
[/B]

Exactly.
There won't be appreciable current flowing between the isolated 'live' and the earth connection.
So, once we have moved the isolated supply off the chassis with the possibility of accidental connection to the live mains (not isolated) circuit, what's the point of the earth conductor? It seems to me that you are really dealing with a two-conductor, isolated system which is not referenced to ground (or neutral).
So once you are on the isolated side of the transformer, you need to pay attention to the fusing arrangement, since you can no longer depend on the main breaker/fuse for protection.
In spite of code 'requirements', isn't the earth really just a decoration after the isolation transformer?

BTW, there is a lot of info on this topic in marine wiring systems, where isolating the boat from the electric currents flowing around the marina can be a big deal.

Slightly (or more) baffled,
John

[zigzagflux]

Originally posted by Iain McNeill
The requirement to tie neutral to earth is only at the power panel. Never do this anywhere else.

Emphatically wrong!! The transformer is classified as a separately derived system, and it is required to ground one conductor. This conductor then becomes named the neutral. There are no exceptions to this; if we need to get a moderator involved in order to address the illegal suggestions that are being made, I have no problem with it.

[Iain McNeill]

This is good, we have a quorum of experience.

Regarding safety ground on the isolated side. I will clarify my point by saying that safety ground is a fundamental protection for persons who may come into contact with the equipment. In the event of an IT pri-sec short, safety ground will still protect the casual user.

What you do in the private of your own lab is your own concern . In this domain I agree that there are benefits to balanced power distribution

zigzagflux:
If you enforce the neutral-ground tie on the isolated side, then there must be some requirement to maintain polarity right? It sounds like you have experience in the power industry that I/we can learn from. Can you give us some more info?