isolation

Status
Not open for further replies.

BCK

Member
I have a few questions on isolation transformers. I have been looking into different models for my lab. Some have ground leads for both primary and secondary while others only have leads for the hot and neutral.

I have read that isolation transformers can help remove ground loop current problems as well as reduce shock hazards, but I don't understand how. Can anyone explain to me how they work, especially what happens with the grounds of the primary and secondary.
 

LarryFine

Master Electrician Electric Contractor Richmond VA
Location
Henrico County, VA
Occupation
Electrical Contractor
The ground is of no relevance with isolation or any other type of transformer. An isolation transformer merely indicates that it has a 1:1 voltage ratio; its purpose is isolation, and not voltage change.

Actually, all transformers are isolation transformers, unless you make a connection between the primary and secondary conductors, as you would with buck-boost transformers.

The equipment grounding conductor is separate from all others, except that it just so happens that the 'neutral' conductor is bonded to the equipment grounding system back at the main service disconnect.

So, just consider that the isolation transformer is a way to effectively "un-do" that bond for your circuit, so there is no direct connection between the secondary circuit and the system's equipment ground, and the earth all around you.
 

iwire

Moderator
Staff member
Location
Massachusetts
LarryFine said:
The equipment grounding conductor is separate from all others, except that it just so happens that the 'neutral' conductor is bonded to the equipment grounding system back at the main service disconnect.

What about the bonding jumper on the secondary side of the trans?

So, just consider that the isolation transformer is a way to effectively "un-do" that bond for your circuit, so there is no direct connection between the secondary circuit and the system's equipment ground, and the earth all around you.

Not sure if that is legal without ground fault indicating equipment.

A isolation trans can still 'isolate' even with a grounded secondary.
 

BCK

Member
This is where my confusion comes in. Why is there a ground terminal on the secondary of many isolation transformers if it is supposed to decouple the two circuits and, if there should be no ground on the secondary, how does this not create a shock hazard? I couldn't find anything in the NEC on it, but I'm sure it's in there.
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
LarryFine said:
Actually, all transformers are isolation transformers, unless you make a connection between the primary and secondary conductors, as you would with buck-boost transformers.

That is not exactly true, if there is an EGC on the primary side bonded to the enclosure and the secondary has a bonding jumper, there is more than magnetic coupling between the systems

LarryFine said:
The equipment grounding conductor is separate from all others, except that it just so happens that the 'neutral' conductor is bonded to the equipment grounding system back at the main service disconnect.

Exactly, and generaly speaking with code compliant bonding the term isolation is just that, a term that is not acurate.

LarryFine said:
So, just consider that the isolation transformer is a way to effectively "un-do" that bond for your circuit, so there is no direct connection between the secondary circuit and the system's equipment ground, and the earth all around you.

Which is a violation unless the proper monitoring is installed (a LIM) and this lab is in a hospital.

Roger
 

iwire

Moderator
Staff member
Location
Massachusetts
roger said:
That is not exactly true, if there is an EGC on the primary side bonded to the enclosure and the secondary has a bonding jumper, there is more than magnetic coupling between the systems

I agree that there is more than magnetic coupling between the two systems but IMO they are still electrically isolated from each other.

With only one conductor from each system connected together there is no circuit between them so one system can not interfere the other.

If I had two generators each 100' from the other both are connected to earth.

Would you call those systems isolated from each other?

How about if I bond both GESs together with a copper conductor?

Are they still isolated from each other?

JMO, Bob
 

BCK

Member
As well as in hospitals, isolation transformers are used often for computer labs to protect the equipment and in audio systems to avoid ground loop noise. Do they pass the ground through from primary to secondary for systems not monitered by an LIM? Is the neutral in the secondary then unbonded and just floating away from ground? Isn't that a problem?
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
iwire said:
With only one conductor from each system connected together there is no circuit between them so one system can not interfere the other.

Bob, an intentional circuit is not the issue, it's the term "isolation".

If there is a surge on the primary EGC it can introduce the same surge into the windings of the secondary. If there is anything tieing the primary and secondary sides together, isolation (in reality) is not acheived.

iwire said:
If I had two generators each 100' from the other both are connected to earth.

Would you call those systems isolated from each other?

Well, it depends on how dry the conditions are and how sandy the earth is. :wink:

iwire said:
How about if I bond both GESs together with a copper conductor?

Are they still isolated from each other?

No, they are common, not isolated

Bennie would have been all over this.
icon12.gif
It's hard to believe it's been just over 3 years ago.

Published in The Oregonian on 8/4/2004.

[FONT=Verdana, Helvetica, sans-serif]quote:[/FONT] [FONT=Verdana, Helvetica, sans-serif]PALMER, BENNIE RAY Bennie R. Palmer of Milwaukie, died peacefully July 28, 2004, while supported by family and friends after a hard fought battle with lung cancer. Bennie was born December 10, 1932, in Wilson, Okla., and married his wife, Lois J. Palmer, 22 years later. Bennie and Lois planned to celebrate their 50th wedding anniversary December 24th of this year. Together, Bennie and Lois raised five children, four of whom reside in the Portland area. During his life Bennie served our country while enlisted with the United States Marine Corps as an AV electrician and continued his work as an electrician for the balance of his career through the IBEW Local # 11 and later, as an electrical inspector for the City of Tigard. Survivors include wife, Lois; son, Billy; daughters, Lynda Gentile; Judy P. Adler, and Raelynn P. Rackerby; seven grandchildren; one great-granddaughter, three sisters, and numerous extended family members. In lieu of flowers, donations can be made to the Oregon Cancer Society in honor of Bennie.
[/FONT]​
Roger
 

iwire

Moderator
Staff member
Location
Massachusetts
roger said:
If there is a surge on the primary EGC it can introduce the same surge into the windings of the secondary. If there is anything tieing the primary and secondary sides together, isolation (in reality) is not acheived.

Roger I may be missing something but if there is a surge on the primary it will effect the output of the secondary regardless of it having a bond or not.

It's hard to believe it's been just over 3 years ago.

I was thinking of something he said not more than a few days ago.

'The NEC wants to shut the power off, the NESC wants to keep it on' (Or something close to that)

BTW, I do remember our discussions on SDS / the connection etc. I may be having a change of heart. :smile:
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
iwire said:
Roger I may be missing something but if there is a surge on the primary it will effect the output of the secondary regardless of it having a bond or not.

Bob, a surge through the windings of the primary will elevate the voltage on the secondary proportionately, but that would be through magnetic coupling.

A surge type of fault on a common condutor (the EGC and bonding jumper) to the primary and then to the secondary would simply be tieing both sides together physically.

BCK, true isolation systems used in hospitals are not allowed for use in other locations that I am aware of.

Roger
 

iwire

Moderator
Staff member
Location
Massachusetts
roger said:
Bob, a surge through the windings of the primary will elevate the voltage on the secondary proportionately, but that would be through magnetic coupling.

Agreed.

But as best as I can figure that is the only way a surge on the primary can effect the secondary.

A surge on the primary side in relationship to the common EGC should not effect the secondary side at all.
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
iwire said:
A surge on the primary side in relationship to the common EGC should not effect the secondary side at all.

Bob, I do agree that it shouldn't, but I still see a common conductor to both sides as defeating the term "Isolated".

I can see an Ungrounded Delta to an Ungrounded Delta as fitting the discription as well as the aforementioned Hospital Isolation type of systems, but I can't agree that all transformers (not talking about auto transformers either) are isolation transformers.

Roger
 

iwire

Moderator
Staff member
Location
Massachusetts
roger said:
but I can't agree that all transformers (not talking about auto transformers either) are isolation transformers.

Thats cool. :cool:

My view changed from 'not isolated' to 'electrically isolated without being physically isolated'.:)

JMO, Bob
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
iwire said:
Thats cool. :cool:

My view changed from 'not isolated' to 'electrically isolated without being physically isolated'.:)

JMO, Bob

I wonder if we can get that included into some formal terminology. :grin:

Roger
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
iwire said:
To late....it has already happened. :grin:

Didn't a CMP already make that happen?

I hate that, i must of missed it.
icon9.gif
:)

Roger
 

dereckbc

Moderator
Staff member
Location
Plano, TX
OK, I have to jump in this as this area is my field of expertise. My comments will pertain only to the subject of which the OP is using the transformer for. So let?s untangle some of the confusion. I will start with why isolation transformers are used.

1. Reduce Common Mode Noise. Isolation transformers are the second best way of reducing CMR, second only to a true dual Conversion UPS or Motor Generators. Your average ordinary everyday transformer be it an isolation, step-up, or step down types will reduce CMR by a minimum of 40 decibels (10, 000:1). Special ferroresonant types made for the purpose can go as high as 160 decibels 10,000,000,000,000,000:1.

2. To establish a new reference point. Notice I did not say ground reference point however that is the most common method used.

3. For poly-phase transformers to prevent harmonics or trap then on the secondary of the transformer and prevent them from backing up. This is why you will see then used in data centers on lighting and rectifier circuits

That is it. Even though the primary feeder EGC maybe common to the Xo on the secondary does not couple the primary to the secondary together. There is no complete circuit for current to flow. So if there were a disturbance on the primary EGC, it has no effect on what is going on in the secondary because even though there may be a spike or rise in voltage, the new N-G bond reference point just floats up and falls, and whatever equipment is connected to the secondary never sees it.

Now let?s go back to that ferroresonant transformer because it can also reduce differential or normal mode noise. These have special isolation shields between primary and secondary windings giving them superior CMR. They also have a resonant tank circuit to more or less buck and boost the output voltage in the event there is some voltage or frequency variation in the primary. So in effect if the input voltage sags, it boost the output voltage, and vice versa so the output remains at a constant voltage.

That is all I got for now.
 

dbuckley

Senior Member
Excellent Dereck.

Now we need to OP to supply some more info on what he intends to supply, and what he thinks installing an isolating transformer will do for him. Quite, how, in detail, you use a isolating tranny has many subtleties...
 

BCK

Member
I was still wondering about the neutral.

The neutral to ground bond is broken and the neutral is then isolated. The neutral will then have no reference and can drift. In my mind this is a problem, that is, having an unbonded neutral. The 'grounded' conductor would then not be grounded, am I right?

I was hoping that the transformer would help reduce ground loop issues and other power supply noise feeding to an o-scope and other signal measurement equipment.
 
Last edited:

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
BCK said:
The neutral to ground bond is broken and the neutral is then isolated. The neutral will then have no reference and can drift. In my mind this is a problem, that is, having an unbonded neutral. The 'grounded' conductor would then not be grounded, am I right?

The neutral to ground bond provides a well identified connection between the transformer secondary and ground. If this is the _only_ connection between the transformer secondary and ground, then breaking this connection would permit the secondary to drift.

However if the neutral to ground bond in the transformer is the _only_ connection between the transformer secondary and ground, then you cannot have a ground loop problem!

You will only have a ground loop problem if there are _multiple_ connections between the secondary circuit and ground.

BCK said:
I was hoping that the transformer would help reduce ground loop issues and other power supply noise feeding to an o-scope and other signal measurement equipment.

The problem that you have here is that your signal measurement equipment has a ground connection on the signal side which is also connected to the electrical system ground. Now you have at least _two_ connections between the electrical system and ground. In theory you could break this loop on the power side by eliminating the ground-neutral bond, and the transformer secondary would remain referenced to your signal side ground.

My preference would be to use instruments with internal power supplies that provide this isolation. Not only does this eliminate the big ground loop between your instruments and the power supply, but it also eliminates the small ground loops between your various instruments. Even better is to provide the necessary isolation at the level of your signal transducers. Now every measurement is a differential measurement relative to the local relevant ground, rather than having a bunch of 'ground' referenced signals with a poor defintion of what ground is.

-Jon
 
Status
Not open for further replies.
Top