isolation transformer? 1

[bigheadted]

can somone breifly explain the term 'isolation transformer'
to me, surely all transofmers with two electrically isolated windings are isolating or does it have something to do with the earth connections, if so what does this do?
thanks
rich

 

[RalphChristie]

See thread238-145570

 

[Engcan]

I apologize for stuffing my question here, but I came across some thing similar. I was on a site visit doing survey for the loads. I found a new installation on all 1o floors and it was all new panels supplied by mitigation transformers 600V/120/208V and will be also called an ISOLATION transformer case ?(Please correct me if I am wrong). When we asked the client about the drawings for this installation, I came across some thing, which I could not reason out. You guys might be able to advise.
The building has 10 floors and each floor has one transformer and when I saw the drawings there was a column in the description of these transformers, which says a phase difference of 0 degress and then 30 degree for next floor and then again 0 degree for next floor transformer. Could you suggest what was happening.
Thanks

 

[dpc]

The transformers with zero degrees of phase shift were delta-delta or wye-wye. The transformers with 30 deg shift were delta-wye or wye-delta.

The intent, I guess, is to reduce harmonics reflected back through the primaries by phase shifting, although I don't see how it going to help much with a bunch of small single-phase switching power supplies that will be all over the map anyway.

But maybe someone will enlighten me.

 

[Engcan]

Thanks for the reply, but honestly I did not exactly get it. Harmonics reduction, I understand but I am sure as the panels supply single phase loads, so the config. cannot be delta -delta as it will not help for single phase loads,
so it can be star-star in case of zero phase shift. But how reduction in hamronics flowing into primary will help.In case of 30 degree phase shift, it seems logical delta -star configuration as the harmonics will stuck up in delta and will not go back. But how about the first option.

Thanks

 

[davidbeach]

Harmonic mitigating transformers can be built in many strange configurations, delta primary and mostly wye secondary with enough zig-zag to get the desired angle. Fifth harmonics though a 30 degree (fundamental) shift are shifted 30*5*(-1), or -150 degrees while the seventh is shifted 30*7*1 or 210 degrees. The (-1) in the fifth is due to the fifth being a negative sequence harmonic and the final 1 for the seventh is due to the fifth being positive sequence harmonic. When you then consider that the source is shifted -30 degrees relative to the load and add than to the above numbers, 5th harmonics from the secondary are shifted -180 degrees getting back to the primary and 7th harmonics from the secondary are shifted 180 degrees getting back to the primary. Adding the fifth and seventh harmonics from a zero and a 30 produce cancellation to the extent that the harmonics match in magnitude and phase angle. Not perfect cancellation, but much more cancellation than would happen if all of the transformers have the same phase shift.

 

[Engcan]

Hi david beach,
I found a reference of your explanation on Square D
website

http://ecatalog.squared.com/pubs/El...ormers/Non-Linear Transformers/7400DB0301.pdf

and it advises feeding two panels with separate mitigation transformers with different phase shifts. I tried to digest it but will this be the right statement to make that,
when we supply a computer, fax or a copier panel through power free from harmonics, this is only true till the secondary of that transformer who is feeding these panels.
The harmonics are still present at the service entrance distribution switchgear and primary of this transformer.
On Page 4 and fig. 3 of this link above,
it advises to feed from same bus two different panels with different configurations, what I could not understand that
if on the service entrance, the main transformer is
DELTA-STAR, there will be no third harmonics in building or system. My limited knowledge tells me about this subject that what ever is on the secondary of the transformer(Configuration) feeding the panel, It will only make a difference in secondary not in primary, so how this arrangement helps. Please correct me if my concept is screwed.

Note: Waross, I would appreciate if you could contribute.

 

[waross]

Engcan
Well david beach has said it better than I could. I am learning here also. That said, I'll go out on a limb and make a couple of comments on the Square D information that may make it easier for you to grasp.
1> Page 8; Source Impedance. Note this is the actual impedance of the transformers, not the percentage impedance. The actual impedance is what limits the current. Transformer pecent impedance is a ratio that is dependant on the actual impedance and the rated full load current.
Consider a 50 KVA transformer and two 25 KVA transformers, all with 2% impedance. The actual impedance of one of the 25 KVA transformers is double the actual impedance of the 50 KVA transformer.
I believe this is to point of the comment re higher impedances in the Square D link.
Splitting a load onto two smaller transformers will result in better impedance mitigation of harmonics.
2> Combining waveforms;
Square D link, Page 7, figures 7 and 8.
To understand these figures, imagine a vertical line through both figures. (or print it out and actually draw a line)
Draw the first line through the first (from the left) index mark on the zero current line. (Through the center of the first large positive waveform.) You will notice that each point on the upper wave form (fig. 7) is a result of the signed addition of the corresponding points on the lower wave form. (fig. 8) At the point where the fundamental wave form is at a positive peak, both harmonics are at a negative peak. When they are added the harmonics are negative and actually subtract. That gives us the notch in the wave form in fig.7.
Fourier Transformation is a mathematical tool that will analyse the wave form shown in fig.7 and derive the component wave forms shown in fig. 8.
A question davidbeach;
There is reference to harmonic distortions affecting other customers on th e same supply.
Is my understanding correct on this
1> With a theoretical infinite supply source, one consumers harmonic distortion will not affect another customer.
2> In the real world, the distorted currents will cause distorted voltage drops in the common source impedance, and this distorted voltage will be the effect on the other users.
3> The common source impedance will be primarily the supply transformer impedance plus the impedance of any shared conductors.
4> The voltage harmonic distortion will be inverted. That is where ther is a peak in current, there will be a corresponding higer source voltage drop. The effects will be that current peaks will be reflected as voltage valleys and current valleys will be reflected as voltage peaks.
Have I got this right?
Thanks in advance davidbeach.
yours

 

[davidbeach]

I have not taken the time to look at the Square D material, but have generally found their stuff to be fairly good.

Engcan, I think you have things backward. The utility sources the fundamental, and the load sources the harmonics. The point of using two harmonic mitigation transformers that have a mutual 30^o phase shift is to cancel the harmonics on the primary of the transformer. As waross points out, voltage distortion comes from current distortion causing non-sinusoidal voltage drops across the source impedance, but the source impedance is everything back to the "infinite bus", transformers, conductors, the whole works. The last transformer may be only a small percentage of that total impedance.

A point to remember on all power quality problems - There is no power quality problem that can not be solve by a sufficiently stiff system. Put in other terms, the lower you can keep your source impedance, the fewer your potential power quality problems. Other problems perhaps, but not power quality.

 

[Engcan]

Thanks for the replay David beach.
You might be right, I have things backward.

After reading and recollecting , what I have been taught about Harmonics, Load cannot be a source of any power,
It is the switching mechanism that makes the source supply harmonics. am I correct, so if I rephrase it, The utility sources harmonics and fundamental because of non-linear loads. Anyways, In regard to source impedance, This square D manual talks about having high source impedance as one of the means of reducing harmonics and you said
"Put in other terms, the lower you can keep your source impedance, the fewer your potential power quality problems"

Waross also mentioned that two transformers have more impedance one and that helps.
Thanks for your earlier comments though.
Regards,

 

[waross]

Hello davidbeach
Thank you for your time and explanations, I am sure there are many forum members in addition to Engcan and myself who are following your advice. I suspect that there are a lot of us who are a little nervous and lacking confidence when faced with a harmonic issue. I find that understanding a concept in words often makes subsequent mathematical understanding and analysis easier. I have tried to avoid mathematics and to frame my questions and suggestions in simple language to this end.
Please continue assisting us, as your time permits, davidbeach.
We have statements from two good sources that appear to be in contradiction.
Are we confusing HARMONIC CURRENTS with the HARMONIC VOLTAGES that result from them in a less than infinite system?

Square D states;
"Source Impedance
Source impedance has the effect of attenuating the crest factor created by a
non-linear load. Once the voltage rises to a specific point, the control
circuitry in the power supply allows a capacitor to be charged. With low
source impedance, the current drawn by the capacitor is high and the
duration of the charging cycle is short. Higher impedance does not allow as
much current to be drawn, and extends the time it takes to charge the
capacitor. This is how the crest factor is reduced, as well as the harmonics.
An example of how this has been done for years is the use of line reactors
or drive isolation transformers that feed drives, and, more significantly, this
is also done every time an isolation transformer is used!"

I understand that this applies to the CURRENT HARMONICS and I understand that these harmonics create VOLTAGE HARMONICS which are the SOURCE OF PROBLEMS FOR OTHER CUSTOMERS.
To mitigate the VOLTAGE HARMONICS the source impedance upstream of the common connection between customers should be as low as posible.
respectfully
Let me try to put this concept in simple phrases;

A relatively high impedance transformer at the last point of feed to a nonlinear (harmonic inducing) load will somewhat mitigate the MAGNITUDE of the CURRENT harmonics that are generated.
Upstream from that, a relatively low impedance will mitigate the magnitude of the resulting VOLTAGE HARMONICS.
The VOLTAGE HARMONICS which are generated in the system impedances (IZ or IR voltages) are the SOURCE OF PROBLEMS FOR OTHER CUSTOMERS on the system.
respectfully

 

[davidbeach]

OK, I see where they are going. Yes, an impedance, the more inductive the better, on the front end of the rectifier will lengthen the duration of the current pulse, diminishing the peak of the current pulse. That diminished current peak will make for for less voltage distortion due to voltage drops across other impedances further upstream. My point about stiffer power systems, which would be lower impedance, is that if there is minimal impedance, there is minimal voltage drop. One is an attempt to reduce the production of harmonics, the other is a means of making the harmonics a moot issue.

In a sufficiently stiff system, the current distortion would not become voltage distortion. What your quote from Square D is doing is moving enough impedance into the device, at which point distortion is not a power system concern, to flatten out the current pulse. If you rely on source impedances shared by multiple loads, you then have issues with each load impacting the other loads. Low impedance for the common portions of the system and a high inductive impedance on the front end of the individual load. Shunt capacitance in addition to the series inductance can also help by supplying energy into the rectifier when the current it being limited by the inductor.

I think waross's summary is reasonably accurate.

The transformers that Engcan was asking about are intended, working in pairs, to cause the 5th and 7th harmonics through one transformer to be 180 degrees out of phase, on the primary, with the 5th and 7th harmonics through the other transformer to its primary. In the common supply system those harmonics then cancel to a large extent. The third harmonics were trapped in the delta primaries, so the primary circuit feeding the two transformers will have a much lower harmonic content than the primary feeding either transformer alone. The primary feeding a single one of these transformers will have all of 5th and 7th harmonics, but none of the third harmonics present on the secondary side.