Transformer Inrush Measurement

[ThePunisher]

We are setting up to measure a transformer's maximum inrush current. One of our testing contractor offered an equipment called: Dranetz PX5 recorder with 3 TR2500 clamps, memory card and software.

My question is a 5ms measurement enough to detect the transformer inrush with harmonics or do we need less than 4ms measuring capability?

 

[Skogsgurra]

The Dranetz is not exactly what I would use for that. Also, if you are concerned about the DC component of the inrush current (you shoud be) you shall make sure that the clamps can handle peak current AND the rather long (hundreds of milliseconds - sometimes seconds) DC component without saturating.

There was a recent similar discussion at

http://www.eng-tips.com/viewthread.cfm?qid=271076&page=3

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ThePunisher]

Thanks Gunnar, any recommended instrument?

How about the HIOKI recorder?

 

[ThePunisher]

Here is a specs for the HIOKI:

http://www.hioki.com/product/pdf/88...D=0&bcsi_scan_filename=8860-50E4-02E-00KR.pdf

The difficulty weare having is that there are testing contractors who offers different instruments in measuring the inrush. I would like to get an idea, what should be consider in order to have confidence that their instrument will really capture the peak inrush.

My thanks for any support.

 

[dpc]

Another consideration in actually trying to measure the inrush is that it will be different each time the transformer is energized, unless you have the means to close all three phases at a precise phase angle.

Shape and magnitude of the inrush event is a function of voltage phase angle and remnant flux in the core, as well as any secondary load. Doing a single measurement is not really going to be definitive.

David Castor

http://www.cvoes.com

 

[ThePunisher]

We are actually going to do the test at several times to ensure that we get enough remanence on the transformer.

Both current and waveform captures will be taken on each phase and the switching will be three phase. With this we may compare and see whether the peak current happened at zero voltage period.

My concern is whether the instrument offered will be capable of measuring the harmonics and DC component associated with the transformer inrush and whether the sensors are properly sized so as not to saturate.

Does anyone had any experience on doing this and any suggested instruments will be appreciated.

Do we need NETA Level 3 Tech to do the job?

 

[Skogsgurra]

The HIOKI is great. I have had several over the years. One of the best producers of high-quality recorders. You can't go wrong there. But it costs.

There are several USB instruments available. I have a few TiePie four channel instruments that can be used in parallel so that you can build 4, 8, 12 channel recorders with them. Sampling and resolution is similar to the HIOKI.

We also use our own ARCUS. It has ten channels and 12 bit resolution. Not as versatile and fast as the HIOKI and TiePie. But extremely simple to use. And adequate for inrush measurements.

Dranetz is more for power quality. Inrush is somewhat different. And if you want to do other things, the HIOKI is your safest bet.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ThePunisher]

Here is an information of what is being offered to us:

http://www.dranetz-bmi.com/products/prodspec.cfm?prod=47

The instrument has an inrush mode and a transient mode. I was thinking on considering a transient mode measurement, which goes down to 5 microsecond measurement. The Tech informed us that we don't need that sensitivity since we are just measuring the peak inrush. I was thinking that a transformer inrush will have harmonic and DC component currents included so a transient wave form capture will accurately indicate the current spike. I am not that familiar with recording instruments so I appreicate an overview on these recorders to guide me as far as transformer inrush measurement is concerned.

Anybody who used Dranetz before for inrush measurement of a transformer?

 

[Bronzeado]

Dear ThePunisher,

Any recoder with a good sample rate can make the measurements. The real problem is the clamp you will use to measure the inrush current. It should have a good frequency response and can not saturate.
So, you need estimate the peak value of the inrush current and use a clamp that not saturate for that peak.

Best regards,

Herivelto bronzeado

 

[electricpete]

...since we are just measuring the peak inrush.

Out of curiosity, what is the purpose of the measurement?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[ThePunisher]

WE have these transformers connected to an upstream UPS which bypasses everytime the transformer is energized. WE had proposed before to connect it to a normal mains with redundancy.

However, one manufacturer claimed that they can manufacture a special design that will yield an inrush of almost 100% FLA. Now we are testing the protoype to ensure that what was spec out by them is valid.

 

[electricpete]

I think you have gotten good advice for test equipment. I have heard people try to use various “peak capture” features that simply give you a single number representing the peak. I don’t like that approach because I don’t know what it represents (instantaneous peak, if so what sample rate, rms peak...if so averaged over what interval, etc). I like to look at a TWF because it gives you a lot better idea what the number represents. Also you can get a feel for the sample rate looking at TWF, and factors such as CT saturation may be evident in current TWF.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Skogsgurra]

TWF? True Wave Form?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ThePunisher]

Actually, I looked at the both HIOKI and Dranetz PX5. The HIOKI recorder is having 20 Mega samples/sec. while the Dranetz is 256 samples/cycle = 15,360 samples/sec.

As I gather, the transformer inrush occurs on a very very short period of time so a high speed recorder like HIOKI would be a good choice. I will check on the clamp rating to ensure that it will not saturate while measuring the inrush. I think the clamps are rated as to the maximum current they are rated to measure.

Does this mean that if my estimated inrush is 720A RMS, I would use a clamp sensor having an accuracy up to 800A?

Regards,

 

[jghrist]

Theoretically, you need twice as many samples per cycle as the highest harmonic you need to record. You don't really need any more that the 7th harmonic, so the 256 samples per cycle of the Dranetz would be more than adequate.

I've gotten very reasonable recordings of inrush using the oscillography of relays with 16 samples per cycle.

 

[Skogsgurra]

Re 800 A.

I would use at least a 1000 A clamp. And make sure it doesn't saturate just above 1000. Most 1000 A clamps are good for 2000 A peak, with a somewhat reduced accuracy.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ThePunisher]

If I get personal technical opinion, can the dranetz PX5 suitable for measuring transformer inrush currents?

This is what is offered:

http://www.dranetz-bmi.com/products/prodspec.cfm?prod=47

 

[rbulsara]

WE have these transformers connected to an upstream UPS which bypasses every time the transformer is energized.

That is a well known issue with UPS and downstream transformers. You are lucky not have blown fuses in the UPS bypass circuit or cause some other damage. It is customary to place the UPS on bypass before energizing the transformers. In data centers, there are means and procedures to prevent automatic re-engergization of PDU transformers, once the power is lost or you can size the UPS accordingly.

But if you can get a unit that limits the inrush within capability of the UPS, that would work too.



Rafiq Bulsara

http://www.srengineersct.com

 

[ThePunisher]

Yes Rafiq, we are luck we did, but we are closely going to monitor people form loading them further before our luck expires and end up tripping breakers and blowing up fuses.

By the way, on the current clamp specifications, it is indicated that:

Basic Accuracy: 1% F.S. (48~65 Hz) (%reading + floor spec).
Useable Frequency: 1Hz~20kHz
Max. Non-destructive Current: 4000A
Continuous Current Range: 2A-200A, 20A~2000A.

I am concerned of why the % accuracy is not mentioned for frequencies outside the 48Hz~65Hz window but within the 1Hz~20kHz window. Any reasons why and how does this affect the measurements.

I asked the testing contractor but they gave me a dead end answer that this is what the manufacturer specified.

 

[rbulsara]

Loading of transformer has very little to do with peak inrush on engergization, that actually sends the UPS on bypass. Some loading may help dampen the transient quicker but won't help breakers tripping on instantaneous or UPS triggering to bypass.

Rafiq Bulsara

http://www.srengineersct.com