Transformer Core de-magnetization

[sjvanhecke]

I have a GSU trasformer that I have someone testing and they did the testing out of order and did a winding resistance test first. Now the power factor readings a all messed up because the core is now magnetized. I know to de-magnetize it you need to use either DC current or voltage. We have a 400A DC source available but you cannot get much voltage across the leads of this piece of equipment. Would a DC welder be better and just bump it to give it a "DC shot"?

 

[Skogsgurra]

sjvanhecke,

I do not think that the resistance measurement caused any problem at all. Your transformer core most likely had a DC field in one or more of the legs when the test started. The reason is that you cannot disconnect from the mains with zero flux in all three legs. Usually you will have more or less DC remanent field in all the legs.

If you really need to de-flux the core's all legs you will need to apply a three-phase voltage that is high enough to saturate the core and then reduce the voltage slowly down to zero volts. Most GSUs that I know about are in the 10+ MVA range so it will not be very easy to do this. And I do not think that you should do so either.

You should find your problems elsewhere.

 

[ScottyUK]

To add some weight to what [blue]skogsgurra[/blue] said, in order to achieve core saturation you will almost certainly need to use the generator attached to the transformer as the voltage source, and then directly control the field current in order to reduce it to zero. This is a pretty unusual thing to want to do, not to mention complicated and potentially dangerous.

What are the symptoms of the power factor readings being messed up? Do all your instruments agree with each other? Cross-check MW, MVAr, PF against each other. Cross-check MW against V, I, and PF. until you find something that doesn't tally.

If the instruments all correlate each other, please post some more information about how the problem manifests itself.



-----------------------------------

Start each new day with a smile.

Get it over with.

 

[jbartos]

Suggestion: The transformer can also essentially have demagnetized the core by cycling it around its hysteresis loop several times using gradually diminishing values of d.c.

 

[electricpete]

It is well known that dc tests should be done last to avoid messing up the excitation test portion of the Doble test. If you have abnormal excitation readings I'd say you have correctly diagnosed the cause (the dc test). If abnormal power factor results, I'm not sure.

I think the procedure for demagnetizing would involve applying your current source in alternative polarity and decreasing magnitude. Exactly what jbartos said. I have heard of it done with a car battery so I suspect your current source will be fine if you have the ability to vary (decrease) the current.

The scenario of residual magnetism associated with sudden deenergization, I believe this is more commonly associated with ct's. (I'm not positive).

I believe in rare/extreme situations it is possible for residual magnetism to lead to tripping of the transformer upon energization.

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

 

[buzzp]

Doble says to use a car battery but some of the guys I work with tried it ONCE. It is very dangerous and you should take some serious precautions if you use the method Doble suggests. We vote to never use this method.

 

[sjvanhecke]

Thanks everyone for you input. What ended up working was taking a piece of equipment called a Omicron CPC100 and putting 200A DC on a winding got rid of the magnetization. Using that piece of equipmnet is mach safer than a car bettery. Thanks again for your help.

 

[jbartos]

Comment: By using the DC power supply, there is good control over the transformer demagnetization and good following the magnetic hysteresis, B-H curve.

 

[electricpete]

What is the hazard with using a car battery?

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

 

[electricpete]

Is it the fact that battery does not have surge supression circuit for breaking the inductive circuit?

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

 

[electricpete]

... and if something goes wrong we could have battery acid all over.

OK. I think I get it.

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

 

[buzzp]

You got it Pete, the arc was such that no one in their right mind (as described to me, I was not there) would ever consider using a car battery. I can ask one of them exactly what happened but I remember someone getting knocked around a bit from a nice big flash/arc.

I could be wrong and maybe this is not a consideration, but seems such a rapid discharge of a lead acid battery could result in bad things happening. I remember hearing this somewhere along the way. However, batteries do get warm when asked to produce currents for starting a vehicle (typical battery is ~700-800A max). Not sure the current draw when attempting to demagnetized a transformer.

 

[busbar]

buzz, the hazard you describe sounds diametrically opposed to that in this electrical “accident”… first entry at

http://electrical-contractor.net/ubb/Forum1/HTML/004077.html

 

[buzzp]

After waiting five minutes for that link to load, I got to read the story. The guy was an idiot. I feel for his family but who jump starts a car with AC power? Beyond me.