Motor Start effect on XFMR 1

[amy86]

We have a system that's running close to the feeder XFMR's Full-load MVA rating:
LOAD-FLOW MVA: 16.0543
XFMR Full-LD MVA: 16.625
When a large motor starts, the starting current increases the load to about 21.2914 MVA which is about 28% more than the rated value of the XFMR, but of course that only lasts for few cycles. My question is, does that hurt the XFMR in any way?

 

[ScottyUK]

Probably not to any significant degree because the thermal time constant of the transformer is very long so it will not see much change. Of course if you now tell us that the motor is 10MVA with a 30 second acceleration time then maybe you do have a problem.

You might want to think about getting more cooling onto the transformer if it is critical to operation: either add fan cooling, or bigger fans, or bigger radiator banks. What are the winding and oil temperatures reading?


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

If we learn from our mistakes I'm getting a great education!

 

[Millbo]

The inrush of start-up current could become an issue over time. What can be occuring is the electro-mechanical forces are trying to push the active part apart. Results can include: loss of clamping force on the windings, deformation in the windings, over-stressed connections and eventual gassing to name a few.

 

[dpc]

I agree with Scotty - I would not be too concerned about this short-term overloading, assuming this motor starting does not happen too frequently and the usual other disclaimers.



David Castor

http://www.cvoes.com

 

[amy86]

Scotty, the transformer's MVA value of 16.625 takes into account added cooling fan system. I'm not sure about the temperatures but I don't believe we're having issues there. I appreciate everyone's comment.

 

[racobb]

You can also look at the ANSI loading curve for the transformer. Don't expect you have a problem.

Alan

 

[jghrist]

I agree with others that you will not have a transformer loading problem. You could have a voltage drop problem, however, during motor starting.

 

[veritas]

Hi

Another issue to consider is that the trfr's overcurrent protection remains stable during the motor startup period. But I agree with the other comments that I would not be too concerned about trfr overloading if conditions are pretty standard.

Also, if designed to a reputable standard then the trfr mechanical fixtures should not be under threat as the motor startup currents are way below fault current values which is where the mechanical fixtures really get tested.

But of course it does no harm to consider factors such as what is the ambient temperature of the trfr? How many motor starts per day are you looking at? Do you have DOL or soft starting? Also, I doubt that the starting currents last only for a few cycles - rather seconds. Are there other motors of similar size supplied by the same trfr that could push up the number of starts per day?

Hope this helps.

 

[RRaghunath]

I agree that the temporary overloading is not a concern. Transformers are supposed to withstand overloading up to 159% of rating for short duration (30minutes if my memory is right) as per IEC.
The bigger concern is the electromechanical forces damaging the winding fixures if the motor happens to be large enough to cause the starting current to reach over say 3times the trafo rated current. I have seen this happening in no. of transformers.

 

[amy86]

Thanks guys...I don't believe the motor is that large. It's 1200 HP motor at 4160v. A soft start would be a good idea I should consider...but not sure that it's worth the investment.

 

[jack6238]

Hey fellows,

Here is the scoop.

You say the transformer is loaded to 16.625 MVA. Here's the question. Is that the 24 hour RMS average?? If not, then that is not the loading!! Calculate the 24 hours RMS value and I'd bet my soul that it is less that 16.625.

And who cares about starting a motor on a transformer and the effect on a transformer life . It's nothing!!!! A transformer does not lose life when a motor is started. A transformer has a statisticl life of 400+ years(See IEEE Gold Book) and it is the RMS loading and primary voltage of the of a transformer that determines the life of a unit. Now, if this is a transformer with an impulse load, hammermill or chipper , then this is a another consideration. This can be calculated.

I do not care if the accel time is 30 seconds or 60 seconds, the life of the transformer is never the cnsideration................it's the effect on the voltage drop at the bus that is the concern. My heavens the transformer is a stout item in the sysem.

Who in the world scares the uneducated with this nonsence?

Jack

 

[ScottyUK]

Hello Jack,

Nice to see some new contributors here.

Have you got any reference for statistical life of 400 years on a transformer, or is that a typo for 40 years? I'd expect paper insulation to be weakened after 40 years, depending on how hot it has run, and at 60 years the insulation will be fragile and may not survive a major through-fault.

Why do you think a hammer mill or chipper is bad for a transformer - because of the cyclic forces caused by the varying current?


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

If we learn from our mistakes I'm getting a great education!

 

[jack6238]

Mr. ScottyUK,
Thanks for the question. And it is nice to know someone reads and pays attention. I myself can trace my linage to England in the 1600's. So here's a special to you.

I did not make a mistake on the statiscal life and yes it is ~ 400 years. IEEE has made an analysis of the reliability of electrical systems The life of the transformer was a part of this study. By the way, Charlie Heising of G.E led, (in my opinion) this effort.

The 400 year life if based on the primary voltage of 15 kV and the reported failure rate of transformers as summarized inthe ANSI/IEEE 493 Gold Book. For example, your plant has 400 transformers, you will have a transformer failure each year. Our local utility (Duke Energy) confirms this data but they say ~300-350- years for the statistical life.

Concerning the hammermill or chipper and the impact on a transformer.

Louie Levoy( actually L.G. Levoy)of GE Schenectady wrote a classic paper on the "Impact Loading of Transformers." The paper concerned the application curves for transformers with pulsating or short time loading. These loads are typical of hammermills and Chippers. I am not saying these loads are bad for transformers. What I am saying is that these loads must be considered when applying transformers...............for they should last well beyond the process life.

Jack

 

[alehman]

Jack,
I echo Scotty's appreciation for your participation.

I would like to clarify that expected failure rate of 1 per 300 to 400 transformer-years is clearly not the same thing as an expected life of 300 to 400 years for any given transformer. The failure rate calculations are based on transformers in service, which are normally replaced before they fail.

Alan
“The engineer's first problem in any design situation is to discover what the problem really is.” Unk.

 

[electricpete]

Let me provide a 2nd echo for Alan and Scotty's comments - welcome.

I'll throw in my two cents on this peripheral issue of transformer life. I suspect the 400 year number is probably an expression of failure rate (one failure per 400 transformer years). But it is probably not supposed to be interpretted as some kind of average life expectancy for an individual transformer.... logic tells us that aging effects would become important before 400 years as Scotty observed.

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

 

[electricpete]

And as I reread it you did describe it as a failure rate. So, I don't think I am correcting or contradicting anything, just trying to provide some discussion to understand the 400 year number.

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