I've installed a 112.5kW TX, 480vDelta/208vWye (reverse wired, approved by UL & manufacturer) at a small plant that only uses the connected load a couple of hours per month. They regularly switch on/off the main 200A CB(no load) feeding the TX to save the cost of keeping the core energized (approx 1kW-hr no-load = $90.00/month). The CB trips everytime at switch-on(in-rush current), and then holds at 2nd try. I've just replaced that CB(wonder why!). Question: Is there a "soft start" device available, or, can I/should I "up-size" the primary CB (and what is allowed by NEC)? The TX is <6ft from the main CB/Meter, and the secondary fused disconnect (80A DEF fuses)is <6ft from the TX. Thanks.
Transformer Inrush Problem
- Thread starter photon
- Start date
- Status
Re: Transformer Inrush Problem
What is the FLA of a 112.5kva transformer at 208v? My figures show 312a for which a 200a breaker is applied.
The data that I have acquired from an old timer and retired dry type distribution transformer design engineer who worked for a leading manufacturer shows that the average inrush current (asymmetrical) for a common 150degC transformer is about 2800a. Knowing that the magnetic trip of a 200a rated breaker is 10x its rating, the breaker is calibrated at 2000a +-20%. You've got trouble right from the getgo. Now, go to a 115degC rise transformer and you can expect the inrush to increase in the neighborhood of up to 20%, that would be over 3000a. Expect 80degC, K4, K13 transformers to be even higher, anywhere from 25-60%.
When designing a lower temperature rise transformer with lower losses and K factor transformers the inrush is higher because the transformer's windings have lower resistance, thus higher inrush current values.
So a 200a breaker paired with a 112.5kva transformer on can expect trouble. If the primary protective device were sized FLAx1.25% the ideal size would be 390a which is not standard so you are then allowed to step up to the next standard rating which is 400a. This would give you a magnetic rating of 4000a
However, going backwards from an inrush of 2800a for a 150degc transformer one could conclude that a 300a breaker may work provided that the magnetic calibration is on the high side.
I have always advised people to stay away from those marginal application because you may end up on the hairy edge and which is an invitation to trouble.
Couple that with a panel that will not accommodate a breaker larger than a 250a frame and you have now backed yourself into a corner.
It is also a big problem when you have based your primary protective device based upon the cable which you have sized based upon the anticipated load without giving the transformer inrush any consideration. Transformer inrush is not going to change.
But, then again, sometimes one gets lucky.
What is the FLA of a 112.5kva transformer at 208v? My figures show 312a for which a 200a breaker is applied.
The data that I have acquired from an old timer and retired dry type distribution transformer design engineer who worked for a leading manufacturer shows that the average inrush current (asymmetrical) for a common 150degC transformer is about 2800a. Knowing that the magnetic trip of a 200a rated breaker is 10x its rating, the breaker is calibrated at 2000a +-20%. You've got trouble right from the getgo. Now, go to a 115degC rise transformer and you can expect the inrush to increase in the neighborhood of up to 20%, that would be over 3000a. Expect 80degC, K4, K13 transformers to be even higher, anywhere from 25-60%.
When designing a lower temperature rise transformer with lower losses and K factor transformers the inrush is higher because the transformer's windings have lower resistance, thus higher inrush current values.
So a 200a breaker paired with a 112.5kva transformer on can expect trouble. If the primary protective device were sized FLAx1.25% the ideal size would be 390a which is not standard so you are then allowed to step up to the next standard rating which is 400a. This would give you a magnetic rating of 4000a
However, going backwards from an inrush of 2800a for a 150degc transformer one could conclude that a 300a breaker may work provided that the magnetic calibration is on the high side.
I have always advised people to stay away from those marginal application because you may end up on the hairy edge and which is an invitation to trouble.
Couple that with a panel that will not accommodate a breaker larger than a 250a frame and you have now backed yourself into a corner.
It is also a big problem when you have based your primary protective device based upon the cable which you have sized based upon the anticipated load without giving the transformer inrush any consideration. Transformer inrush is not going to change.
But, then again, sometimes one gets lucky.
brian john
Senior Member
- Location
- Leesburg, VA
Re: Transformer Inrush Problem
Not sure where this transformer is located (what geographical area) but leaving a dry type transformer off is not always the best idea. In high humidity areas the insulating material can absorb moisture degrading the transformer.
I was informed by a transformer engineer that the inrush for reversed wired transformers (approved or not) have a higher inrush that a transformer specifically wound for 208 to 480/277 operation.
As for soft start yeah you could utilize a 3 phase variable transformer to energize the windings. This would be a questionable installation and costly to install.
Not sure where this transformer is located (what geographical area) but leaving a dry type transformer off is not always the best idea. In high humidity areas the insulating material can absorb moisture degrading the transformer.
I was informed by a transformer engineer that the inrush for reversed wired transformers (approved or not) have a higher inrush that a transformer specifically wound for 208 to 480/277 operation.
As for soft start yeah you could utilize a 3 phase variable transformer to energize the windings. This would be a questionable installation and costly to install.
Re: Transformer Inrush Problem
Yes, the inrushes would be higher as it is proportion as are the pri. and sec. voltages. So it appears high if the pri. sec. ratio is not taken into consideration. A 112.5kva, 480y-208y/120, 150degC transformer would have a pri FLA of 135a and sec. 312a. Anticipate the inrush to be about 1200a for the HV and 2800a on the LV.
If the cable feeding the transformer has been sized to carry the FLA of the transformer and the breaker feeding the transformer is sized accordingly you should never have a problem, either using a transformer to step up or step down with common 150degC rise transformers and you shouldn't with most others.
It's when the transformer is oversized for the load for which money is saved by running smaller cable and using a smaller breaker that starts to be an issue because the transformers inrush does not change.
My study was the result of a request made to me by a transformer marketing manager because no one had ever addressed it before. He had been confronted with nuisance tripping issues which was largely misunderstood. Combining my knowledge in industrial molded case circuit breakers and the anticipated inrushes of dry type distributions transformers of various designs (150, 115, 80degC, K-4, K13, etc). I put together a spread sheet using pri. only, and pri. if sec. protection is provided, with breaker magnetic trip values and calibration range.
Because my data is limited to that provided by one transformer design engineer I would welcome any credible inrush values from any other transformer manufacture to include in my data base.
Yes, the inrushes would be higher as it is proportion as are the pri. and sec. voltages. So it appears high if the pri. sec. ratio is not taken into consideration. A 112.5kva, 480y-208y/120, 150degC transformer would have a pri FLA of 135a and sec. 312a. Anticipate the inrush to be about 1200a for the HV and 2800a on the LV.
If the cable feeding the transformer has been sized to carry the FLA of the transformer and the breaker feeding the transformer is sized accordingly you should never have a problem, either using a transformer to step up or step down with common 150degC rise transformers and you shouldn't with most others.
It's when the transformer is oversized for the load for which money is saved by running smaller cable and using a smaller breaker that starts to be an issue because the transformers inrush does not change.
My study was the result of a request made to me by a transformer marketing manager because no one had ever addressed it before. He had been confronted with nuisance tripping issues which was largely misunderstood. Combining my knowledge in industrial molded case circuit breakers and the anticipated inrushes of dry type distributions transformers of various designs (150, 115, 80degC, K-4, K13, etc). I put together a spread sheet using pri. only, and pri. if sec. protection is provided, with breaker magnetic trip values and calibration range.
Because my data is limited to that provided by one transformer design engineer I would welcome any credible inrush values from any other transformer manufacture to include in my data base.
Re: Transformer Inrush Problem
Yes, 400a is what the correct size would be as stated in my previous post but quite often, if a panel used that can only accommodate up to a 250a frame then watch the finger pointing.
Even better yet if one applies the pri. when sec. protection is provided, that's 312x250%+780a which means that a wooping 700a macho breaker can even be used , ref table 450.3(B). However, the way that I read it you may have to read between the lines a bit because this would relate to the 25' tap rule and as such the OCPD can not be located more than 25' from the sec. OCPD when measure through the transformer.
I doubt if this allowance is used very much though as 125% is usually more than adequate.
Yes, 400a is what the correct size would be as stated in my previous post but quite often, if a panel used that can only accommodate up to a 250a frame then watch the finger pointing.
Even better yet if one applies the pri. when sec. protection is provided, that's 312x250%+780a which means that a wooping 700a macho breaker can even be used , ref table 450.3(B). However, the way that I read it you may have to read between the lines a bit because this would relate to the 25' tap rule and as such the OCPD can not be located more than 25' from the sec. OCPD when measure through the transformer.
I doubt if this allowance is used very much though as 125% is usually more than adequate.
Re: Transformer Inrush Problem
Thanks to all for good advice. I've decided to install a 400A breaker, and a heavy duty switch on the primary side to prevent arc damage to the breaker (I don't think it is switch-rated) for connecting/disconnecting the X-former. BTW, the conductors are all sized for maximum load capacity (I always try to plan for something added later to every system - CYA and I don't like doing the same job twice!). Thanx again.
Thanks to all for good advice. I've decided to install a 400A breaker, and a heavy duty switch on the primary side to prevent arc damage to the breaker (I don't think it is switch-rated) for connecting/disconnecting the X-former. BTW, the conductors are all sized for maximum load capacity (I always try to plan for something added later to every system - CYA and I don't like doing the same job twice!). Thanx again.
Re: Transformer Inrush Problem
Please don't waste your money on a disconnect switch unless it is required by the NEC of for additional convenience.
Because of the way breakers are designed it should handle that transformer load like a breeze..
Most are not familiar with UL489 tesing requirements but here's an overview for a 400a breaker in part: Overload test of 50 operations at 600% normal current at rated voltage followed by 25 operations at 200%, 1000FL and 5000NL operations at 4/min. After endurance tests the breaker must again pass a calibration test at the 200% and 135% ratings.
So don't underestimate the robust qualities of a breaker.
Please don't waste your money on a disconnect switch unless it is required by the NEC of for additional convenience.
Because of the way breakers are designed it should handle that transformer load like a breeze..
Most are not familiar with UL489 tesing requirements but here's an overview for a 400a breaker in part: Overload test of 50 operations at 600% normal current at rated voltage followed by 25 operations at 200%, 1000FL and 5000NL operations at 4/min. After endurance tests the breaker must again pass a calibration test at the 200% and 135% ratings.
So don't underestimate the robust qualities of a breaker.
- Status