Unusual Electrical Load 1

[richanton]

I have a furnace type tool that is being designed. It's electrical load profile is 480 volts, 3 phase, 200 amps constant load, but when the tool process is in use, the current spikes to 840 amps for 6 seconds out of every minute. This process lasts maybe an hour or two, then the tool would be back to 200 amps for another hour, then process begins again. Is there any way the breaker or the cable could be sized smaller than the 840 amps you would normally size for?

 

[Skogsgurra]

It depends somewhat on where in the World you are.

Breakers and cables usually can take six seconds overload on a reglar basis, provided the duty factor isn't too high. At 10 % duty cycle, it would probably not be necessary to use anything 'thicker' than what you would use for the steady state load. One size up is probably sufficient.

Many motor loads with high inertia have the same characteristics with higher current peaks than you have. But, it depends on what code you have to obey.

Gunnar Englund

http://www.gke.org

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

 

[rcw retired EE]

Will a 300A breaker trip on 840A after 6 seconds? Proably not.

Will a cable rated for 300 amps be protected within code by a 300A breaker? Yes.

Check the breaker trip curves. Also look at the IEEE Brown Book on cable protection where it describes short term overloads.

Since it is a furnace load, you may have to upsize the cable for ambient temperature derating anyway.

 

[cranky108]

The breaker curves are usually coordinated for larger motor inrush. So why can't you coordinate for this higher current for the time limit?
You should also look at the reset time to be sure you don't ratchet into a trip.

 

[richanton]

Thanks for the ideas. What I probably have to do is get the tool designer to profile the cable temperatures to see if they return to normal during the 54 seconds when it is not drawing 840 amps. It would seem to me that if they don't return to normal, then the unit could eventually trip the breaker.

 

[alehman]

The NEC dosen't allow undersizing for a single intermittent load. You would be required to size the feeder and circuit breaker for 840 amps.

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

 

[magoo2]

The cable effects can be evaluated in terms of its load's duty cycle. With 840 A for 6 s every minute, what is the load for the other 54 s? Determine the average over a minute and you'll have a number that is applicable for the cable.

You've already got some comments about the breaker. I'd have to look at an actual breaker curve to see what the trip levels are.

 

[Skogsgurra]

Do not use the average. Use sqrt((I1^2*t1 + I2^2*t2 ...)/total time) i.e. the square root of the mean square aka RMS value.

With 840 A 6 seconds and 200 A for 54 seconds you will have 326 A thermal load for cable and also for the breaker.



Gunnar Englund

http://www.gke.org

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

 

[rcw retired EE]

Gunnar - Good idea, I should have thought of that, because we size cables that way on varying loads. At the end of the minute, the cable can't tell if the heat came from 840A and 200A or from a continuous 326 A.

The ETAP program used to have a cable heating module that predicted operating temperatures from varying loads. When we ran intermittent repetitive loads through it, the cable temperature ratcheted up and stabilized at the same temperature as when we ran the equivalent I2t continuous amps through it.

Still, use a minimum of 125%-140% margin and check for ambient temperature derating. If you use parallel cables, add an extra 5-10% derate for unequal load sharing.