momentary loss of power may drop motor starter, restart immediately or delay restart?

[UASteve]

I am designing a PLC control panel for a train tunnel ventilation system. I am mandated to provide my panel with power from 2 small UPSs that will be powered by presumably separately derived systems. I have a design that can keep the UPS power separated for most components but I will need to 'combine' the UPS power at some point to run some of my IO. When I say 'combine' I really mean switch by contactor or automatic transfer switch. My question is should I go for a very quick transfer time or a slow transfer to let the large induction motors I am controlling either come to a stop or dissipate energy enough to attempt a restart. I found an off the shelf ATS that will transfer in max 30mS even if the line power is out of phase or I can 'roll my own' slow transfer switch that drops power long enough for my PLC to see it is dropped and then lockout my motor starters for 10 seconds or longer....sorry for the long winded post.

 

[waross]

I would delay a restart of any motor of 10 HP or larger.
Some may delay the restart of smaller motors.
30 ms is two cycles.
Two cycles is probably the worst case of an out of phase transfer that may reasonably be expected in the field.
PFC capacitors will make the negative effects of an out of phase transfer even worse.
As well as mechanical transients that may break shafts, the electrical current transients may exceed the ratings of the supply devices and may exceed the normal starting current surge of the motor.
Short version: GO WITH A DELAY. Allow extra time if there are PFC capacitors in the motor circuits.
Example: In soft wood lumber kilns the fans were typically reversed every 2 to 4 hours.
To avoid plugging a still turning motor, a delay of one or more seconds was common to ensure that the motor was completely stopped before being reversed.
I would accept 10 seconds for motors without PF correction.
I would consider more time if there are PF correction capacitors involved.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

Good logic there.

That mass of air will likely not even stop in 10 seconds.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[davidbeach]

UPS for the IO, not UPS for the fan is how I read it.

Any Protection Engineer knows how you solve that one. Two PLCs, each with its own IO, each on its own battery (no need to go back to AC from the DC in the UPS). Either can do everything; all you need is one to be working to have the whole thing working just fine.

 

[itsmoked]

Or at least "fine enough".

Keith Cress
kcress -

http://www.flaminsystems.com

 

[davidbeach]

Ok, of course, fine enough. Perfect is the enemy of good enough.

Two types of employees in the utility business. Those who escaped from consulting and those who escaped the whole consulting thing although. The latter may not be as scarred as the former, but the former have a much better idea of what it means to have a good enough solution. Learned to stop when 80% of the project had consumed 80% of the budget to avoid having the remaining 20% of the project consume the other 80% of the budget.

If you’ve been in consulting you’re laughing, if you’ve never been in consulting you’re trying to make sense of that equation.

 

[jraef]

It’s not so much the torque on the turning load that is the issue, AC induction motor starting torque is generally lower than the peak torque (Break Down Torque) anyway, so drive shafts and such should be designed to handle that. The bigger problem, and the reason for the delay on restarting, is because of the time it takes for the flux in the motor to decay so that the motor is not regenerating at the instant the new source is connected. Reconnecting before the magnetic fields dissipate can be like connecting two generators that are out of synch; there can be a HUGE voltage and/or torque spike, way beyond the capability of anything to handle either one.

I learned this the hard way once in observing a 500HP 460V fan motor twist the shaft off on a transfer switch test scenario. I had a delay timer in the starter set for 3 seconds, but the user didn’t like the delay when added to the time it took the generator to come up to speed and voltage, so he turned it down all the way (0.1sec) against my recommendations. I was young and green, so I had the delay there because I had been taught to use one, but couldn’t articulate the reasons off the top of my head. When they tested it, the generator was already running by the time they commanded the transfer, so there was almost no delay and the shaft snapped like a pretzel stick.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden