Power cycling line side of drives

kolyur · Jun 12, 2012

I've noticed that many VFDs and servo drives I've used have a warning in the manual that the line power should not be cycled too frequently. Usually it specifies some amount of time, say 30 seconds to several minutes, that the device should remain 'off' before applying power again. What's the purpose of this delay? I suspect it has to do with capacitor drain/charge time but how can it be detrimental to the drive?

The reason I ask is that I'm using a simple stepper motor drive in my current project that doesn't have a typical 'enable' signal. (When line power is applied, the motor immediately has holding current.) I'm trying to decide how to implement the emergency stop function. I'd like to put a contactor on the line power but the manual states that power should be off for 15 seconds before re-applying. That will usually be the case with an E-stop, but not always. I'm concerned about whether occasional quick power cycles will significantly reduce the life of the drive.

I've already contacted the manufacturer and they had no good advice other than restating the 15-second rule from the manual. They did suggest using the "All Windings Off" drive input, which releases holding torque, but I've discounted that because it is an active low signal (signal on = motor off) which is not appropriate for a safety function.

brucechase · Jun 12, 2012

Yes, short cycling power is detrimental to drives, in my experience. I have seen this on mainly the older AB drives (which is mostly what I used).

What I've done is put the safety contactor on the output of the drive and at the same time to kill the enable to the drive via the safety contactor.

Now many will say that you need to actually kill the enable BEFORE you open the drive output contactor. There are several ways to do this if you use a Safety PLC or a Safety controller. You have the option to delay the signal by a small time (say .5 seconds) before you open the contactor. I have never actually done the delay. After installing the output contactor, we definately saw a big improvement of the life of the drive.

DamianInRochester · Jun 12, 2012

The biggest reason that this is bad for most capacitor bus based drives is because of the charging circuit. Most smaller sized drives utilized a resistor and contact on start-up to limit the inrush of current when first applying power. Once the bus is up to voltage they often close a contact across the resistor to eliminate the drop from that point on. This resistor is typically weak link because they are generally not sized for any type of rapid duty cycle.

The larger the drives get the more likely they will utilize some type of thyristor controlled charging circuit to bring the cap bank up to voltage. But it is a much higher cost component. From my experience this tends not to suffer as much from cycling power, but you will also have the issue, like you pointed out, that there is another resistor that is usually used to drain the bus voltage on a power down, and this can get overheated as well.

DamianInRochester · Jun 12, 2012

brucechase said:
Yes, short cycling power is detrimental to drives, in my experience. I have seen this on mainly the older AB drives (which is mostly what I used).

What I've done is put the safety contactor on the output of the drive and at the same time to kill the enable to the drive via the safety contactor.

Now many will say that you need to actually kill the enable BEFORE you open the drive output contactor. There are several ways to do this if you use a Safety PLC or a Safety controller. You have the option to delay the signal by a small time (say .5 seconds) before you open the contactor. I have never actually done the delay. After installing the output contactor, we definately saw a big improvement of the life of the drive.

I have used this technique as well with pretty good success. If you have a common DC bus system that also needs to be zoned you almost have to do it this way. I always have the delay in there (albeit tiny ~100ms) so I can't speak to how things go without the delay.

kolyur · Jun 12, 2012

brucechase said:
What I've done is put the safety contactor on the output of the drive and at the same time to kill the enable to the drive via the safety contactor.

Now many will say that you need to actually kill the enable BEFORE you open the drive output contactor. There are several ways to do this if you use a Safety PLC or a Safety controller. You have the option to delay the signal by a small time (say .5 seconds) before you open the contactor. I have never actually done the delay. After installing the output contactor, we definately saw a big improvement of the life of the drive.

Yes we do this on most of our servo drives and VFDs. I use a safety relay with off-delay contacts to drop the enable immediately, then open the load-side contactor about 1/4 second later.

It works great but I was just trying to avoid the extra wiring in this instance. Now I think I'll stick with what works. Thanks for the tips.

The Plc Kid · Jun 12, 2012

brucechase said:
Yes, short cycling power is detrimental to drives, in my experience. I have seen this on mainly the older AB drives (which is mostly what I used).

What I've done is put the safety contactor on the output of the drive and at the same time to kill the enable to the drive via the safety contactor.

Now many will say that you need to actually kill the enable BEFORE you open the drive output contactor. There are several ways to do this if you use a Safety PLC or a Safety controller. You have the option to delay the signal by a small time (say .5 seconds) before you open the contactor. I have never actually done the delay. After installing the output contactor, we definately saw a big improvement of the life of the drive.

Actually opening the load side without first ensuring that the IGBT's have stopped conduting buy disabling the drive is worse than cycling line power but both are bad. The best method is to use safe off if you have a drive that supports it. Most newer rockwell drives do and some other vendors do also.

Using a contactor with a pneumatic delay is the way it used to be done in many cases and monitor the contactor state in your safey system. Safety contactors are good for systmes with a high safety rating.

DamianInRochester · Jun 12, 2012

The Plc Kid said:
Actually opening the load side without first ensuring that the IGBT's have stopped conduting buy disabling the drive is worse than cycling line power but both are bad. The best method is to use safe off if you have a drive that supports it. Most newer rockwell drives do and some other vendors do also.

Using a contactor with a pneumatic delay is the way it used to be done in many cases and monitor the contactor state in your safey system. Safety contactors are good for systmes with a high safety rating.

Dropping the enable to the drive should shut off the IGBT almost instantly. So it is more than likely that the transistors are off well before the contactor contacts physically separate. I agree though, I wouldn't want to count on that timing. Giving the contactor a delay is the safe bet.

I often actually end up with two tiers of delay. The first delay is to start decelerating everything to a stop rapidly, then drop the enable, then open the contactors about 100ms later.

From the drives standpoint I don't see a difference as far as protecting the drive between droppping an old fashioned enable signal as opposed to an STO. They both accomplish the same thing. The STO just meets more stringent safety requirements and is supposed to be unprocessed.

DickDV · Jun 12, 2012

Returning to the original posters subject, the capacitor inrush current when powering up is limited thru a resistor.

A resistor can take ten times its rated wattage if starting cold but that wattage comes down rapidly to its continuous rating if there is any chance of it not fully cooling before being hit again. It is for that simple reason that the time delay or the per-hour limit is imposed.

If the manufacturers would use the proper wattage resistor, the limitation wouldn't be there but that would be physically bigger and more expensive.

The Plc Kid · Jun 12, 2012

Well you also have to consider the load. I almost all estop situations you want your motor to come to a aggressive but controlled stop. There are very few cases where you want to break the connection between the drive and motor and the motor is stil rotating under load.

I have a lot of stuff that also uses safe speed monitors to control safe off actuation in the drive so when the estop is hit a very aggressive dc injection brake occurs or a physical brake comes on and in some cases both injection to stop and physical to hold. Then at 0 speed the drive isolation contactor opens or safe off engages.

shooter · Jun 13, 2012

An Estop is not allowed in this place as the driver will stay active too long.
You must use the disbale input, but that is no problem as the driver will keep its power.
You can use a seperate break here,
If you put a contactor behind the driver the driver will not keep its psoition so restart is difficult. and the outputs could burn as the is no load to them.

curlyandshemp · Jun 13, 2012

brucechase said:
Yes, short cycling power is detrimental to drives, in my experience. I have seen this on mainly the older AB drives (which is mostly what I used).

What I've done is put the safety contactor on the output of the drive and at the same time to kill the enable to the drive via the safety contactor.

Now many will say that you need to actually kill the enable BEFORE you open the drive output contactor. There are several ways to do this if you use a Safety PLC or a Safety controller. You have the option to delay the signal by a small time (say .5 seconds) before you open the contactor. I have never actually done the delay. After installing the output contactor, we definately saw a big improvement of the life of the drive.

This is a situation where we use a time delayed safety relay. These safety relays have two types of contacts, immediate and timed.
When the safety circuit trips, the drive enable signal is dropped out immediately through the immediate contacts. The timed safety relay contacts are used to drop out a safety contactor on the load side of the drive.

GlennA · Jun 13, 2012

You say the drive does not have an 'enable' signal. Does it have an option to program an 'interrupt' signal ? Can you use an input to trigger a 'maximum deceleration stop' ? Use the normally closed contacts of a relay in the safety circuit to drive this signal, so that the relay must be energized to reset the stop signal. And you could use a discrete timer to delay the power-up. But it would have to get power from somewhere else so that it could time while the power to the driver is off.

kolyur · Jun 13, 2012

When an E-stop is triggered, the stepper controller will immediately ramp down the motor. At that point it's stopped but still has power applied (holding torque). The question is, what then? I could either break the line side of the drive (main power) or the load side (between drive and motor). If I do the latter I'd have to enable the "All Windings Off" input first to remove current from the motor. It just strikes me as overly complicated: (1) ramp down motor, (2) enable AWO, (3) short delay, (4) open contactor.

GlennA · Jun 13, 2012

kolyur said:
When an E-stop is triggered, the stepper controller will immediately ramp down the motor. At that point it's stopped but still has power applied (holding torque). The question is, what then?
Depends on your load. If the load will fall due to gravity you may want to maintain holding torque. If it is safe, or safer, to have the load free to move, remove the motor power.

OkiePC · Jun 13, 2012

If you break the line side, then when recovering homing will be necessary, correct?
If you take the added steps to properly break the load side, can the drive recover without homing?
That "AWO" signal sounds like the one that would warrant early break/late make aux. contacts on the load side contactor.

I have a preference for breaking the load side of drives and keeping the caps warm and pre-charge resistors cool. I have seen too many drives go poof due to button happy operators. I have at times used a PLC driven relay to prevent a safety circuit from being reset too frequently. This punishes e-stop abuse with downtime which can be a good thing, but it is still not the ideal thing to cause downtime because you are forced to let a pre-charge resistor cool down.

Power cycling line side of drives

kolyur

Lifetime Supporting Member + Moderator

brucechase

Member

DamianInRochester

Lifetime Supporting Member

DamianInRochester

Lifetime Supporting Member

kolyur

Lifetime Supporting Member + Moderator

The Plc Kid

Member

DamianInRochester

Lifetime Supporting Member

DickDV

Member

The Plc Kid

Member

shooter

Member

curlyandshemp

Lifetime Supporting Member

GlennA

Member

kolyur

Lifetime Supporting Member + Moderator

GlennA

Member

OkiePC

Lifetime Supporting Member

Similar Topics