Primary/Standby VFDs serving one motor

I have a client with a requirement to keep a motor running in the event of a VFD failure. They do not have redundant motors, but will have redundant VFDs serving the same motor. This is the first time I've seen this and I am left wondering if there needs to be true isolation relays on the field (motor) sides of the VFDs isolating them from one another. Or, is it a non issue to have 1 VFD driving the motor while the other is commanded off with their motor terminals tied together electrically.

russrmartin said:

I have a client with a requirement to keep a motor running in the event of a VFD failure. They do not have redundant motors, but will have redundant VFDs serving the same motor. This is the first time I've seen this and I am left wondering if there needs to be true isolation relays on the field (motor) sides of the VFDs isolating them from one another. Or, is it a non issue to have 1 VFD driving the motor while the other is commanded off with their motor terminals tied together electrically.

Click to expand...

I think you would need an isolation contactor on the output of each drive.

russrmartin said:

I have a client with a requirement to keep a motor running in the event of a VFD failure.

Click to expand...

if the motor still running at 1200 rpm for example then you switch on the redundant VFD what will happen?

The intent is for both VFDs to follow the same command signal, but only one is enabled at a time. Seems like the VFD manufacturer is supposed to provide isolation hardware but I have not seen their submittal to confirm. I was just curious what the consequences are if one VFD is enabled and the other is not while connected together. I can visualize the potential issues if they both are trying to control, but when one is disabled I was unsure of what likely would happen.

russrmartin said:

I have a client with a requirement to keep a motor running in the event of a VFD failure. They do not have redundant motors, but will have redundant VFDs serving the same motor. This is the first time I've seen this and I am left wondering if there needs to be true isolation relays on the field (motor) sides of the VFDs isolating them from one another. Or, is it a non issue to have 1 VFD driving the motor while the other is commanded off with their motor terminals tied together electrically.

Click to expand...

Some VFDs are designed to accomplish this, it is referred to as "N-1 redundancy". In effect, the VFD power section is split in two and run in parallel, each power section sharing equally and feeding the same motor. If one power section goes down, there are two choices of functionality:

1. If the total drive power is matched to the motor size, the remaining power section is only capable of 1/2 of the motor power, so the drive automatically reduces the frequency to limit the motor current to what the drive can handle. This is often used in semi-critical pumping situations where SOME flow is better than NO flow.
2. If the drive is designed as 2X the motor size, then either power section can power the motor fully, providing 100% redundancy without compromise. So normally both power sections are on and sharing the load, but if one goes down, the controller just disables the other and all of the power shifts over to the remaining drive seamlessly.

The key to this working is that there is ONLY ONE control board, so it is controlling BOTH power sections simultaneously and there is zero possibility of the outputs not matching. All it does when it detects a failure of one side is to stop giving it a firing signal.


In order to have two COMPLETELY separate VFDs running the same motor, the ONLY way to not have the drive transistors blow up is to have total isolation between them. If not, the psuedo-AC output from drive 1, if even SLIGHTLY different from that of drive 2, will cause the transistors in drive 2 to self commutate and vice versa, which will destroy them all. So you MUST have two isolation contactors, one on the output of each VFD, mechanically and electrically interlocked so that only one drive can be connected to the motor at the same time while not allowing the two outputs to "touch". The drawback to this is that because of the need for isolation between the drive outputs, there WILL be a time lag in switching from one to the other. That must be weighed against the reasons for requiring the redundancy.


Flying Start functionality would also be a requirement so that the backup drive can catch the spinning load, but most good quality drives have that now (under some name or another).


What size VFD / motor are we talking about here by the way?

Thanks for the detailed answer. Good to satisfy the curiousity without having to do some potential destructive tests. Lol.

The drives range in size and are serving fanwall arrays in air handling units serving various critical spaces. Total HP load is < 100HP in all cases.

russrmartin said:

Thanks for the detailed answer. Good to satisfy the curiousity without having to do some potential destructive tests. Lol.

The drives range in size and are serving fanwall arrays in air handling units serving various critical spaces. Total HP load is < 100HP in all cases.

Click to expand...

OK, too small for any N-1 drives I am aware of then, so you are going to have to go with the isolation version.

Sometimes, the system jraef mentions, or one similar to it, is described as a drive with VFD bypass. It's essentially a bypassed drive as is common in HVAC but, instead of an across-the-line starter on the bypass side, there is another VFD. And, yes, full output side isolation between the drives is essential.

I've seen bypassed drives with softstarters in place of the starter, as well. This is more common as the hp gets above 200hp.

I just completed several systems like DickDV described. I used isolation contactors on the line and load side and separate over current protection for each drive. Flying start is a must if you want automatic switching. Line side isolation is a must in the event of a fault on the primary drive. http://literature.rockwellautomation.com/idc/groups/literature/documents/at/150-at002_-en-p.pdf. Page 65

Line isolation is technically unnecessary, but it DOES allow you to remove the defective drive for servicing without shutting down the back-up system, so it's a very good idea for systems that cannot shut down.

russrmartin,

I would use isolation contactors with a manual switch on the contactor for safety. turn the contactor off when working on the other vfd. If there is no isolation to the vfd's and one shorts out, how will you change the defective one out without turning off the other vfd?

What is the consequences of back feeding power to the drive?

You MUST be 100% certain that both drives are phased the same input wise and produce the same output phasing, otherwise risk possible damage to the output of the backup drive.

regards,
james