Multiple VFD's fed from Bus Duct Systems load calcs

[mull982]

We are looking at a preliminary design where a customer will have several VFD's ranging from 75Hp all the way up to 450Hp at 480V fed from Bus Duct in their facility. The Bus Ducts will likely be fed from Switchgear breakers and the bus duct will be used to allow the customer the flexibility off easily adding drives to the process as necessary.

This particular customer rents space in their facility and provides the availability for their customers to connect drives to their system as needed for the particular application area the space is rented for. So with that said there is not an exact count of the drives that will be installed on the Bus Duct and thus the customers will add the drives as their customers rent space in the facility. The customer has provided us with a rough count of drives and drive sizes for another part of their facility which is similar. Most of these drives will be operating at high frequencies (450HZ or so...so likely not large motor loads) but some of these drives are in the 400 and 450HP range.

I was curious how others would go about the load calculations for this application since the size and quantity of the drives are not fixed but rather will be added as necessary. Would you simply provide Bus Duct feeders from switchgear with each feeder sized for a maximum HP total to allow the drives to be added as needed up to the HP Total? Or would you use the similar existing drive counts/sizes as the basis for sizing feeders with each feeder being specified for a particular drive size & count? Obviously the large 400 and 450HP drives can have a large impact on some of the feeder sizing so I was just curious what approaches others would taking to laying out this distribution.

Curious to hear other considerations for this application.

Thanks

 

[petersonra]

Make the best guess you can and move on. There is no code section on predicting the future.

If your best guess suggests you might be a little tight on bus capacity, maybe you up the bus size, just in case.

You might want to consider how you would go about adding additional bus work down the road in case it is needed, and accommodate that addition.

 

[mull982]

Make the best guess you can and move on. There is no code section on predicting the future.

If your best guess suggests you might be a little tight on bus capacity, maybe you up the bus size, just in case.

You might want to consider how you would go about adding additional bus work down the road in case it is needed, and accommodate that addition.

When calculating feeder and bus capacity for proposed VFD's do you use the VFD input rating adding them up and using 25% of largest as you typically would for sizing feeders?. If VFD input ratings are not available then do you just use FLA of drive HP rating?

 

[mull982]

What kind of diversity factors would you use in a case like this where the connected load is unknown? The similar list of VFD Hp's that we were provided total about 6,000hp which if you add these all up comes to a transformer size of about 6MVA which feels like overkill to me.

Any suggestions on how to include a diversity factor for transformer if we based design of an estimated connected HP size?

Because these drives operate at high frequency does this effect the load diversity calculated for given drive's HP.

 

[Jraef]

What kind of diversity factors would you use in a case like this where the connected load is unknown? The similar list of VFD Hp's that we were provided total about 6,000hp which if you add these all up comes to a transformer size of about 6MVA which feels like overkill to me.

Any suggestions on how to include a diversity factor for transformer if we based design of an estimated connected HP size?

Because these drives operate at high frequency does this effect the load diversity calculated for given drive's HP.

I don't think you can apply a diversity factor to this. Diversity factoring has to do with a known maximum demand of an entire system vs actual demands of individual loads within it. In your case, because the system design is not fixed (meaning people can ADD loads that were not there before), it's not possible to determine an actual demand in my opinion.

You do have a fixed maximum demand though, limited by your service transformer. Then in each individual feed to a bus duct, you have the maximum limit of the bus duct itself. But with a randomly variable load profile that you have no control of, it can be dangerous. At the very least, each bus duct must be limited to the expected load on it, and that should be worst case in my opinion, because it sounds as though you will not have direct control over load profiles. So if you have a 1000A bus duct, you cannot allow more than 1000A of VFDs to be connected to it.

Also one big issue that you left out of this will be harmonics. Lots of individual drives that have no harmonic mitigation are going to rob your service transformer of capacity if nothing else, so that can be a bigger limiting factor than you may realize. You may need to insist that each VFD load be harmonically mitigated, or have each tennant be responsible for owning and maintaining an Active Harmonic Filter to keep your demand on the service under control.

Were it me, I would make it impossible for anyone to plug into the bus duct (by perhaps using locking stab covers?) without coming to someone in charge of looking at ALL of the issues on each individual feeder AS WELL AS the effect it has on the system as a whole.

Output frequency of the drives to the motors is not a factor in any of this by the way.

 

[mull982]

I don't think you can apply a diversity factor to this. Diversity factoring has to do with a known maximum demand of an entire system vs actual demands of individual loads within it. In your case, because the system design is not fixed (meaning people can ADD loads that were not there before), it's not possible to determine an actual demand in my opinion.

You are correct, the system is not fixed. I was simply going to apply a 70% demand factor which is a number I have used on past projects to come up with the demand from the connected load.

You do have a fixed maximum demand though, limited by your service transformer. Then in each individual feed to a bus duct, you have the maximum limit of the bus duct itself. But with a randomly variable load profile that you have no control of, it can be dangerous. At the very least, each bus duct must be limited to the expected load on it, and that should be worst case in my opinion, because it sounds as though you will not have direct control over load profiles. So if you have a 1000A bus duct, you cannot allow more than 1000A of VFDs to be connected to it.

This is a new service which will be supplied from and existing 138kV feeder, so we are dictating the size of the 480V service transformer. We can make it any size as needed for the demand (within reason)

Also one big issue that you left out of this will be harmonics. Lots of individual drives that have no harmonic mitigation are going to rob your service transformer of capacity if nothing else, so that can be a bigger limiting factor than you may realize. You may need to insist that each VFD load be harmonically mitigated, or have each tennant be responsible for owning and maintaining an Active Harmonic Filter to keep your demand on the service under control.

Great point and definitely something that will be considered.