Open Delta Buck Boost Connections

[Sparkologist4hire]

Hello

I am working on a project at a local hospital. The utility service is 240/120 3 phase 4 wire (high leg) and the emergency generator is 208Y/120 3 phase 4wire, 208v line to line and 120V line to neutral.

We are adding a new 150 amp 208Y /120v panel board for the IT server UPS and HVAC. The engineer is calling for (2) 5kva buck boost transformers, wired OPEN DELTA to convert the utility service to match the generator service before the new transfer switch.

The only response I got from the engineer from my RFI was the buck boost transformers are wired open delta.

This is the closest wiring diagram I can find and the engineer has given no more information than I have already typed. At best this looks like 120/240v 3 phase 4 wire service and it looks to me like the voltage from "A" phase to Neutral is still 200+ volts.

I am also concerned about the size of the buck boost transformers only being 5 kva and the service load is 150 amps.

 

[winnie]

See http://www.jeffersonelectric.com/s3/site/pdf/Buckboost-wiring.pdf for various wiring diagrams related to three phase buck-boost systems.

I believe that your concerns are correct; an 'open delta' buck-boost system would change the line-line voltage from 240V to 208V, but would not correctly alter the line-neutral voltages.

I can imagine autotransformer (generalized buck-boost) connections that would take 240V high leg to 208V balanced wye...but they would be very non-standard. I think that a proper isolation transformer is in order.

-Jon

 

[GoldDigger]

As drawn that is not buck/boost. It is a straight isolation transformer. So nothing can be connected to the second X2 and the two transformers together need to be rated for the full load power, not just the change in voltage.
Whether you start out with an isolation transformer or not, buck/boost always ends up looking like an autotransformer.

Tapatalk!

 

[Sparkologist4hire]

Thank You

Thank You

Thank You both very much

Wiring diagram #12 from the Jefferson buck boost looks like what he is wanting, but I do not think this will give us a 120/208 4 wire Y sub panel, from the 120/240 4 wire delta service.

As I suspected the size of the buck boost transformers is not large enough either.

John

 

[winnie]

The sizing of 'buck-boost' transformers can be quite surprising.

The reason is that the KVA required of the transformers is just that of the _difference_ in voltage between the two systems.

If you could get a 208:32V transformer (not one of those on the Jefferson list), then you can get an exact transformation from 240V to 208V.

Imagine that you have a 150A, 208V single phase load, or 31.2 KVA.

The transformer would only be carrying a 150A 32V load, or 4.8KVA.

So if there weren't problems with the neutral voltage mis-match, then a pair of 5 KVA transformers might actually be large enough for the load being served.

-Jon

 

[GoldDigger]

The thing that makes buck boost particularly odd when trying to arrive at a high-leg four wire 240/120 is that you still need three boost windings even for an open delta, since you have to boost both A and C separately to keep them balanced with respect to the center tap neutral, and then somehow boost B from the winding which is still connected to the original A point.
Basically you cannot really make the numbers and angles work out.
It is easier to get the proper geometry by using a large full voltage ratio transformer for the side of the delta.

Tapatalk!

 

[Sparkologist4hire]

Simple Solution

Simple Solution

The thing that makes buck boost particularly odd when trying to arrive at a high-leg four wire 240/120 is that you still need three boost windings even for an open delta, since you have to boost both A and C separately to keep them balanced with respect to the center tap neutral, and then somehow boost B from the winding which is still connected to the original A point.
Basically you cannot really make the numbers and angles work out.
It is easier to get the proper geometry by using a large full voltage ratio transformer for the side of the delta.

Tapatalk!

How about suggesting they do not install any transformers and simply avoid utilizing the (High Leg) B phase at the IT Sub panel for any 120v branch circuits, and label the panel as: 120/240 4 wire Delta (Normal Operation) and 120/208 4 wire Y (Emergency Operation)?
I believe the UPS and Condensing units will accept 208 or 240v three phase.

 

[Smart $]

If all the loads on the panel are 208V, either 1 or 3 phase, it would work. 120V loads are possible on the load side of the UPS, provided it has its own output isolation transformer and derives the secondary neutral.

Two caveats: you'd never be able to power a 120V load from the 208/120 panel, and the load on the 240/120 service would unbalanced... which isn't a problem as long as other loads somewhat balance it out.

 

[Smart $]

With that said, I fail to see why "people" mismatch equipment and utility configuration when it is just so much easier to match 'em.

The only scenario where I can see it might be an advantage if they knew the service configuration was going to change sometime in the near future.

 

[kwired]

The other problem here is load balancing - maybe not as much of a problem for the utility feed but can be for the generator.

If you have a high leg system from utility and have limited load on that high leg but a total load of say 10kVA, that doesn't mean a 10kVA generator @ 208/120 is going to be adequate to supply your load in the unbalanced manner it is connected.

 

[texie]

With that said, I fail to see why "people" mismatch equipment and utility configuration when it is just so much easier to match 'em.

The only scenario where I can see it might be an advantage if they knew the service configuration was going to change sometime in the near future.

Gee, I thought it was just me that felt this way. I'd be looking at the whole picture here instead of these contortions.

 

[Sparkologist4hire]

Unbalanced load

Unbalanced load

Both systems are four wire, so we will be able to feed any 120v loads while on utility power from the A and C phases.
The amount of unbalanced load is minimal, because the UPS and condensing units are three phase. The only 120v loads are a couple lights, HVAC alarm / damper motor. My guess is that the rest of the generator 120v loads around the hospital are causing a serious imbalance as the "B" phase must be very light in comparison to the A & C phases.

My guess why the mismatch in power sources is probably because an engineer figured that the main service for the hospital would be changed out as well in the future. One would hope anyway....

Thank you to everyone for the great replies. I have been reading this forum for years and I am happy I finally joined the club.

John

 

[kwired]

My guess why the mismatch in power sources is probably because an engineer figured that the main service for the hospital would be changed out as well in the future. One would hope anyway....

I'm with you on that. A hospital generally is not all that desirable to operate from a 240 volt high leg system. Other then HVAC and a few other motor loads the high leg typically would not have much connected to it, unbalancing the system. Unless this is a pretty small facility, but even then it would likely still be preferred to switch to a wye system in most cases.

I had a small hospital that I did frequent work in and it still had enough load that a delta supply wouldn't have been all that desirable to have around. Most of what was required to be on emergency system was 120 volt loads - really complicates the generator used if your three phase is going to be that unbalanced because of a high leg, or requires a larger then otherwise necessary generator.

 

[Sierrasparky]

I see how the buck boost would work. Transformers in that type of confuguration are not usually determined by the Kva for total power. I think I had a post on that a ways back.

 

[GoldDigger]

I see how the buck boost would work. Transformers in that type of confuguration are not usually determined by the Kva for total power. I think I had a post on that a ways back.

Buck boost would work fine for a corner grounded delta or for s wye. Or for a wye source with only delta loads. It just does not add up for the high leg delta with a center tapped neutral on one phase.

Tapatalk!

 

[Smart $]

Buck boost... It just does not add up for the high leg delta with a center tapped neutral on one phase.

...going to a 208Y/120 3? 4W hmy:

I'd love to see how the engineer intends to make it work if there are 120V loads fed off the panel.

 

[Sierrasparky]

I have not worked on a open Delta in about 20 years , Isn't the connection from farthest leg from the 120 grounded connection give you 208?

 

[GoldDigger]

...going to a 208Y/120 3? 4W hmy:

I'd love to see how the engineer intends to make it work if there are 120V loads fed off the panel.

Just going to a wye with only two of the three phases serving 120V loads actually would be trivial (with one possible exception later). The problem would be how all of the 240V single or three phase loads react to getting 208V instead.
That and the little problem of neutral current in former 120/240 MWBCs.

Tapatalk!

 

[Smart $]

Just going to a wye with only two of the three phases serving 120V loads actually would be trivial (with one possible exception later). The problem would be how all of the 240V single or three phase loads react to getting 208V instead.

Yeah... wouldn't be a problem if they split off the 120V loads to a separate panel and picked off 120/240 1? 3W before the buck transformers... but that would would require a separate transfer switch to pick up the 120/208.

By the time and expense to go through all that, may as well put in a full bore 240/120V?208/120V isolation transformer.

That and the little problem of neutral current in former 120/240 MWBCs.

Well he said new panel... should we assume new circuits???