Paralleling generators at sub-distribution panel

[davva]

with reference to attached sketch

we have a distribution system that is currently operated with two generator plants that are independent, i.e. the 690V switchboard operates independently of the 480V switchboard.

Somebody has come up with the idea of reusing an existing sub-distribution feeder from the 480V sub-panel of the 690V system and using it as a tie to the 480V main switchboard as dashed on the drawing.

This will give increased operational flexibility in certain situations.

Just looking at the arrangement makes me a bit nervous as I have not seen this happen before.

Obviously consideration needs to be taken into account of fault levels to make sure all equipment is suitable.

In addition there needs to be some form of synchronising control implemented.

But I have reservations regarding power flow and concern of how to prevent overloading of the cable tie or components in that connection - as you will see it is a cable of limited cross section.

I wonder if my concerns are warranted or if it is actually simpler than I think. If so what considerations and schemes need to be implemented to make sure this is a reliable and robust upgrade?

appreciate any help/comments/suggestions.

 

[rbulsara]

The scheme is technically feasible and your concerns can be addressed with proper design and controls. This is not something out of ordinary.

As for the pros and cons of the solution and justification of the design, you need to ask your consultants.

Rafiq Bulsara

http://www.srengineersct.com

 

[waross]

Protecting the cable is easy, just use a breaker that is sized to trip when the cable capacity is exceeded.
The challenge is designing a control scheme that will avoid tripping the breaker.
Fault levels should be checked but given that islanded generators generally have lower fault levels than grid fed transformers and including the impedance of the interconnect cable, the chances are good that you will be safe.
Generator control: How are your generator governors controlled? Are you using a simple droop control system or are you using generator load control panels?
First, frequency of both generator groups must be exactly equal. Further more, the power flow between the generator groups will depend on the phase angle between the two generating sources.
Now, when you add a generator to lineup, it should be quite simple. Adjust the governor until the scope is turning slowly clockwise and at 12:00 o-clock close the breaker. Then adjust the governor until the incoming generator is sharing the load with the others.
Here you will have to adjust the speed/frequency of more than one set, and then adjust several governors to get the proper balance. While you are doing this, try to stay as close to 50Hz as possible.
But that's old school. You are probably using panels capable of controlling the governors directly on several generators.
The best solution may be a load monitoring panel which will send an error signal to the load control panels for both groups of generators.
The simple solution will be to decide how much power you want to transfer on the inter-tie cable and generate a bi-polar error signal. The error signal will be sent as is to one group control panel and sent inverted to the other group so that one set of governors are slightly advanced while the other set is slightly retarded.
That takes care of power flow. Now we will have to balance power factors to avoid excessive reactive currents in the inter-tie cable.
It is do-able, but bring your check book.
These issues are present with many grid inter-tie schemes however the solutions that are economical for a system transferring GigaWatts of revenue power may be quite expensive to provide convenience for a small system.
That's my take. I'm hoping that catserveng shares some comments. There may be an economical panel available that will do all this.

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

 

[davva]

Thanks Waross, yes these were my concerns.

I don't know the details on the load sharing/governing on each system at present and I would think the kVAR load sharing is maybe not even included.

You mention "Further more, the power flow between the generator groups will depend on the phase angle between the two generating sources". Are you referring to the 1MVA transformer? This is a delta/delta transformer at present.

 

[WoodrowJWeen]

Not exactly sure what you're trying to accomplish. Another limiting factor will be the 1MVA xfmr. Might an ATS arrangement be simpler and not involve synchronizing?

 

[catserveng]

This looks like a marine system with a propulsion bus and a house bus. You want to be able to use house power for the propulsion bus and the other way around?

For 6 generators the Woodward EasyGen would be a good choice. With that said there needs to quite a bit more information gathered and things looked over for a succesful retrofit.

There are a number of other control systems, such as Basler and Deep Sea that would likely also work, your best bet may be to contact a local engineer familar with the type of system you're trying to retrofit and talk to the vendors to get an idea of the best course of action.

Hope that helps.

Miek L.

 

[waross]

Phase angles. When two generators are paralleled, the sine waves may be exactly in phase. If the throttle is opened on one of the generators, the sine wave will advance slightly ahead of the sine wave of the other set. We are talking about plus or minus a few degrees. This is the angle that influences the power sharing. Nothing to do with transformer wye:delta phase shifts.

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

 

[davva]

yes catserveng you are near enough correct in describing the philosophy.

Thanks all, it seems as I thought that it needs some careful consideration.