Power Transformers In Parellel

[Mbrooke]

What advantage would be reasonable in a scheme using 4 or 5 smaller power transformers operating in parallel vs 2 larger units run normally open as typically seen in the wild?

For example four 12/16/20MVA transformers vs two 40/50/60MVA 115-23kv transformers.

Similarly five 15/20/25MVA vs two 60/80/100MVA 230-34.5kv transformers.

60 and 100MVA loading respectively to allow for failure of one unit during the hottest of days (assume load transfer disabled or not available to other substations).

In running calculations the fault current is about the same for both scenarios.

I know that one large transformer is cheaper in total than many smaller transformers, however, I can not help but notice the capacity of each smaller unit is more utilized per dollar through out the in service life.

This concept tends to apply to generators, underground cables, interconnecting auto transformers, ect but rarely to distribution substation applications. Is there a technical prohibition?

 

[waross]

Did you consider the cost of bus bars, switch gear,protection and real estate?

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[sushilksk]

Two advantages may be N-1 criteria & transportaion issues in remote / hilly regions.

 

[Mbrooke]

More real-estate yes, but the rest isn't to bad- ie the bus segments can be of a lower current rating since trafos and distro bays alternate.

 

[Mbrooke]

N-1; with the parallel design their is less trafo capital cost in theory since the loss of any one unit is smaller than a large 60MVA unit, meaning each is loaded closer to their FA/FA rating instead of both 60MVA units never being loaded above 30MVA 99.995% of the time.

 

[sushilksk]

One reason may be capacity addition over the period of time. Say 1st year only two 20MVA transformer (considering N-1) and later on adding 20MVA transformers depending on load requirements. say after 10 years there may be 4 nos. 20MVA transformers in parallel.
Use of 2 Nos. 60MVA at the begining is not justified in this case.

 

[Mbrooke]

You make a really good case regarding steady load growth.