Current-limiting fuse to allow transformer replacement

[lavm]

Hi all,

I've recently done a short-circuit analysis on EasyPower to evaluate the impact of replacing a 1.5 MVA xformer by a 2 MVA. What came out is that the old FPE 50H breakers cant withstand the new Icc of over 50kA (about 55kA) so I am considering two options:

1. Replace every breaker by an appropriate model ($$$);

2. Add a current-limiting fuse at the secondary of the xformer.

I'd like to know if option 2 is viable and if the fuse will blow at every single fault. What I'd like is for the fuse to blow ONLY when the fault current is about to get higher than 50kA and let the breakers open if it's not the case.

So what do you think?

Mike

 

[tem1234]

Can you rise the impedance of the 2 MVA transformer to lower the SC current?

 

[davidbeach]

Nope, not permitted by code. Series ratings have to be established by testing and that would exceed the cost of replacing all the breakers (and probably destroy a few breakers in the process). The NEC foolishly leaves open one other possibility but nobody of any competency will touch that exception because it requires one to state that they know the unknowable.

 

[dpc]

I agree with David Beach - it's unlikely that those old breaker were ever tested in combination with a modern CLF upstream, so I suspect that's not an option.

I'd just replace the entire secondary lineup rather than trying to replace every breaker.

Another option would be a current-limiting reactor but I suspect you'd need a lot of reactance, and they have their own problems.

 

[wareagle]

lavm
What is the impedance of the transformer you have?
Why can't you order one with a higher %Z?

 

[rbulsara]

Try what wareagle said. (david and dpc are right though).

What voltage is this at? 55kA sounds high for a 2MVA units (assuming 480V) unless you have other sources connected. Have you taken the source impedance behind the transformer into account?

 

[waross]

I saw an elegant solution to this problem years ago.
The designer calculated that 100 feet of cable would add enough impedance to drop the available fault current to a level that the breakers could safely interrupt. He then spec'ed that every breaker must be fed with no less than 100 FT of cable. This was the poor man's series reactor. If you are close to the rating of the breakers, you may make up the difference with cable impedance.
I have also seen series reactors used.

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

 

[lavm]

Thank you all for your answers.

The transformer was installed before I was hired, they've just never calculated the impact on the equipments. There are many similar cases (20 transformers + 100+ branch breakers) covered in the analysis so I won't list every impedance but it should be around 5.5%.

The Icc sometimes reaches 60kA so adding cable length is out of the question for most cases.

dpc I don't understand what you mean by replacing the secondary lineup. Does it mean rising the % of the tranformer?

Thanx again for the help.

Mike

 

[itsmoked]

He means the breakers.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[dpc]

Yep, the breakers and the supporting structure, everything downstream of the transformer. Those breakers you have are getting pretty old as is the bus and compartments the breakers attach to. If the breakers are underrated, how about the bus bracing of the switchgear itself?

 

[davidbeach]

Bill, I've done the cable bit before. More than once I've had to specify that the new feed, from a new service, to an existing panel has to run at least a certain distance. Usually much more than the most direct route but the cable cost is far less than a panel replacement cost, plus I knew that the utility number was something that would never happen in a real fault so I was happy to get just below the panel rating.

 

[waross]

Thanks for the support, David.

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