Reducing Fault Level Issues

[ukgraduate]

Hello,

We are looking at reducing the fault level on a new system in order to save costs on cables and MCC boards. The fault level is only an issue on the MV side. We do not want to put in reactors as it will increase the losses. We can’t increase the transformers impedance upstream as they are the utilities, the utility cannot provide a guarantee to not parallel their transformers so we are designing on the basis of maximum fault level. So we are looking at putting in fault limiting fuses on the incoming feeders. We want to reduce the fault level from 25kA to around 16kA.

Can anyone tell me what are the issues/disadvantages or things we need to investigate before limiting the fault current? Protection grading shouldn’t be a problem as our plant has not been built yet and we will grade at a later stage. I am thinking of voltage stability etc?

 

[cranky108]

Consiter what happens if the current limiting fuse blows. How much of your plant will be dark.

How much losses do you expect from reactors? If you keep the PF near 1 then they should not be that high.

 

[dpc]

Fuses don't really reduce fault current. They simply clear a fault before the peak current occurs. So the only way this could work is is these main incoming fuses clear before anything else.

I don't think 25 kA of fault current is going to have much impact on the cost of the cables.

In the US at least, certain circuit breakers are tested in combination with upstream fuses and this "series rating" can increase the rating of the circuit breaker dramatically. But again, this eliminates any reasonable chance of coordination.

In today's systems, 25 kA is not a lot of of fault current. It is fairly easy to buy MCCBs with 42 kA of interrupting, even in smaller sizes.

 

[rcw retired EE]

If you have the budget, put in a reactor and short out the reactor with current limiters designed for this purpose. G&W's CLiP assembly is one product we have used in this application.

It uses a buss bar fitted with explosive charges to short out the reactor and a current limiting fuse. A CT senses the fault current and triggers the shaped charges to sever the bus bar. That transfers the load to the current limiting fuse which blows and places the reactor in the circuit.

No reactor losses.

But the cost is probably more than just going with 65 KAIC gear.

 

[m3ntosan]

run split prefault and put in place auto-switching scheme on couplers, sections etc to couple up site for transformer loss

May you grow up to be righteous, may you grow up to be true...

 

[jraef]

dpc, he is looking at the MV side.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[dpc]

Sorry, I missed that little tidbit - saw references to MCCs.

But fuses still don't reduce fault current even on the MV side.

 

[ukgraduate]

Thanks for the help so far.

I am aware that the fuses dont reduce the fault only limit it. The idea behind the fuses was to put them on the incoming side of the MV SWB in order to use smaller cable sizes on the outgoing side of the MV switchboard. We have about 15km worth of cables.

Can someone tell me if there is a fault on one of these cables and assuming it pulls enough fault current to blow one of these fuses. Will all four have to be replaced, will they have started to fuse and hence cannot be relied upon. I am assuming that they wont all blow at once but maybe they are damaged? We have calculated that if 1 fuse blows we should be able to still run the whole plant.


Also can anyone point me in the direction of some good literature on reactors i have spent ages on the web trolling for how they work to reduce fault current and how they are installed on the network. (Our PF is assumed to be around 0.95)

From the sounds of it the fuses are the cheapest way to reduce (limit) the fault enough to use smaller cables. Are there any issues with limiting the fault that I need to consider?

 

[jghrist]

In my experience, the sizing of MV cables is seldom determined by fault currents. At least as far as the conductor size is concerned. The cable shield could limit fault current capacity for Ø-ground faults if there is no separate ground wire return.

A possible problem with reactors would be voltage regulation. The reactors will increase voltage drop.

Can someone tell me if there is a fault on one of these cables and assuming it pulls enough fault current to blow one of these fuses. Will all four have to be replaced, will they have started to fuse and hence cannot be relied upon. I am assuming that they wont all blow at once but maybe they are damaged? We have calculated that if 1 fuse blows we should be able to still run the whole plant.

What four are you talking about? Fuses? Why four fuses? If one phase blows, you don't want to run the plant. The possibility of doing so is a problem with using fuses for protection.

But fuses still don't reduce fault current even on the MV side.

What's the difference between reducing fault current and limiting fault current? If the available fault current is above the current-limiting range of the fuse, it reduces the current.

Is there such a thing as series rated protection at MV?

 

[cranky108]

Voltage problems can be fixed with capacitors.

Did I miss something, how do you run a plant with a phase missing?

There was a scheme I saw once with a current limiting fuse and a reactor. The fuse blows and requires all the current to flow through the reactor. It kept fault currents down without the problems of the reactor.