Shunt Fault Current Limiting Reactors

[vtpower]

I am looking for sort of the same thing that mpparent was looking for. In my situation, I have introduced a couple new 115/34.5kv sources to our subtransmission. Off of our subtransmission we actually have a few feeders that are directly tapped to the 34.5 kv through a recloser and regulators. With these new sources, the fault current will be over the rating of most of our distribution equipment, which is rated at 12kA. In order to not spend a ton of money to change all of our distribution protection, we will be installing fault current limiting reactors. I am curious if there is any information with this type of set up and also what kind of losses are introduced on the system, and nominal voltage drop if any? Thanks for the help.

 

[scottf]

The losses will depend on the reactor design and manufacturer. It's really tough to give estimates on that kind of thing...you'll need to get quotes from reactors manufacturers (Ritz and Trench) to be sure.

Voltage drop across the reactor will be a function of the impedance/inductance of the reactor, which will be based on how much impedance you need to add to get your fault current down below 12 kA.

 

[dpc]

Once you determine the amount of impedance you need to limit the fault current, the voltage drop at normal load is easily calculated.

To determine the losses, you will need the X/R ratio of the reactor from which the resistance can be determined. This will vary quite a bit depending on the reactor size and construction. Conrad St. Pierre does have a graph of typical X/R ratios in his "A Practical Guide to Short Circuit Calculations" if you have that. If not, I'd just contact suppliers as Scott suggested. Once you have the resistance, losses are just I^2*R.

You can expect X/R ratios in the range of 15 to 100.

 

[scottf]

dpc-

Any bid that you get for a reactor will list out the total losses in kW, so you shouldn't need to do any calculations. Also, the losses of a reactor aren't simply an I^2 * R calculation.

 

[dpc]

scottf,

Sometimes we need to model existing reactors for which no resistance data is available.

What losses occur in an air core reactor other than resistance losses?

 

[robthecob]

We used 1% reactors at 12.5kV to reduce fault current on the 12.5kV-480V substations from 70,000 amps to 64,000 amps. The power factor also dropped 2% with these reactors.

 

[ScottyUK]

Hi dpc,

If the reactor is iron cored such as the type installed at our site for fault limitation, there are reactor iron losses to consider. Other than that, I can only suggest skin and proximity effects modifying the DC resistance. On conductors of this size these effects would be minimised by using a Roebel type winding I suspect.


----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[scottf]

Losses are also a function of the winding design (magnetic coupling effect) and can also be effected by using insulators with steel caps, etc...

 

[PowerfulStuff]

ScottyUK,
Can you use iron core reactors for fault current limitation? I had the idea that they would saturate under and become ineffective under fault conditions.

PowerfulStuff

 

[scottf]

You wouldn't typically use iron-core reactors for current limiting applications. Normally only air-core. Air-core reactors maintain a linear response with current, i.e. no saturation.

 

[ScottyUK]

The installation and calcs were done by a third party as part of a turnkey job. I'd not thought about it much because it was installed before my time, but you have me wondering now.

I guess the iron core could be designed so that it didn't saturate under the highest available fault level, but without doing any calcs that is going to be a big reactor. I'll do some more digging on this when I get back to work. If what you're suggesting about saturation can happen in our installation then we have potentially got a really big problem with fault level on our 11kV system.

FWIW the reactor is the approx size of a 2.5MVA transformer, oil filled iron cored, rated for ~2500A continuous. System is resistance earthed and has approx 60kA fault on the unlimited side of the reactor. I'm quoting from memory here, so I'll re-post when I have a chance to look at this properly.



----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[scottf]

ScottyUK-

What do you mean by "has approx 60kA fault on the unlimited side of the reactor"?

I don't understand what "unlimited side" means.

 

[ScottyUK]

Hi Scott,

We have an 11kV bus which was extended to accomodate an embedded generator on our 11kV network, primarily for black start capability but also capable of commercial operation. The addition of this third source to the two existing infeeds caused the fault level of the original board to be exceeded. To overcome this problem a new bus section was added, with a fault-limitng reactor as the bus tie between the original and new sections. The new bus has two infeeds, one from the embedded generator plus one moved from the original board, while the older section now has only one. The 'unlimited' side of the board - and I apologise for the poor choice of words - is the newer section which has a very high fault level because of the embedded generator plus a grid connection. The older part of the board which has the lower fault rating has the fault infeed from the new section limited by the reactor. On the occasions when there are only two infeeds available the reactor is bypassed.

As an aside, the voltage regulation on the older section of board suffers badly when the reactor is in service because the grid transformer regulates its terminal voltage which is connected to the new section of board. Motor starting causes some severe volt-drops when the reactor is in service. The whole arrangement is, frankly, not a pretty solution.



----------------------------------

If we learn from our mistakes,
I'm getting a great education!