Protection of Air Core Dry type Reactors connected to a Transmission Line

[Hakakiar]

Hello,

We are installing a 30MVAR 230kV shunt air core dry type reactor to be connected to a Transmission Line for voltage control, and I am trying to determine protection requirements for the reactor. I have the following questions:

1) What kind of protection should be provided for shunt dry type air core reactors connected to a transmission Line?
2) Should there be dedicated reactor protection provided or the line protection (used to protect the line) can be used to protect the reactor as well?
3) Should auto-reclose be allowed to reclose onto a faulted reactor? This case would apply where no fault discrimination is available because the line protection is also used to detect faults in the reactor. The advantage of this is that no standalone CT would need to be installed for the reactor (less cost).

We usually use oil immerse reactors for this type of application and have never used air core dry type reactors. It seems that they are not very popular in the industry as it has been a bit of a pain to get feedback from anybody.

Thanks all.

 

[144x]

you can find the answer to your questions in the following references:

1- TECHNICAL BROCHURE
Title:
Protection, Monitoring and Control of Shunt Reactors-CIGRE WG B5.37

2- C37.109-2006 - IEEE Guide for the Protection of Shunt Reactors

3- IEEE Guide for Automatic Reclosing of Circuit Breakers for AC Distribution and Transmission Lines

 

[submonkey]

Hi,

A few questions:-

Is the reactor at one of the line ends, or somewhere in the middle?

Will it have an earthed or unearthed starpoint?

What kind of line protection are you using? Is optical fibre comms
available?

Thanks,
Alan

 

[bacon4life]

Sometimes things are unpopular because they are a bad idea. What has recently changed about reactor technology that drives your decision to pioneer the use of 230 kV air core shunt reactors?

In general, the choice to use reclosing weighs the likelihood of a successful reclose against the possibility of increased collateral damage at the fault location. Although air core reactors pose much lower fire risk than oil filled reactors, I don't see how you would achieve turn to turn fault protection using just line relays.

 

[Hakakiar]

Hello 144X,

Options 1 and 2 provide useful information but only related to dry type air core reactors used in tertiaries of power transformers and not connected to Transmission Lines. However some of the criteria apply to both applications. There are some industry papers that offer more insight but still not convinced. I would like to see what other have actually done in this application to make a more informed decision and be able to justify it.

Thanks.
Hans

 

[Hakakiar]

Hi Alan (Submonkey),

Each station (line end) will have a Y shunted solidly grounded reactor. Another utility is in charge of the remote end, so I don't know if the remote reactor will be air core or oil immersed type.

The line protection will be redundant distance with POTT (Permissive Overreaching Transfer Trip) as the main protection element. So communications between the terminals will be required.

According to guides and papers the risk of a phase to phase or a phase to ground fault in this type of reactor is close to non existent due to the physical distances between phases and the amount of insulation to ground. The only concern are turn to turn faults, that seem to be actually happening in the industry due to the switching duty of the reactors, which my application will also have (the daily switching, I mean).

However, detecting reliably a turn to turn fault on time seems very difficult and maybe unachievable; and even if this fault is detected, the reactor may be already damaged and will require replacing. If the turn to turn fault is left unattended, it will develop into a phase to ground fault.

Because the reactor is connected to the line at the terminal, the protection for the line will detect the phase to ground fault at the reactor and will clear the fault.

The other factor into this is that if I let the line protection clear the ground fault at the reactor, then auto-reclosing will happen into the faulted reactor. For me this would be a no-no scenario and I would provide dedicated protection for the reactor, if only to know that the fault happened there; but my boss is challenging me to really justify this need. Providing dedicated protection will mean that stand alone CTs will be required at a cost.

What do you think?

Thanks,
Hans

 

[Hakakiar]

Hi bacon4life,

The dry type air core reactor at this voltage level was not available in the past when the criteria for line connected reactor was implemented in the utility I work for. So when they appear nobody thought in using them until now, and not be me by the way, but I have to provide protection for it anyway.

As you implied the main advantage of this type of reactor is that there is not oil involved, so no risk of catastrophic explosion, or environmental damage and lower maintenance costs. And yes, the question is should I try to provide turn to turn protection?

Actually one manufacturer has recommended that only phase overcurrent be provided.

Thanks,
Hans

 

[submonkey]

Hi Hans,

This is a very interesting problem. My previous employer used air cored
reactors fairly extensively, and I did spend some time considering how
to protect them.

I generally dislike the idea of discounting certain fault types just
because those present can't envisage how they might arise. I've seen
a lot of things in life which I wouldn't have predicted. Busbars at
330kV have massive spacings, yet these are still protected.

I do agree that faults between turns are the most likely failure case,
and these are difficult (but not impossible) to detect.

You've obviously considered the possibility of adding HV side CTs,
presumably with the intent of installing differential protection or
similar across the reactor. Sensitive high impedance differential
protection would be a good thing to install. It should be noted that
differential will not detect faults between turns.

How does one detect faults between turns? One method I have considered
is to install two identical shunt reactors on each phase, and detecting
the current unbalance on the neutral side - a bit like a capacitor neutral
unbalance circuit. This method carries some cost because it requires two
smaller reactors. I haven't personally tried it yet, and would be interested
to hear if it has been used by others.

In your position of having to justify the expense to your boss, it might
be helpful to consider what can be done to improve the situation at minimal
cost.

Suppose you install a set of CTs at the neutral end of each reactor - these
could be standard resin cast CTs on unistrut, and could be added for a few
$k at most. You could use these to:

- Apply differential protection to the reactors and line by using a line
differential relay with an extra CT to subtract the reactor current.
i.e. for a single phase, trip when:
Ia_local_line + Ia_local_reactor_neutral + Ia_remote_line + Ia_remote_reactor_neutral >> pickup
Many relays designed for 1-1/2 breaker applications have extra local CT
inputs which could do this.

- Detect a change in the reactor impedance by comparing the output of the
line VT to the measured neutral current. You could do this with a spare
distance element. This method has potential to detect faults between turns.

- Detect open circuit faults which may be the end result of faults between turns.

I haven't tried the method above personally either.

Mostly these reactors have ended up with differential or overcurrent, and so
far I haven't seen one trip.

Please let us know what you end up with.

Thanks,
Alan