Rating of CB for protecting PF Cap Banks

[majesus]

I have a question about a low voltage PowerPack J frame CB that will be used to protect a PF correction CAP Bank.

I was instructed to find the "capacitor interrupting rating" for this CB. So I'm thinking... what the heck is the capacitor interrupting rating of a CB? I was told (briefly) a CB has trouble detecting an abnormal condition and also extinguishing the arc when the PF is 0.

For me, when current is flowing, the CB will be able to detect an abnormal condition by the:
a) Thermal effects caused from electron flow
b) magnetic effects caused by the change of electron flow

How does the PF effect this? What about arcing current, how can arcing becoming problematic when the current leads or laggs the voltage by 90?

What about if current was being detecting via a CT and relay, would the PF then become an issue?

Finally, negleting the issues of a capacitor high energizing currents, short circuit rating and continuous rating, what is also required for consideration when sizing a CB for protecting a PF cap bank?

 

[jghrist]

The interrupting capability of a breaker depends on the power factor because of the voltage across the contacts after the current is interrupted. A high voltage can lead to restrike which can lead to higher voltages when capacitors are re-energized. When switching capacitive current, restriking can result in voltages as high as 2.5 times normal line-to-neutral crest voltage.

 

[majesus]

Ah I see what you mean, so this is an issue of transient due to the interaction with the capacitive bank and the system's reactance from a restrike..
Thanks for the help.

However, what I'm trying to visualize is the CB's max capacitor interrupting rating. How is setting a limit on the current flowing through the protection device reducing the chance of a restrike?
The ability of an interrupter to interrupt current
without restriking is determined by its contact material,
contact design, and gap dielectric field strength. IE: arching is a voltage issue over contact area, not current.

B) How does PF come into play with all of this?

 

[davidbeach]

All interruption occurs at a current zero. At unity power factor you have zero voltage across the parting contacts at the current zero. At zero power factor (you'd have similar issues switching an inductor) you are at maximum voltage across the parting contacts during the current zero. That voltage across a very small opening - the contacts have only just begun to part - tends to initiate an arc across that gap.

 

[jghrist]

I'm sure that this isn't a completely technically correct response, but this is how I think about the issue for it to make sense:

The interrupting ability is dependent on both current and the transient voltage. Clearly, a breaker would have no problem interrupting a circuit where the load was switched off at the next device. The current interrupted would be very small, but almost totally capacitive, the capacitance of the cable.

The transient voltage across the contacts is related to the power factor. The current is interrupted at a current zero point. With 100% power factor, the voltage is also zero at that point. With 0% power factor, the voltage is at a maximum.

 

[majesus]

Thanks guys for pointing me into the right direction. I found a very good read here:

http://www.ele.auckland.ac.nz/archives/reports2005/pdfs/Power Systems/proj_62_vkau001.PDF

Refer to 5.1 Restrike on Page 2 which talks about how the current and voltage angle affects restriking.

The pdf refered me to "IEEE Guide for the Protection of Shunt Capacitor Banks" IEEE Std C37.99-2000

An important consideration involving application of circuit breakers or circuit switchers for capacitor
switching is the transient overvoltage that may be generated by restrikes during the opening operation. At
current zero, the capacitor is left charged to nearly full-peak line voltage. Little recovery voltage appears
across the switching device contacts at this instant, and the capacitance-current arc is usually interrupted at
the first current zero after the switching device contacts open. After interruption, the normal frequency alternation
of the voltage on the source side of the switching device results in a recovery voltage across the open
contacts, 0.5 cycle later, approaching twice the peak line voltage [see Figure 45(a)]. If a breakdown were to
occur at 90° in Figure 45(b), the capacitor voltage immediately attempts to equalize with the system voltage.
The circuit is oscillatory. At the first peak of the transient, the capacitor voltage will, depending on damping,
overshoot by an amount approaching the difference between the two voltages immediately prior to the
restrike. This high transient overvoltage may damage equipment. If the current is interrupted at the first highfrequency current zero, the transient voltage peak is trapped on the capacitor bank. The recovery voltage
reaches a value greater than that following the first interruption. However, the contacts have moved farther
apart, and the buildup of dielectric strength may prevent additional restrikes.

Thanks for helping me understand about the restrike problem with CB's and Cap Banks. My last question is my original... what is the Capacitor interuption Current Rating of a CB and how does it tie into all of this.

 

[majesus]

If anyone knows, can you explain what is "Capacitor current Interrupting Duty" as specified in Table 2 "Specifications for Circuit Breaker" on page 3 of the following pdf:

http://www.ieso.ca/imoweb/pubs/caa/caa_SIADraftReport_2007EX-333.pdf

Basic Insulation level 250 kV peak BIL (min)
Rated Maximum Voltage 46.6 kV rms
Rated Continuous Current 1200 A
Fault Interrupting Current 20 kA
Rated Momentary Current 54 kA peak
Capacitor current Interrupting Duty 370 A rms
Rated Transient Recovery Voltage duty for restrike free
capacitor switching
113 kV peak
Rated Transient Inrush current with 0.5 mH series reactor 4.42 kA peak@1.4 kHz

 

[jghrist]

To get more specific than "this is how much capacitor current the breaker can interrupt", you would have to get into the industry standard requirements for circuit breaker rating and testing. The nominal current for the 21.6 MVAR bank at 46 kV is 271A.

 

[majesus]

Morning jghrist, the reason I ask is because I am getting curious. I want to know the reason why it is necessary to have two ratings for a cap bank load. What's the reason for having what seems to be "a derated CB." As in mentioned case:

1) Continuous Current = 1200 A
2) Current Interrupting Duty = 370 A rms

I'm reviewing the IEEE Guide for the Protection of Shunt Capacitor Banks IEEE Std C37.99-2000 to see if I can find an answer or at least clues.

 

[Marmite]

It's normal for capacitor banks to be rated for 30% above nominal current. It looks like this is where the 370A comes from. Nominal current is 283A. The switchgear associated with the capacitor would in theory need to interrupt upto 130% nominal current also.
Regards
Marmite

 

[dpc]

I believe the ANSI C37 breaker standards have requirements for capacitor switching.

Lower ratings for cap bank switching are normal for any type of circuit breaker. This is most difficult task for any breaker due to the leading power factor of the load. It is just a physical reality.

The NEC Article 460 requires conductors and disconnecting means for capacitors be rated at not less than 135% of the cap bank's rated current to account for harmonic currents, etc.

 

[majesus]

Thanks guys for the input, but it's not what I'm "exactly" looking for.

I'm looking for the reason why the CB continous rating has to be derated for use in Cap Banks. Probably because of the 0 pf and the potential re-striking issue, but why is the current being derated? How does reducing the amount of current flowing through a CB reduce re-stricking? My guess is that by de-rating the CB for Capacitor bank application, you are essentially choosing a bigger CB for switching duty. Hence the CB's contact point would have more surface area thus reducing the chance of arching since it is a function of voltage over contact area.

However, that is my guess... something I came up with. Is it right or wrong? I am looking for the answer.
I read IEEE Std C37.99-00, now I shall read C37.06.

 

[apowerengr]

Breaker ratings (like fuses, switches and other overcurrent stuff) are designed at a specified (either by the mfg. or by standards) X/R ratio. Capacitor bank X/R approaches infinity. So, what the breaker manufacturer is trying to tell you is even though you have a 1200 amp breaker which can be used to switch loads on and off up to 1200 amps the switching duty is at the specified maximum X/R ratio, when considering loads having a higher X/R than the standards under which the circuit breaker were manufactured and tested to causes a derating of the device. Switches have the same limitation, you can have a 1200 amp gang-operated switch with arcing horns which the mfg. will tell you can break the magnetizing current on a transformer up to some (relatively small) value of transformer MVA (5% of the load rating for example). This is also a reflection of the load break rating based on an X/R ratio which is greater than the standard requirements. S&C's 600 amp alduti-rupter distribution load-break switch is only rated for switching 600kVAr at 15kV (~20 amps) but 600 amps at the rated X/R or less.

Load-break is different than fault-break as the internal mechanisms are different. A piece of equipment has a fault interrupting rating (also stated at a maximum X/R ratio) and a close & latch rating which are different but it's the same device and the same current carrying contacts.

Does this address your question?

 

[majesus]

Hi Apowerengr, Thanks for the input!

I understand what you mean, but as you mentioned: "the magnetizing current." The X/R ratio that we speak of, isn't that the ratio of inductive impedance/resistance?

It is simple to visualize the difficulty when interupting the current of an inductive load due to the fact that I=di/dt can not instantaneously change. As the X/R becomes largers and larger, that is why you are required to derate the CB.

But for a CB that shall be applied for a capacitor bank, I don't see how current will be an issue because the flow can instantly stop (assuming L of the system ~ 0)
That's where my question lies: why for a CB must the capacitor's current be derated?

Note: Another thing, I was talking to Schneider this morning, and they said the LV CB's used for Capacitor bank application have no issues in regards to this (which is great, as my application is for 600V).

It is only when MV and HV CB are used, that you it becomes more application specific where specific duty CB are required for Capacitor Banks. IEEE C36.07-2000 address various CB continous ratings and their dereated current ratings used for Capacitor Bank.

 

[davidbeach]

Read about capacitor switching in Greenwood. Nasty stuff. Sure, the current can change instantaneously, but the voltage across the open contact doesn't decay the way it does with any other type of load. The current zero happens at maximum voltage, that is an important fact.

 

[majesus]

Allen Greenwood's Book on Electrical Transients In Power Systems is awesome and I did read this afternoon the section 5.3 on Capacitor Switching as well as Section 6.4 on three phase capacitor switching. It has an excellent explanation on Oscillation, restriking and how shoot through can significantly increase the voltage on a cap.

It explains the cause and derives equation to use for calculations. But it didn't explain why a HV CB would require specific duty ratings and a LV CB does not. Nor could I find an explain why derating a CB's continuous current would solve these problems.

Maybe I'll have to wait until I am at some sort of circuit breaker presentation and ask an expert.

 

[Marmite]

This is the text from IEEE C37.012 concerning switching capacitors:-
4.7.1 Capacitor Bank Current.
Circuit breakers are to be applied according to the actual capacitance current they are required to interrupt. The rating
should be selected to include the following effects.
1) Voltage Factor. The nameplate reactive power rating of the capacitor bank, in kilovars, is to be multiplied by
the ratio of the maximum service voltage to the capacitor bank nameplate voltage when calculating the
capacitance current at the applied voltage. This factor can be as large as 1.1, since capacitors can be operated
continuously up to 10 percent above the capacitor rated voltage.
2) Capacitor Tolerance. The manufacturing tolerance in capacitance is -0 to +15 percent with a more frequent
average of -0 to +5 percent. A multiplier in the range of 1.05 to 1.15 should be used to adjust the nominal
current to the value allowed by tolerance in capacitance.
3) Harmonic Component. Capacitor banks provide a low-impedance path for the flow of harmonic currents.
When capacitor banks are ungrounded, no path is provided for zero-sequence harmonics (third, sixth, ninth,
etc), and the multiplier for harmonic currents is less. A multiplier of 1.1 is generally used for a grounded
neutral bank and 1.05 for an ungrounded neutral.
In the absence of specific information on multipliers for the above factors, it will usually be conservative to use a total
multiplier of 1.25 times the nominal capacitor current at rated capacitor voltage for ungrounded neutral operation and
1.35 times the nominal current for grounded neutral operation.

See also the attached Cahier Technique from Schneider.These are available on Schneiders website, but you have to register so I've attached the pdf.

Regards
Marmite

 

[JensenDrive]

It is interesting that the breaker is derated for capacitor current, but I never really asked myself the particular mechanism that leads to the need to derate it. If I had to guess, start by taking a look at Figure 5 of the New Zealand paper. At the moment current crosses zero, if current is to be interrupted, the contacts are already well apart, there HAS to be internal arcing going on prior to the interruption, and the voltage across the arc is higher, compared to what would exist for a resistive current interruption. Also, the internal arc resistance over the short current zero appears to need to rise at a much higher rate to interrupt a capacitive current, compared to a resistive current. This implies that the contacts must be farther apart at the zero current point to interrupt a capacitive current, compared to a resistive current. This implies the internal arc is going on longer prior current interruption in the capacitive load. This implies higher contact wear and heat. I would guess the derating is the result of higher Varc*Timearc during the capacitive interruption process.