Zone Selective Interlocking

[timm333]

I am trying to figure out how does zone selective interlocking (ZSI) work. In traditional coordination (without ZSI), for a downstream fault the downstream breaker will trip instantly and the upstream breaker will not trip. And if we have coordination with ZSI, then again the downstream breaker will trip instantly and the upstream breaker will not trip. So what is the difference? How does ZSI reduce the arc flash energy by reducing the trip time of the downstream breaker? Thanks.

 

[SG-1]

If the downstream breaker fails to trip a signal goes to the upstream breaker that overrides the time delay, causing it to trip ASAP.

 

[timm333]

But the document at the link: https://www.ewh.ieee.org/soc/ias/psp/minutes/salt_lake/ETU_definitions.pdf

4th paragraph from top says that without ZSI, the downstream breaker will still trip (with time delay). My question is that why do we have to give time delay to the downstream breaker (when no ZSI). Why we cannot trip downstream breaker instantly (when no ZSI) ?

 

[d0nut]

It is assuming there are more breakers downstream of the subject breaker, and that the subject breaker has a delay to coordinate with the downstream breakers.

 

[jim dungar]

The issue has to do with faults between the upstream breaker and the downstream breaker.

For coordination the upstream breaker must be set sufficiently high (enough time delay) in order to let the downstream breaker clear a fault.
But if the downstream breaker does not clear the fault, do to it failing, or more likely to a fault occurring on its line side conductors, the upstream breaker will follow the time delay associated with its selectivity settings.
Now in a ZSI system the upstream breaker trips as fast as possible, ignoring its coordination, unless it receives a restraint signal from a downstream device. This way an arc flash event can be cleared by the upstream device with minimal delay.

 

[ATSman]

I agree with Jim but would like to put in my 2 cents.
ZSI enhances the coordination of circuit breakers in an electrical distribution system. Rather than relying on the TCC and settings of the individual breakers to provide this function, ZSI sends a blocking signal to the upstream breaker(s) in a multi level feeder system to allow the breaker closest to the fault to trip. If this breaker does not clear the fault then the next upstream breaker is commanded to trip, and so on. Consider this scenario: Most large commercial buildings only have GF protection on the Main Breaker(s) only because it's required by the NEC, but not the feeders, due to cost. Consider this scenario: In a building with 277V lighting an electrician is changing out a lighting ballast hot. He shorts the hot wire to ground, not only does the 20A branch panel breaker in the room trip, but the 480V 4000A main breaker 40 floors down in the basement power room trips on Ground Fault (typically set at 10% or 400A) Why? Because a #14 wire can momentarily carry hundreds of amps before a GF trip and that fault current can travel that far in an instant. Customer calls us and says " Their's something wrong with the GFR on my main breaker, can you come and test it?" Sure but the GFR is only doing its job, your system is not designed with muti-level GF protection with ZSI. I can't tell you how many times this has happened to us.

http://www.eatoncorp.com.au/ecm/groups/public/@pub/@oceania/@elec/documents/content/pct_372886.pdf

file:///C:/Users/user/Downloads/PU01202002E.PDF

 

[jim dungar]

ZSI enhances the coordination of circuit breakers in an electrical distribution system

I never consider ZSI when I am deciding on the settings for coordination.
The breakers need to be coordinated without ZSI.

In your example the upstream ground fault would never get a blocking signal and thus would trip immediately, unless the downstream breaker was supplied with GF ZSI. However if the upstream device had been properly coordinated with the downstream 20A device no downstream ZSI would have been needed. I cannot remember a time that I have not been able to coordinate a single main, 1200A and larger, GF with a 30A maximum branch breaker.

Yes, ZSI enhances the performance of a coordinated system, it just doesn't replace it.

 

[ATSman]

I agree and disagree. A 20A breaker does not have the capability of ZSI and cannot send a blocking signal. I was just stating what we see in the real world as to nuisance tripping of a main breaker due to a lighting ballast short. But your statement that the main would trip immediately is incorrect. A GFR device has a pickup setting: 200A-1200A and time delay setting of 0- 1.0 sec and typically the time is set between 0.3 to 0.5 sec to allow for downstream devices to clear the fault. By design the only time a GFR would trip instantaneously is if the fault occurred on the load side main bus of the 4000A breaker its protecting. In my example both the 20A breaker and main breaker tripped because according to PE's I have worked with "You cannot coordinate an instantaneous fault" , it is a crap shoot and usually all breakers in the lineup, trip. I was only emphasizing the importance of GF protection on every feeder level or at least a 2nd level to prevent tripping of the main breaker when a lighting ballast shorts.

 

[jim dungar]

... according to PE's I have worked with "You cannot coordinate an instantaneous fault" , it is a crap shoot and usually all breakers in the lineup, trip. I was only emphasizing the importance of GF protection on every feeder level or at least a 2nd level to prevent tripping of the main breaker when a lighting ballast shorts.

But an instantaneous fault for a 20A breaker may not be an instantaneous fault for a 1200A breaker. It is amazing how much current choking effect as little as 25' of #12 can have.

For 15 years, I have been able to create breaker settings that would have large main GF coordinate with 30A max breaker (sometimes even higher).
The problem has been that rarely were the settings ever implemented on the project. It was not unusual on 'small-medium' projects for the electrical contractor to be off the site before the coordination study was finally even performed, so who knows if the breakers were ever adjusted.

 

[timm333]

I understand that the ground short circuit protection (GSCP) and the ground fault protection (GFP) are not the same thing. GSCP is for high currents (from phase to ground) and these are cleared by the regular phase breakers. Whereas the GFP is for low leakage currents (for example arc currents) and these are cleared by the ground fault sensor.

These are two different things. Why do we still coordinate the ground fault protection of the upstream breaker with the phase protection of the downstream breaker?

 

[paulengr]

I understand that the ground short circuit protection (GSCP) and the ground fault protection (GFP) are not the same thing. GSCP is for high currents (from phase to ground) and these are cleared by the regular phase breakers. Whereas the GFP is for low leakage currents (for example arc currents) and these are cleared by the ground fault sensor.

These are two different things. Why do we still coordinate the ground fault protection of the upstream breaker with the phase protection of the downstream breaker?

Because with a large enough ground fault it appears as the same thing.

 

[paulengr]

I am trying to figure out how does zone selective interlocking (ZSI) work. In traditional coordination (without ZSI), for a downstream fault the downstream breaker will trip instantly and the upstream breaker will not trip. And if we have coordination with ZSI, then again the downstream breaker will trip instantly and the upstream breaker will not trip. So what is the difference? How does ZSI reduce the arc flash energy by reducing the trip time of the downstream breaker? Thanks.

It’s not the downstream breaker that’s the problem. An arc flash sticker uses the trip time of the upstream breaker.

Ultimately if we know the arcing fault current we can just set instantaneous on all breakers just below that value. This gives us minimum trip times. But coordination is an issue since all breakers attempt to trip simultaneously on an arcing fault. So with ZSI all upstream breakers wait one extra cycle. If the breaker recognizes an instantaneous trip and no downstream breaker assets ZSI, then it trips. So typically all downstream breakers are set to instantaneous anyway. An upstream breaker needs to wait for the downstream breaker trip time if ZSI is not used. So with 1 cycle to recognize a fault and say a 3 cycle breaker we have to wait at least 4 cycles plus a safety margin or 5 cycles or about 60-80 ms to trip on the next upstream breaker and another 60-80 ms per layer after that. This is extremely fast. Most older engineering recommends at least 0.2-0.3 seconds between layers. With ZSI we only wait 1 cycle or 16 ms between layers. Arc flash is linear with time. So even with just one upstream breaker it can clear an arc flash in 5 cycles compared to 10 cycles, cutting incident energy in half. If we are using older slow breakers or inverse time curves and cutting from 0.24 seconds to 0.08 seconds it’s 1/3rd and this is being conservative. Generally speaking it’s much better than this.

The inherent problem with ZSI is implementation. Need all breakers to be same manufacturer and need to run the ZSI wiring. It’s not very practical for most situations. You won’t find ZSI built into say ILine breakers especially smaller sizes or most breakers used in MCCs for instance but ZSI is no discriminatory..,it does not work if even one breaker does not support it. A maintenance switch gives you better performance (the one cycle delay is eliminated), costs less, and requires none of the things ZSI requires. You can put one for instance on a feeder breaker but keep existing panel boards and MCCs. The trade off is that it’s not automatic. I’m a service contractor for a very large area. To date I have only read about ZSI in books and articles. In the hundreds of plants I’ve been in, none of them have ZSI or they abandoned it. On the other hand maintenance switches are becoming so common they are offered as standard on many trip units.