High impedance faults

[eng.dork]

Does anyone know much about high impedance faults? From my understanding they are very hard to detect because of the low current that occupanies them unlike low impedance faults where the current is very high. Wait...is that right? I overheard my boss talking the other day about how much he hates high impedance grounding...why is this a problem? I would think that having this type grounding system that functions correctly would be benefical.

 

[Dark Knight]

Recently we had a fault on a switch where one phase stayed open while the other two were close. It was "wicked" scary. Explosion and pieces of insulators flying evrywhere. The open phase was a high resistance path. My boss had me calculating the effects of that kind of fault. First we calculate the pre-fault conditions and then the voltages during the fault.

I mention this, and I don't know if that will answer your question, but during that fault the voltages will be reduced. By Ohm's Law if the resistance is high, in order to keep the voltage level the current will rise as crazy. That is the reason why most equipment get fried during that kind of situation(high resistance faults)

It is very interesting. We had half voltage on the low side of two phases and full voltage in one phase. Costumers on the full voltage phase: Never saw it. Costumers on the other two phases and three phase costumers: Not happy.

If someone wants to see the calculations just PM me. I will scan my sheet and send it to you.

About the benefits of highh impedance grounding, and shooting out of my pockets, the situation is that there are places were we have to detect zero sequence components and then is when the HIG comes to benefit. I will dig some more on that. Stay tuned.

 

[grover]

Low impedance faults will cause a LOT of current, and will trip the breaker. High-impedance faults can still release a lot of energy, but if the impedance is high enough, it's won't trip the breaker, and the fault will continue unabated until something catestrophic happens. Proper calibration and careful setting of overcurrent devices is important in limiting the danger of high-impedance faults; EG, if you have a 1600A switchboard with 200A of load, set the OCP at 300A, not 1600A. Otherwise, you could have a 1400A fault, and never know it until the fire dept arrives.

For your second question, instead of solidly bonding the neutral to ground, it can be high-impedance bonded, where the neutral is still connected to ground and under normal conditions will maintain a potential very close to 0 volts, but under ground fault conditions, the high impedance reduces fault the maximum potential fault current.

 

[trainee]

Low impedance faults will cause a LOT of current, and will trip the breaker. High-impedance faults can still release a lot of energy, but if the impedance is high enough, it's won't trip the breaker, and the fault will continue unabated until something catestrophic happens. Proper calibration and careful setting of overcurrent devices is important in limiting the danger of high-impedance faults; EG, if you have a 1600A switchboard with 200A of load, set the OCP at 300A, not 1600A. Otherwise, you could have a 1400A fault, and never know it until the fire dept arrives.
For your second question, instead of solidly bonding the neutral to ground, it can be high-impedance bonded, where the neutral is still connected to ground and under normal conditions will maintain a potential very close to 0 volts, but under ground fault conditions, the high impedance reduces fault the maximum potential fault current.

I echo the first part of your statement but can you please explain your example in more detail. What is OCP? and why is it set at 300A? What risk of setting this OCP would there be to open cct the faulted phase?

Thanks.

 

[mudpuppy]

I echo the first part of your statement but can you please explain your example in more detail. What is OCP? and why is it set at 300A? What risk of setting this OCP would there be to open cct the faulted phase?

OCP is short for overcurrent protection. What he is saying is that it is better to make your protection as sensitive as possible in order to avoid damage. If you know the highest load current in the circuit is going to be 200 A, then set it somewhat above the load, rather than based on the maximum load the circuit is capable of carrying (2000 A in this case).

Now, if I may get on my soapbox. The concept Grover is describing is widely viewed as appropriate--it prevents damage and possible injury due to impedance faults, which may have a current slightly above load current. But, along comes NERC (the North American Electric Reliablity Corporation) with the results of their investigation of the blackout of 2003 (recommendation 8a), saying that this concept should *not* be used on transmission lines. Their rationale is that maybe some of the lines that tripped during this catatrophic failure would not have tripped if their protection had been set much less conservatively. And if fewer lines had tripped, fewer people would have been blacked out. In the example above, NERC's recommendation would be to set the protection to trip at 3000 A, (rather than 300 A or even 2000 A) as long as this setting would trip for a bolted fault.

 

[Flyer_PE]

Now, if I may get on my soapbox. The concept Grover is describing is widely viewed as appropriate--it prevents damage and possible injury due to impedance faults, which may have a current slightly above load current. But, along comes NERC (the North American Electric Reliablity Corporation) with the results of their investigation of the blackout of 2003 (recommendation 8a), saying that this concept should *not* be used on transmission lines. Their rationale is that maybe some of the lines that tripped during this catatrophic failure would not have tripped if their protection had been set much less conservatively. And if fewer lines had tripped, fewer people would have been blacked out. In the example above, NERC's recommendation would be to set the protection to trip at 3000 A, (rather than 300 A or even 2000 A) as long as this setting would trip for a bolted fault.

Just to add my $.02. The appropriate setting is dependent upon your objective. There is a compromise that must be reached between safety and reliability. Which way you bias the settings will depend on the individual situation. For some motor operated valves in nuclear facilities, accident conditions will cause the thermal overloads to be automatically bypassed. The thought being under those conditions, the motor is expendable so long as it puts the valve in the correct position to mitigate the accident.

Our company has been doing a lot of arc flash analysis lately. Several of our clients are opting for dual settings for their protective relays on the medium and low voltage switchgear. The "normal" setting is biased toward reliability and maintains normal coordination with the other protective devices. The "maintenance" position lowers the settings to reduce the duration of an arcing fault. The down side of the maintenance mode is that you are not maintaining coordination and a downstream short may very well take out the whole bus rather than just the branch feeder.

Jim

 

[trainee]

interesting...yeah, arc flash is a whole different world that probably deserves its down thread. I've seen a little device that can throw the relays into "maintenance mode"...but like you said, its only when you have to open a switchgear.

With respect to OCP and relays and settings....lately, I've been seeing a lot of "time dial" and "multiplier".

What do those mean?

Thanks.

 

[Dark Knight]

interesting...yeah, arc flash is a whole different world that probably deserves its down thread. I've seen a little device that can throw the relays into "maintenance mode"...but like you said, its only when you have to open a switchgear. With respect to OCP and relays and settings....lately, I've been seeing a lot of "time dial" and "multiplier".

What do those mean?

Thanks.

Some overcurrent relays have a time dial function. You will recognize those on the diagrams because they use the 50/51 designation. 50 means instantaneous overcurrent protection and 51 is time delayed overcurrent protection. What this does is allows some time, very short time, to other protection elements to clear the fault before sending a trip command. The multiplier, and if I did not missunderstand your question, is a magnitude that help you to select the current level you want your relay to operate.

For example...

Your pick up current is 5 Amps so at that value your relay will start to work but will have a time delay.(remember that this current is provided to the relay by Current CTs.) Your multiplier is 2, or 2X, so that means that when your current reach 10 Amps your relay will operate at a specific time. It will not be the same speed or time of operation is you have, lets say, 4 times tap or 4X.

These times can be calculated using a formula and depending on the type of curve the relay uses (Inverse or Very Inverse for example)

Hope I made any sense.

 

[mudpuppy]

The appropriate setting is dependent upon your objective. There is a compromise that must be reached between safety and reliability.

Good point. I don't disagree with that at all.

 

[Dark Knight]

Found this paper and it is related to the topic.

Enjoy it.

High Impedance Detection