Is there a standard definition of "minimum available fault current"?

[constantlylearning]

I’m embarrassed to ask this, but can anyone tell me if there is a definitive guide to what is meant by “minimum available fault current” on transmission systems? Traditionally, when I’ve been asked for this and I do not get more information from the requesting party, I give the fault current/system impedance based on the following conditions:

[ul]
[li]All generation local to the bus in question is turned off.[/li]
[li]The strongest line source OR strongest transformer source is out of service.[/li]
[li]I’ve seen some suggestions that a fault impedance should be included. However, I feel that this is not correct. The requesting party can take the system impedance data and model and add fault impence if they choose.[/li]
[/ul]

For “maximum available fault current”, I always turn all generation on and keep all lines and transformers in service (that would normally be in service).

Finally, I think it is always a good idea (though I’ve not always followed the practice) to give the requesting party system impedances rather than fault currents. Of course, with one, you can derive the other. But, if they can’t make the conversion, they probably shouldn’t be applying the data.

 

[cranky108]

When I get this question, my answer is zero.
It's a dumb question because it does not convey what is exactly expected.

On every fault current question, I automaticaly ask what do you want it for. If they don't know, then I can't give them an answer.
The problem is people who use computer forms just ask the question for fault current without understanding what is really needed or what is being calculated.
Someone created these forms, but never took the time to write a discussion of what the program is calculating, assuming, or needing. Insert value here.

Some consultants will send a form asking for all sorts of things. And they are mostly things they don't need. I just throw the forms away, and call them to ask what are you using it for.
They will waste your time, but they don't care.

Push back and call them and ask what they need it for. Take the time to understand why they need it.

Strange how much we try to not talk to each other, despite the fact there are more phones in this country than people.

 

[constantlylearning]

Cranky, I get what you are saying and I agree. Having said that, usually they are asking for the purpose of an arc-flash study. There are a variety of system conditions ranging from mass outages due to storms all the way up to N-1 contingencies. Of course each can give varying levels of fault current.

So, if someone wanted to do an arc-flash study, what system condition would you give them? Incidentally, these are transmission-level impedances, so they wind up having a relatively small impact on fault current when compared to the transformer impedance. But, I like to be as precise as I can.

I really do appreciate your candor though. On a side note, you wouldn't by chance be a fan of John C. Dvorak would you?

 

[dpc]

If you are talking about utility source fault data at the transmission level, I would give them the maximum fault current and the fault current with the largest single source out of service. If this is a distribution feeder, there generally is not enough difference to matter, so I don't bother.

 

[cranky108]

Who is John C. Dvorak? After a day in the office, my eyes are tired.

If they want a fault study for Arc-Flash, then they should state that. If they want a motor start study, which usually includes a minimum credidble system, then state that.If it's for a ground grid study, so on.

I would think for an Arc-Flash study you would want to know the system maximum to set the upper limits of the study. But I have so few transmission study requests.

I think the last one was for pipeline cathotic study.

For distribution, we always say here's the transformer typical impedance for this size, it is suggested you use inifite bus for the high side. But you can't always tell how much the customer wants to shave off the fault current to justify a lower cost gear. It seems foolish to possibily undersize equipment which can hurt people. Makes personal injory lawers wealthy though.

 

[constantlylearning]

I am not an arc-flash expert, but I believe the philosophy is that a lower system impedance will result in a lower fault current. A lower fault current will take longer to clear, as it is closer to the Y-axis on the Time/OC curve. The idea is that a longer exposure to lower-current arc-flash is worse than a shorter exposure to high-current arc-flash. Again, I am not an expert in this area.

Side note: John C. Dvorak is an old-school tech journalist who started his career as a chemist. He's sort of know as a curmudgeon and he has a very dry wit. Your comments, "When I get this question, my answer is zero. It's a dumb question because it does not convey what is exactly expected.", made me think of him. He's one of the few tech-journalists that I find credible and not a complete idiot.

 

[dpc]

It has to be calculated - you can't automatically conclude that the maximum (or minimum) fault current will result in higher incident energy levels. So multiple scenarios must be considered. That is why the question is being asked.

 

[constantlylearning]

Yes, I agree. I should have said "COULD" be worse. But, back to the original question. Any idea what the "standard method" amongst utilities would be for specifying minimal? I'm not aware of any guide or standard that spells this out.

 

[constantlylearning]

DPC, I do note where you speak of the largest single source being out. That does seem like a good idea, but are you aware of any guide or standard that spells this out?

 

[dpc]

There's no standard practice that I'm aware of. Arc-flash operating scenarios are generally based on single contingencies (if that), so they are looking for the situations that could reasonably be expected, not the End of Days (or as we call it around here, the Cascadia Subduction Zone Event). As a point of comparison, arc-flash calculations assume all breakers will operate properly and make no allowance for breaker failure.

In most situations, the system on the secondary side of their main transformer is not going to be greatly impacted by changes in the utility fault duty. The buffering impedance of the transformer dominates.

 

[cranky108]

Because most transmission is open air, or in cable or iso-bus, does this come up very much? Arc-flash is compounded if it is in a box, and normally is taken as three phase.
So in transmission there usually isen't much compounding and three phase events.

So where is this a concern? I know there is always a concern, but it is much reduced on the transmission system.

Or are you looking at the sub-transmission system?

 

[constantlylearning]

Cranky, it varies per request. Usually, the customer who is requesting the data is doing so for use inside of their facilities, on the LV side of the substation transformer. We get asked about system impedance on the transmission bus. So, I'm assuming the customer is adding the step-down transformer impedance, along with the impedance of any transformers they have. In transmission, we will sometimes model the substation transformer to verify that protective relaying can't "see" through the transformer, but we do not provide LV fault currents or impedances, except in unusual situations.

Thank you everyone for a great discussion. It's always good to get feedback from fellow industry professionals.

 

[7anoter4]

We are following IEC 60909, usually. In IEC 60909-0 1.1 "Scope" it is stated:
“In general, two short-circuit currents, which differ in their magnitude, are to be calculated:
-the maximum short-circuit current which determines the capacity or rating of electrical equipment; and
-the minimum short-circuit current which can be a basis, for example, for the selection of fuses, for the setting of protective devices, and for checking the run-up of motors.
However, as shown in ch. 2.2 "Calculation assumptions” some of the parameters are neglected as for instance: d) Arc resistances are not taken into account.
Nevertheless it is mentioned further:
"Despite these assumptions being not strictly true for the power systems considered, the result of
the calculation does fulfil the objective to give results which are generally of acceptable accuracy."
In ch. 2.5 "Minimum short-circuit currents" there are some "conditions" also:
Voltage factor will be cmin, choose the system configuration and the minimum contribution from power stations, motors shall be neglected, resistances RL of lines shall be introduced at a higher temperature.

 

[tinfoil]

I always ask what they want the data for.

If it is for an arc-hazard study, then I just give them the max available and tell them to simulate different fault impedances, from zero to a large value.
This will take them through the same range of total impedances at the point of the arc.
That way, they still have to work out the worst case I^2*t energy transfer, and my utility cannot be held liable if they use any minimum number I give them improperly.

If they are doing motor start, then I zero out any local generation and assume worst-case LIKELY scenario for lines out of service for their bus of concern.

 

[cranky108]

-the minimum short-circuit current -- and --minimum available fault current -- are not the same thing, at least in my mind. And the IEC what ever is never stated as the basis for the request.

The problem is the request isen't stated clearly, so I can't be sure what is being asked. It's common to have the same issue for internal requests.

It's always better to ask and clarify what the study is being used for. Because running a study usually dosen't take that long, but running many studies because the customer isen't clear about what they want is a waste of company resources (and my time).

It's like writing a letter, get to the point, and state what you want clearly.

 

[unclebob]

We put much work in modeling power systems, with cable lenghts, ampacities, breakers models, settings and whatnot... and the most important value of all, the fault current is the one that is that most difficult to obtain. You end up doing 10 scenarios from 10 kA to 35 kA to get that worst case.

 

[constantlylearning]

Thank you all for your input. We've decided that whenever possible we will give the requestor a network equivalent reduction model at the delivery point, stating what the conditions of the model represent. We still ask what they are using the model for, but often they aren't able to give a complete answer. In this way, they can use the model in as many scenarios as they fit.

Even if they do not have the software to use the model, the Thevenin impedance of the system should be enough to allow them to use the data as they need.

 

[Skythian]

In my understanding, minimum available fault current is the amount of short circuit current coming from service entrance only while all the other sources and/or inductive loads are turned off. Alternatively, maximum available fault current assumes all motor loads and generators are running and contributing to the fault.
Screen shot from Short Circuit Analytic software below gives an example of different type fault currents including minimum, maximum, three-phase, phase-to-phase etc. that can be derived from comprehensive short circuit analysis:

The maximum available fault currents would be used for calculating total arcing current for incident energy and arc flash boundary calculations. The minimum available fault current would be used for calculating arcing current through the protection device and determining arc duration as a function of the device time-current characteristics. Screen shot from Arc Flash Analytic software below exemplifies an application of both the maximum and minimum fault currents (Available 3-ph Short Circuit Current (ASCC) and Part of the ASCC thru Protection Device respectively)

 

[tinfoil]

To reiterate:

If the purpose of requesting the minimum available fault current is for arc-hazard analysis (and nowadays, that is almost every request for me), our company policy is to NOT PROVIDE IT!

There is a significant block of 'consultants' that are just plugging in numbers to get 'an answer', and we fear the measures they might take as a result. The actual arcing current at the time of an incident can depend on a lot of factors. The actual fault impedance has a tremendous impact on this current.

I am aware of a case where the 'consultant' took our minimum number, simply neglected the possibility of a non-zero fault impedance at the point of analysis, and did their analysis. But the consultant did not actually find the 'worst-case' for total delivered energy (which was at a lower current and a longer event duration because of the specific time-current curve of their breaker relays).

My utility does not want to be liable for such poor analyses, nor do we want to take on any burden or responsibilities to review the work of others. So, we don't give out 'minimums'.


If you are asking for other purposes, such as voltage flicker / motor starting or Distributed Generation, that's different.

 

[constantlylearning]

This has been an excellent discussion. I tend to agree with tinfoil's response. It doesn't make sense for a utility to specify a minimum, as it truly can be anywhere between the maximum and zero, depending on circumstances. Thank you all for your input.