Source Impedance and AFC

[wwhitney]

A few questions:

1) Suppose an engineer tells me that at a certain point in the electrical system, the calculated Available Fault Current on a 240V 2-wire circuit is 12 kA, and that calculation is accurate. Does that mean that it makes sense to say the source impedance is 240V / 12kA = 20 milliohms?

2) By which I mean that if you take, say, a 100A load and connect it at that point, the upstream voltage drop will be 100A * 20 milliohms = 2V? Likewise for any current from 0A to 12 kA?

3) Then if so, is the converse done in practice? I.e. if you need to know the AFC at a certain point in the circuit, you could load the circuit by 100A or 300A or whatever, measure the resulting upstream voltage drop, compute the source impedance, and then compute the AFC?

Thanks,
Wayne

 

[electrofelon]

A few questions:

1) Suppose an engineer tells me that at a certain point in the electrical system, the calculated Available Fault Current on a 240V 2-wire circuit is 12 kA, and that calculation is accurate. Does that mean that it makes sense to say the source impedance is 240V / 12kA = 20 milliohms?

2) By which I mean that if you take, say, a 100A load and connect it at that point, the upstream voltage drop will be 100A * 20 milliohms = 2V? Likewise for any current from 0A to 12 kA?

3) Then if so, is the converse done in practice? I.e. if you need to know the AFC at a certain point in the circuit, you could load the circuit by 100A or 300A or whatever, measure the resulting upstream voltage drop, compute the source impedance, and then compute the AFC?

Thanks,
Wayne

A very interesting question. I am going to guess that overall impedance is pretty messy and not linear? Just a WAG, I'd be curious what the EE's have to say.

 

[charlie b]

You have lumped the impedance of the source with that of the circuit wires. That’s OK, and it's really not messy. But yes, your analysis is valid. Do we do that in practice? No. We model the circuit, inserting any impedances we know, including that of the service transformer, and let the computer do the rest.

 

[Julius Right]

I don't think the source impedance could play any role either in the short circuit calculation or the voltage drop, and that is for such a low voltage. The utility must supply the voltage within the limits per ANSI C84.1-2020. From this limits you can judge the voltage drop.
On the other hand, for short-circuit currents, the short-circuit power and the R/X ratio are all you need to continue the calculations.

 

[kwired]

You have lumped the impedance of the source with that of the circuit wires. That’s OK, and it's really not messy. But yes, your analysis is valid. Do we do that in practice? No. We model the circuit, inserting any impedances we know, including that of the service transformer, and let the computer do the rest.

But a length of conductor will have a rather fixed impedance. Impedance of the source changes regardless when you replace the transformer with one with different design details. Doing that will change the available fault current at the source terminals but may not have all that significant of a change at the other end of the conductors depending on size and length of the conductors.

OP suggested 100 amp load for test current. Assuming infinite supply ability you still should have varying impedance across the conductor depending on size and length of the conductor as well.

 

[jim dungar]

A few questions:

1) Suppose an engineer tells me that at a certain point in the electrical system, the calculated Available Fault Current on a 240V 2-wire circuit is 12 kA, and that calculation is accurate. Does that mean that it makes sense to say the source impedance is 240V / 12kA = 20 milliohms?

2) By which I mean that if you take, say, a 100A load and connect it at that point, the upstream voltage drop will be 100A * 20 milliohms = 2V? Likewise for any current from 0A to 12 kA?

3) Then if so, is the converse done in practice? I.e. if you need to know the AFC at a certain point in the circuit, you could load the circuit by 100A or 300A or whatever, measure the resulting upstream voltage drop, compute the source impedance, and then compute the AFC?

Thanks,
Wayne

The voltage drop concept is theoretical but not practically correct.

Impedance is complicated and should be broken into its resistive and reactive components before being combined with other impedances. Most field millivolt measurements are looking a resistance across contact surfaces, they are rarely done to measure impedance from a source.

The value given to you by an engineer may be a partially made up number rather than something you can actually measure. Rounding up is very common, either to reach a typical value, like 10kA or 22kA, or to eliminate insignificant digits.

 

[wwhitney]

You have lumped the impedance of the source with that of the circuit wires. That’s OK, and it's really not messy.

Yes, I guess in (1) I should have said something like "upstream" impedance, which would be the sum of the source(s) internal impedance(s) and the wire impedances.

Thanks, Wayne

 

[wwhitney]

The voltage drop concept is theoretical but not practically correct.

Good, I like to start with theoretically correct, and I'm at least aware of the practical issues you mentioned. Probably there are other practical issues I'm not aware of.

But this (complex number) linear relationship between current drawn and upstream voltage drop observed can't really hold all the way from 0A to the AFC, can it? I thought "core saturation" for transformers is one of these practical issues, and that once you enter that regime, the transformer's internal impedance would effectively change.

Thanks,
Wayne

 

[electrofelon]

Good, I like to start with theoretically correct, and I'm at least aware of the practical issues you mentioned. Probably there are other practical issues I'm not aware of.

But this (complex number) linear relationship between current drawn and upstream voltage drop observed can't really hold all the way from 0A to the AFC, can it? I thought "core saturation" for transformers is one of these practical issues, and that once you enter that regime, the transformer's internal impedance would effectively change.

Thanks,
Wayne

I also was thinking about an upstream transformer. In particular I thought about the process of measuring transformer impedance, where the secondary is shorted and the voltage is raised until rated current is flowing. I would like to see a voltage vs current graph of that. Would it be linear?