Eaton Bussman fault current app

[Dsg319]

In the below image that I have circled which should be chosen when finding available fault current? I know the -10%z tolerance will give the high value as in worst case scenario. But Isthere a way to know which tolerance to choose on a specific transformer?

 

[Carultch]

Transformer impedance is something that the manufacturer cannot know until after it is manufactured and tested. They can build to a certain specification for how to arrange the winding and core geometry, but the impedance will inevitably vary from the intended value. You specify a nominal value to the manufacturer, and they build within a certain tolerance of that value. The transformer manufacturer likely has data on what their track record on the tolerance on impedance would be, which could be the input to this calculation.

 

[jim dungar]

Rarely are impedances actually determined when the transformer is less than 500kVA unless it is a special design.

Most transformer impedance will fall into a typical design value which can be obtained from manufacturers literature.
Dry type transformers typically have more ranges, such as <3, 5-75, 150-300kVA, than do liquid, like oil, filled units. Utility transformers often have impedances which are really low because a utility doesn't like paying for the higher losses often associated with higher impedances.

 

[Dsg319]

Transformer impedance is something that the manufacturer cannot know until after it is manufactured and tested. They can build to a certain specification for how to arrange the winding and core geometry, but the impedance will inevitably vary from the intended value. You specify a nominal value to the manufacturer, and they build within a certain tolerance of that value. The transformer manufacturer likely has data on what their track record on the tolerance on impedance would be, which could be the input to this calculation.

So do transformers typically have tolerance labeled somewhere on them? Or is that something you would have to contact the manufacture about like you said?.....I’d imagine the best thing to do when figuring available fault current would be to use the -10%max fault being that it will give the highest value.(highest available fault current)

 

[Dzboyce]

My local power utility publishes this list of transformer available fault currents. That way we don't have to call them every time we do the calculations

 

[Dsg319]

My local power utility publishes this list of transformer available fault currents. That way we don't have to call them every time we do the calculations

That’s nice!

 

[ron]

The ANSI standard for impedance tolerance is +/- 7.5%, so I typically choose the -10% max fault in that calculator, or I calculate by had what -7.5% would be and select 0% change and enter the value with -7.5% already.

 

[infinity]

The ANSI standard for impedance tolerance is +/- 7.5%, so I typically choose the -10% max fault in that calculator, or I calculate by had what -7.5% would be and select 0% change and enter the value with -7.5% already.

Would someone care to explain what this means to the non-engineers in the room?

 

[Carultch]

Would someone care to explain what this means to the non-engineers in the room?

I'll leave it to the reader to look up the formal definition of transformer impedance. For now, it is simply a performance parameter of the transformer.

Suppose the impedance you specify is 5%. This is the manufacturer's goal for the impedance they build, according to how you specified it. This is the nominal value. Just as when you tell a machinist to cut a bar to 10 centimeters, the nominal dimension is 10 cm, but it may not be exactly 10 cm when you receive it. Given a tolerance is +/- 1 mm, it may be as short as 9.9 cm or as long as 10.1 cm. A +/-7.5% tolerance on this impedance, means impedance could be as high as (1 + 0.075)*5%, or it could be as low as (1 - 0.075)*5%

Translate 7.5% from a percent to a decimal, and get 0.075.
Add it to 1, and multiply that quantity by the nominal 5%. That's the upper limit on expected impedance of the transformer.
Subtract 0.075 from 1, and multiply that quantity by the nominal 5%. That's the lower limit on expected impedance of the transformer.

So 5% +/- 7.5% could be as low as 4.625%, and as high as 5.375%.

The tolerance of +/-7.5% is not an incremental change from nominal 5%, but rather a multiplicative change from 5%. Incremental changes in percents, are called percentage points. If it were an incremental change, then nominal 5% minus 7.5 percentage points, would leave a possibility of negative impedance, which isn't possible.

 

[WasGSOHM]

And if your meter has an accuracy of +/- 1 percent, it may mean that 95% or 99% of the time the readings will be within 1% of the true value.
In principle, at least. . .

 

[wbdvt]

My local power utility publishes this list of transformer available fault currents. That way we don't have to call them every time we do the calculations

Please be aware that the values provided by the utility in that table are infinite bus fault currents not available fault currents. The infinite bus are fine for determining what AIC and SCCR ratings are needed for equipment but should not be used for arc flash calculations.

 

[Carultch]

Please be aware that the values provided by the utility in that table are infinite bus fault currents not available fault currents. The infinite bus are fine for determining what AIC and SCCR ratings are needed for equipment but should not be used for arc flash calculations.

It's my understanding that they *could* be used for arc flash calculations or any calculation that depends on the available fault current, but it is overkill to do so, and may result in excessively high outputs of the arc flash calculations that don't reflect reality. "Infinite bus" refers to assuming infinite available fault current on the primary side of the transformer, as a worst case scenario value that sets a boundary on the available fault current at the transformer secondary.

 

[wbdvt]

IEEE 1584 requires the use of available short circuit current for incident energy calculations as it has been shown many times that use of infinite bus short circuit currents to determine incident energy values result in erroneous values. In some cases it can lead to lesser incident energy values than actual values, resulting in wearing arc rated PPE that does not provide the proper protection. In other cases, it can lead to higher incident energy values than actual values resulting in wearing arc rated PPE that is more than needed. So you could be under protected or over protected.

Why take the chance with safety? Use the correct values for what the situation is.

 

[Carultch]

IEEE 1584 requires the use of available short circuit current for incident energy calculations as it has been shown many times that use of infinite bus short circuit currents to determine incident energy values result in erroneous values. In some cases it can lead to lesser incident energy values than actual values, resulting in wearing arc rated PPE that does not provide the proper protection. In other cases, it can lead to higher incident energy values than actual values resulting in wearing arc rated PPE that is more than needed. So you could be under protected or over protected.

Why take the chance with safety? Use the correct values for what the situation is.

The utility might just give you the infinite bus value anyway. I've had experience asking for the available fault current, and the respondent just told me the specs of the transformer and what I could infer from them.

I'm confused as to how using the infinite bus value could result in lesser incident energy values. Do you have an example of how this would work?

 

[jim dungar]

I'm confused as to how using the infinite bus value could result in lesser incident energy values. Do you have an example of how this would work?

Higher levels of fault current could be within the range of a protective device operating in its instantaneous region. The short clearing time may result in a low incident energy level. A low fault current may result in the protective device operating relatively slowly with a resultant high incident energy.

This is one of the reasons why IEEE recommends using the actual faulty current.

 

[wbdvt]

The utility might just give you the infinite bus value anyway. I've had experience asking for the available fault current, and the respondent just told me the specs of the transformer and what I could infer from them.

I have experienced this also as typically the first person you talk to is a Customer Service Rep and they just have infinite bus tables. You have to explain what and why you need and ask for an engineer. I have had one case where I had to get the state PUC involved as the utility gave ridiculous reasons on why they could not provide the information such as it was proprietary information. After educating the PUC on this issue, the PUC directed the utility to provide the information that I was requesting.

Bottom line is that you have to be persistent to get the proper information especially where safety is concerned.

 

[don_resqcapt19]

Bottom line is that you have to be persistent to get the proper information especially where safety is concerned.

Because of potential liability to the utility when their numbers are used in 70E calculations, they really need to give you the complete dynamic range of what their system can supply and the engineer must then make the safety related calculations based on the worst case. The electrical distribution systems are dynamic as their configurations change for various reasons. Some of these configurations result in higher available fault current and some in lower.

 

[paulengr]

It's my understanding that they *could* be used for arc flash calculations or any calculation that depends on the available fault current, but it is overkill to do so, and may result in excessively high outputs of the arc flash calculations that don't reflect reality. "Infinite bus" refers to assuming infinite available fault current on the primary side of the transformer, as a worst case scenario value that sets a boundary on the available fault current at the transformer secondary.

Nope.

It turns out to do the opposite in most cases.

Breakers and fuses at the distribution level at “typical” arcing currents are inverse time. So if the actual current is say half of what we expect the time increases exponentially much longer than just doubling.

Arc energy which is what arc flash calculations predict is a product of arc power and time. The actual arc power calculation is complicated but close to what you expect...half the current will be approximately half the arc power.

But because time is far more than doubling before the breaker/fuse trips the total arc energy will be much more even if current is half of what we expect.

With short circuits if we are below the rated short circuit of the equipment (SCCR) we don’t care, so using a theoretical number that may be too large is harmless. With arc flash quite the opposite happens.

 

[paulengr]

Transformer actual %Z may deviate a lot from name plate on smaller dry types. For instance GE doesn’t make their own transformers. They put out a spec and then various lesser known companies bid on say making all the dry transformers of a certain size for GE for 12 months. On hand wound transformers doing tests like a TTR is pretty standard and it may be marked say 1. and the decimal is stamped when the testing department tests it to make it say 1.7%. On machine wound transformers they all get marked the same number and any given transformer might be a little off.

Even then the ANSI transformer standard essentially has a fixed %Z for all oil filled transformers with the same voltages regardless of size. It is very hard to meet the spec on smaller (under 500 kVA) and larger (over 5-10 MVA) transformers. Utilities often just use the ANSI number as an approximation and in their specs.

I know this because I was the engineer for a large mine with hundreds of transformers. I often had to get quotes and frequently the transformer manufacturer would throw up red flags about %Z specs. The closer you are to a common design the more likely they are to get pretty close but it was pretty common to order say 5.5%Z and get one marked say 4.2%Z. We were usually low because the transformers were moved a lot so we ordered them with copper coils instead of aluminum because they held up better.

I have tried purposely ordering say 10%Z for arc flash purposes. What I found out quickly is that %Z isn’t that easily controlled.