System study information from utility

[JBD]

What is the sense in purchasing and using power system analysis software, like SKM, if utilities will not provide accurate "as currently installed" information?

Yes I understand that at some future date they may upgrade the transformer and other equipment feeding a facility, but how about today's data. Yes their grids are very dynamic with some clearing times, Z, and X changing constantly, but they must use some "fixed" values to set their own relays and fusing. If they feed a location with a dedicated transformer, can't they be more definitive with the values at the secondary of the transformer?

Maximum design (future) fault currents are absolutely correct for choosing equipment with the proper fault ratings. But maximum ratings may not provide the worst case incident energy for arc-flash. Even device coordination may be easier to accomplish if some fault levels are unattainable. Fault clearing times are not as important but if I have to tell the utility the TCC of my equipment why can't they tell me any thing about theirs?

In a recent situation a facility asked for an 800A 208Y/120V 3phase 4wire service. The utility said it had to be designed for a potential fault current of 57kA based on the maximum potential transformer that might ever be installed. The actual utility installation is a dedicted 75kVA padmount based on the estimated load profile of the service. The utility refuses to provide %Z, X/R or fault data for the installed transformer.

 

[MJJBEE]

If you are planning to do an Arc Flash study you may be able to get more realistic data by explaining to the utility engineer the problems with using false values for arc flash studies. I know my company will provide fault data if you make it clear they will only be used for analysis of the existing system not design.

 

[davidbeach]

If you want good data from the utility, take service at the distribution voltage. After doing a few of those you will have a good feel for the general stiffness of the utility's distribution circuits.

The utility is placed in a difficult position if they provide a prospective fault value that they can ever exceed; who becomes liable if equipment is installed that meets the originally available fault current and is then damaged during a later fault with higher fault currents. Even with notice to the customer, there would be problems; would you want to get a notice from the utility that they were going to increase the fault current and you had to replace $1000s worth of service and distribution equipment?

Once you have set equipment ratings based on the supplied utility information, you are free to use your engineering judgment to run other studies with other source impedances.

It can be a definite frustration, but it's how the game has to be played.

 

[JBD]

MJJBEE, All well and good, except utility will not provide any data except for maximum design fault levels. This data is all but meaningless except to purchase correctly rated equipment.

David, no one is asking for fault levels that cannot be exceeded. What I want is fault levels and/or source impedances based on what the utility actually installed. As far as system changes affecting (negating) studies, there is a greater probabilty the customer will make a change to their own system, than the utility changing out a dedicated transformer feeding the customer.

I have no problem with maximum design levels for equipment selection. I have no problems with not being given clearing/operating times. But I need some realistic fault currents from them. I also understand not working service equipment live, but what about downstream from that point?

The attitude from several utilities is: Yes arc-flash is a hazard and yes maximum fault values are useless for calculating incident energy but we are not going to provide any data, use the tables in NFPA70E instead. Great, one thing the tables in NFPA70E require is "available short circuit current".

If no one has a solution to getting installed data from the utility does anyone have a better defense, when a worker is injured, than "we did the best we could with data we knew was bad".

 

[DanDel]

I agree. I have had similar problems with the utilities in my area, over the past few years, and they are getting worse.
Another real problem with maximum fault conditions is when you need real values to run a load flow or motor start study. In this case, the minimum contribution values are necessary for a conservative result, not the maximum.
I have found that the utility will not release this data because of liability, and/or because the facility is not a primary customer. But when I ask for values for a primary customer, I only get values from the utility bus upstream from my customer, and no idea about the transmission/distribution connections in between.
I have tried to explain to the utility that the issue of an actual calculated value is necessary and can be released with any number of disclaimers, but to no avail.
I have had the most success when the client requests the information from the utility directly. In fact, the utility will not release data to us directly anymore anyway.

 

[davidbeach]

JBD, I don't disagree with your points; I've been there and done that and wished I had much better data. How about doing a staged fault test? ;-)

I worked on one industrial remodel project where I started out using assumptions of the as installed service condition and discovered some equipment that was over-dutied (from the original installation). Ultimately the only information the City would accept and that the utility would provide was based on ultimate worst case; which pushed the equipment replacement much further into the system than the problem actually existed. This added 10s of $1000s to the cost of the project, but not a thing I could do about it.

Try this, use something like a 200MVA source with an x/r of 8 ahead of a transformer with an impedance of 5.32% (low end of the ANSI standard range) and model the actual installation using that. Connect that source in parallel with the source the utility defines and use one or the other depending on what studies you are running. If you want to be really diligent, start at 250MVA and then go down in 25MVA steps and use the source that gives you the worst arc-flash results to print the labels.

 

[jghrist]

If the legal department of the utility was made aware that designing to the maximum fault levels provided by the utility could result in arc-flash injuries, they may change the utility policy.

 

[MJJBEE]

Exactly my point. The utility I work for will give real values only if you promise (in Writing) not to design anything with those values. For design you must use the maximum values.

 

[resqcapt19]

I think as more and more people start to use 70E, that this problem will only get much worse. Most utility systems are too dynamic to provide a fault value that is suitable for use in arc flash calculations. I expect that the leagal departments of more and more utilities will advise their engineering staffs to provide only maximum fault current with a disclaimer that these currents are only for equipment selection and not for use in arc flash calculations.
Don

 

[rbulsara]

I am not sure I can agree with all concerns of JBD.

If you installed a 800A main board, you can not argue that you will not be using it to its capactiy. 75kVA tx is good for only 208A at 208V. And utility co. is right in telling you maximum current they can provide. Yes the, equipment you install must meet those requirement. 57kA is not excessive at all. If you ask the same question in a large city, they will tell you 200kA min at 208V (because of spot networks).

I am not sure what is the point of knowing actual fault levels today, when it should be desgined for maximum available within reasons. Plus before you go mad at rest of the world, do the calcs using 57kA and see how bad it could be. And if you have enough experience, you need rely on them to assume reasonable impedance of a 75 kVA transformers. Look up a few manufactures data and pick the lowest value..like 3% and make a note on you study to that effect to cover you liability.

As for the settings of upstream devices I have always been suscessful in gettinng accruate data from the utilties and in one case convinced them to change their fusing to meet the requiremts once we shared the information.

 

[JBD]

Don,

Are you saying arc-flash should not be considered on systems connected to utility grids? Are you hinting that this section of 70E is not needed?

Maybe what I learned 30 years ago is no longer valid, but I believe that utilities can provide realistic ranges of fault currents on the secondary side of "customer" transformers. Utility fault clearing times are not part of incident energy calculations other than at the line side of main overcurrent devices. So even if we classify all service entrance equipment as "Dangerous" we still need to know the available current from downstream equipment.

Without a knowledge of the range of available current how can any valid load flow study be conducted. Or should we always assume the utility can provide an infinite amount of motor inrush current?

 

[resqcapt19]

JBD,
I think that 70E is very much needed, but I don't think that there is any way to correctly apply it when the power is supplied by an utility. I think that there fault levels are so dynamic that you cannot correctly pick PPE to permit live work. The biggest problem is that a lower available fault current from the utility may result in a higher incident energy and a higher level of PPE.
Don

 

[JBD]

Don,
How can any incident energy calculation or table lookup be performed without a realistic value from the utility? I am not talking about live work at the service equipment. I am concerned about PPE requirements for testing the presence of voltage at other points in a facility that has a utility connection.

rbulsara,
I have no problem with the maximum design level for the selection of equipment.

The problem is as Don mentioned, lower available fault currents cause longer clearing times of protective devices resulting in higher icident energy. This is a major problem where the protective device does not enter it's "current limiting" range. In their SPD02 bulletin, Bussmann says "In some cases, using conservatively high bolted short-circuit currents may result in lower incident energy than what is possible. This is dependent upon the time-current characteristics of the overcurrent protective devices."

In my example at the utility max level of 57kA, a 400A RK1 fuse might release an incident energy of .25 cal/cm2, but at the more realistic level of 4Ka (208A/5.0%Z) the energy released could be 10cal/cm2 instead. now what happens when we lower the current even more by considering cable impedance.

 

[mc5w]

There is a rural power copperative just south of Columbus, Ohio that has the nameplates of their padmounted distribution transformers on the OUTSIDE of the transformer. This make disclosure of transformer data rather easy.

A 75 KVA is rather puny for this size of service but at any rate a typical US 102Y208 volt transformer has a Z of 1.8% to keep voltage drop low. Oil filled transformers of 500 KVA and less single phase are around 1.8% to 2.2% Z to keep voltage regulation reasonable.

All your asking is that they let you read the transformer nameplate. First Energy charges $200 to come out and unlock a transformer and your utility probably charges something similar when service conductors need to be changed. However, First Energy has a number of padmount transformers where the internal nameplate was removed at their warehouse and placed into a reference binder in the belief that the internal nameplate needed to be protected against the weather or some other damnfool reason. I would then need to look up the First Energy inventory number on the outside and have one of their paper pushers find the nameplate back at the warehouse.

 

[resqcapt19]

JBD,
In talking with a utility person on another forum, I was informed that there is no way that his utility can provide fault current information that is suitable for use in arc fault incident energy calculations. They only provide a maximum available fault current for equipment selection and provide a warning on this document that says "this information is not to be used for incident energy calculations".
I really don't understand how a utility can provide a fault current suitable for incident energy calculations. The distribution connections that feed current to any given point on their system are very dynamic and each possible combination of distribution connections will result in a different fault current and a different incident energy.
Just for an example, the plant where I am assigned is served by a 34.5kV feeder. This feeder can feed our plant from either of two utility substations. One of the utility substations is feed by a 138kV line, the other by a 69kV line. We have no way of knowing what line is feeding our service, and I would expect that the available fualt at our service changes based on which substation is feeding us. They also sometimes use the 34.5 line as a tie between their subs and this too would change the available fault current.
It is very possible that each of these different fault currents would result in a different level of PPE for live work.
Don

 

[rbulsara]

JBD:

In essence it is up to you to make reasonable assumptions as professional engineer, and this is not the utility co. or the software's problem. Yes, whatever assumptions or judgment you make may vary from the actual; all you have to show is you did the best you could in your professional judgment. This is all after all comes down to liability. As professional as long as you can show that you were not negligent, you should be OK or it is the best you can do. No one is immune to litigation.

And if the client is really concerned either for safety or liability, they should not be working on a live equipment or use highest PPE. Anything else amounts to nothing, if there is indeed an accident and there is litigation.

 

[JBD]

Don,
Why can't the utility give you all three possible values. Then you could determine the worst case PPE. If I had a self sufficient facility with parallel generators, I would have to run multiple scenarios to determine the worst case.

rbulsara,
I don't want my clients to work on live equipment. But, they must treat all circuits as live until 0V (actually <50V) is confirmed with test equipment. The highest level of PPE does not cover the most dangerous conditions.

Maybe we just need to start asking for a utility disconnect instead of relying on lockout/tagout.

 

[resqcapt19]

This whole thing is a "catch 22". You must prove that the circuit is dead or locked out before working on it. You must assume that the system is live until proved dead by testing. You must have suitable PPE to do the required testing. You must have the actual fault currents to pick the correct PPE. You can't assume worst case, because in many cases that will exceed 100 cal/cm^2 and there is no PPE available. Even if there were PPE for that level of incident energy, there is no PPE for the arc-blast, and many experts in the field say that above 40cal/cm^2, the only thing that PPE does is give you the possibility of an open casket instead of a closed one.
Until there is a way to get "real" or "actual" utility fault currents, there is no way to comply with OSHA or NFPA 70E. This issue will be a huge source of income for personal injury lawyers.
Don

 

[rbulsara]

rescapt:

actual utility curernt is not the only variable..so you last agruement is the reason for not trying to complying..and you do know actual utilty current within reasons, like in this original case it is a 75 kva tx, how much accutae one have to be..

I do agree it is a difficult or inaccurate method, but it was discussed in one of the NETA magazine/periodicals on this subject, we always deal with evolving science, we do the best we can with available tools. That is all is expected from a professional.

 

[resqcapt19]

rbulsare,
I'm not saying that we should not try to comply. I'm just saying at this time there is no way to be sure that the PPE is suitable for the application. I don't believe that we do know the utility current with enough accuracy to correct apply the requirements of 70E. If we start always requiring worst case PPE (40cal/cm^2 or higher), we will have even more resistance to the use of PPE than if we can correctly specify the PPE. This area needs a lot of work before it will be commonly used in the field.
The biggest problem is not with the very high fault currents, but when the available current is reduced for some reason, that will lead to longer clearing times and a higher incident energy.
Don