calculated vs measured

Status
Not open for further replies.
M

megloff11x

Senior Member
Is there a justification in the NEC for using measured current load for sizing circuit breakers, conductors, and disconnects, rather than the standard tables and calculations?

For example, let's say we have a machine that contains numerous motors and other devices. Going by the NEC and based on nameplates and datasheets of what is inside, it should hypothetically be sized for 130A disconnects and 2/0 wire. But also hypothetically, according to the recently calibrated and properly functioning meters, it draws less than say 25A RMS under maximum load. Why can't I use a 50A disconnect, 30A breaker, and 10AWG wire?

While erring on the side of safety is good, what happens when I build a factory with scores of similarly overrated machines? I have the power company diverting more resources from the grid to the building, maybe even adding facilities to handle me. I put in huge switchgear. I run fat wire and buy expensive disconnects and breakers, and when I'm done, I'm using a fraction (1/6th) of the current I asked for.

Had it been sized for reality, the power company would not have been involved, the wires and switchgear would be less expensive, and I would say that the plant would still be as safe as can be. Plants do allow for de-rating the overall based on capacity, but each station must be fully rated in most cases.

The people who rated the sub-sub-assemblies erred on the side of caution. The sub-assembly erred further. The whole machine erred further still. Then the plant with 60 or more similarly caution erred stacked machines followed suit, and so did the power company. When several layers each adds 15-25% on top of each other, it gets big in a hurry.

In this hypothetical case, we just spent a lot of money adding infrastructure that is also subject to misfortune. I can make a car almost utterly crash proof, but it will empty your wallet at the gas pump, and crush any other cars on the road, not to mention wear out the roads faster.

My question is, what part of NFPA 70 and 79 allow for measured values of performance to be used in lieu of calculations and table derived results, and what liability is incurred when this route is used? Or must you go strictly by the tables and calculations?
 
Re: calculated vs measured

You know there are a lot of people who will probably give you a more profound answer, but here's my opinion.

The reason I believe they don't go by measuring or the demand load is because the demand tends to flucuate. Sometimes the demand will be 60A, then will be at 100A. To my knowledge, motors can go up to 8 times their running current when starting, and then your breaker will be too small so it's sized 2.5, 1.75, or even 8 times what size it should be. Also what happens when you want to add loads to a panel, but you have measured each phase, and now you have to replace the panel and the feeder because it's exactly to the demand load. The peak demand load can be deceiving and can change throughout the use of the equipment. I think there are a lot of valid reasons, and for the most part the the code is a safe guard to those. What do you think? :)


Lady :)

[ January 05, 2005, 03:33 PM: Message edited by: lady sparks lover ]
 
Re: calculated vs measured

I think there is some dispensation that can be taken for sizing of feeders when equipment cannot operate simultaneously.

The problem with taking advantage of this is that you never know what someone will do down the road.

Measurement for a short period of time does not always give a true picture of what the potential load could be at any one time. Its not unusual for an industrial machine to be loafing along at half speed because they don't need the production. Bringing it up to full speed can substantially increase the energy requirements.
 
Re: calculated vs measured

megloff11x
Power companies shouldn't, and usually don't, use your breaker sizes to determine the anticipated load on their system. On the other hand, for new work, and individual loads, you can't use the measured amount. In general, you must use the nameplate values for appliances or industrial devices, values in NEC tables for motors, etc.
If you are ading load to an existing service or feeder you may use measured values per 220.35.
Jim T
 
Re: calculated vs measured

There is nothing that prevents you from using a smaller breaker for motor load. You are not required to install a breaker sized at 175% or 250% of the FLA of the motor. You are required to use a conductor that is 125% of the FLA of the motor as listed in the tables of 430 . If you can get the motor started with a smaller breaker, you may use it.

[ January 05, 2005, 04:44 PM: Message edited by: bob ]
 
Re: calculated vs measured

If you are designing a new building, you do not have any measured values to use in the design process. I, for one, would not accept the notion that you have another factory down the road, and a similar machine draws XXX amount of current, as the basis for sizing feeders at a new factory.

As to the utility, they know that the NEC service calculation methods are conservative. That is why they will give you a service that is sized for 40 to 60 percent of the calculated load that you give to them.

If you are designing an upgrade to an existing building, the NEC does have a method for using measured values to determine if any given feeder (or the service itself) has room for adding load.
 
Re: calculated vs measured

Charlie B.,

I'm not trying to be smart, but If I need 1000 amps I'm going to get around 500?

Edit: Are humans so bad at being honest or accurate that even in an engineering context we have to have 50 to 100% BS cusions?

[ January 05, 2005, 05:34 PM: Message edited by: physis ]
 
Re: calculated vs measured

I'm not trying to be smart, but If I need 1000 amps I'm going to get around 500?
Click to expand...
Physis
The utility will supply all that you can handle.
The point Charlie is making it that they are not required to size their wiring by the NEC. They use experience to estimate the maxium load that will occur and provide capacity to fill that
requirement. If at some time their facitities are not adequate, they simply go to a bigger size.
 
Re: calculated vs measured

Quote from physis: Are humans so bad at being honest or accurate that even in an engineering context we have to have 50 to 100% BS cusions?

sam -

Yeah, sometimes. Keep in mind my experience is 99 44/100 percent industrial.

Here is an example:

Suppose we are making a rather large mod. We're adding enough equipment I expect the consumption to increase maybe 25%. During the design, one of the mechanicals comes to me and says, "Our calculations show this pump takes a 30hp motor, but it will have to run up in the service factor - say at 105%. Is that okay?"

Not good. I like it if the motors are all sized perfectly and run at 100%, but leave the sf for all the pieces of the calculations that are a little fuzzy.

So we put in a 40hp and when it comes on-line it is running at 70%.

Did we do wrong? Well most of the money we spent on the high efficiency motors just went down the toilet. The motor leads are oversize. The starter is okay, we would have had a size 3 anyway. At least is wasn't a 50hp to 60hp change, that would have meant a change from a size 3 to a size 4.

But the motor runs cool and it will maybe last 10 years. Load it up to 105% and it sure won't. And next year when ops figures out how to get another 10% production, there is headroom left in the distribution and drivers to handle it.

Multiply this scenario by 50 motors ranging from 10hp to 200hp and add two 2200hp motors and guess where we are.

Now suppose we design this right down tight - right up to 90%+. Motors are hot, switchgear is hot. Maintenance crews are hot :mad: . And next year when we want to bump production, we can't accommodate with out a significant upgrade.

On a personal accountability note, say I'm a really good at the electrical and I'm working with really good mechanicals and we size 90% dead on, 5% is oversize and 5% is undersize. So when we hit startup, we can't come up to full production cause we have 5 out of 100 pumps (heck even 1 out the 100) that won't pull what they need to - and we have to do rework to get full production. How much credit do you think we get for designing a tight, efficient system?

Suppose instead we designed with 10% to 50% headroom, 10% in some places, 50% in others. Everything comes up working per plan. Everything is running cool. Maintenance doesn't look like it will be a headache. Life is good. I've never had anybody complain about the relatively small added cost (compared to the total project) when everything comes up running and nothing is marginal.

Now lets look at this from the utility point of view. We told them the truth, and gave them the data, and likely our screw-up are even with our screw-downs, so they size the service to fit the load - 25 to 50% down. We don't want them to add any more than they have to cause someone (that would be us) has to pay for the extra equipment.

So five years later, we have added some equipment, tuned up the process and are now halfway into our headroom. Now the utility transformer is running hot and the regulation is starting to suck (please use the precise engineering definition). So we ask the utility, "What do you have in mind?"

They respond, "Yeah we've been seeing this coming. We can either change the transformer or add a parallel unit. It will take us a couple of months to get ready. When do you want the take the outage."

So, did any of us do so wrong adding our 50% BS cushion and the utility taking our data and sizing the feed down to what is required?

Generally, the customer would rather we are 50% over on the electrical distribution than 1% short of meeting production.
 
Re: calculated vs measured

Originally posted by physis: If I need 1000 amps I'm going to get around 500?
Click to expand...
I agree with Bob?s response. But I will add that you do not actually NEED the 1000 amps. It is more correct to say that ?The results of your service calculation, as performed in accordance with NEC Article 220, show a service load of 1000 amps.? The utility will have more experience than you or I, in the matter of converting a service calculation into a ?need.? That experience is what leads them to give you a service transformer that is 40 to 60 per cent of the results of the service calculation. It is also what leads them to give you a transformer that meets your actual needs.
 
Re: calculated vs measured

Just to add to Charlie b's response:

A lot of the time the transformer is overdriven as well. The they will overdrive the transformer up to 40%, so when they give you a 1000KVA transformer, and you need 2000KVA it's most likely a combo of demand loads and the fact that the transformer can take more load that it's rated for.


Lady :)
 
Re: calculated vs measured

I'm really comfortable with Coulter's deliberate assessments. That's rational.

It just looked a little scary, people coming along seemingly arbitrarily adjusting things.

It still seems to amount to a BS cushion, which I don't suppose you could get by without in the real world.
 
Re: calculated vs measured

You still seem to be confusing "need" with "results of 220 calculation." It's probably best to avoid using the word "need," in this context. Better to say, "It the calculation results is 1000 amps, then it is likely that the actual, measured load, throughout the life of the facility, will not exceed 500 amps."
 
Re: calculated vs measured

No, I'm better now Charlie, I understand. :)

Edit: The last post was more or less sarcastic.

[ January 07, 2005, 03:40 PM: Message edited by: physis ]
 
Status
Not open for further replies.
Top