Electrical Load Study?

[iwire]

But there is no demand factor in the code for industrial MCCs.

Maybe not but there are allowances for noncoincident loads.

 

[Smart $]

Maybe not but there are allowances for noncoincident loads.

...and no concise method.

Has anyone ever noticed how the text uses double speak? Noncoincident loads by common definition would be loads which would not be in use simultaneously. Seems like the original proposer tried to squeeze the definition into the requirement.

 

[don_resqcapt19]

Maybe not but there are allowances for noncoincident loads.

Only if the logic and or wiring prevents the loads from being on at the same time. That not the case for most of the industrial MCCs that I see.

 

[Smart $]

Only if the logic and or wiring prevents the loads from being on at the same time. That not the case for most of the industrial MCCs that I see.

That's an interpretation not in evidence, i.e. the requirement doesn't actually say that. Also, it uses the word "unlikely", which opens the door to alternative interpretations.

 

[iwire]

Only if the logic and or wiring prevents the loads from being on at the same time. That not the case for most of the industrial MCCs that I see.

Hmm, in your case something is keeping them from running at the same time or the feeder would trip.

I agree with Smart$, its entirely up to interpretation.

 

[don_resqcapt19]

Hmm, in your case something is keeping them from running at the same time or the feeder would trip.

I agree with Smart$, its entirely up to interpretation.

The process design keeps everything from running at the same time under normal conditions, but does not prevent everything from running under abnormal conditions. Also the process/mechanical guys often use motors with more horsepower than the load requires, so the load is less than the NEC calculation because of that too.

 

[Lady Engineer]

The process design keeps everything from running at the same time under normal conditions, but does not prevent everything from running under abnormal conditions. Also the process/mechanical guys often use motors with more horsepower than the load requires, so the load is less than the NEC calculation because of that too.

This is not completely true. There are duty pumps and standby. Just because the connected load is 1000A, it does not mean that the load will ever reach that. We don't use the cycle on and off method, because in the case of Sandy in NJ, the duty pumps were running full blast at most of our stations, and guess what not one of the MCCs were overloaded.

That's why we have process guys, and the control panel will prevent the duty and stand by pumps from running at the same time, and there's a fail/stop method in the control panel just in case the CP PLC malfunctions. Most pumping stations and water treatment plants will always have redundancy, because pumps do fail, but it's built into the control for the pumps not to start until the others have stopped. Then we go into manual or remote operation, from a work station if PLC fails. Also, if they choose the method of having a pump start, prior to the other pump stopping, we do build that into our calcs, but most WTP have generators, so we must be careful about our sizing anyway.

As I said before, we do have have non-code based calcs for doing plant service sizing. The NEC deals with dwellings and some commercial buildings for service entrance conductor sizes. However, you can gauge your needed service size based on the pump being standby or duty, and previous behavior of that plant or other plants. I've NEVER seen a MCC overloaded, but if you add up the connected load, sure it appears that way.

 

[Phil Corso]

Gentlepeople...

For a fairly complete discussion on 'How to size a Xfmr' including treatment of "spare" components, such as might be found in Secondary-Selective Distribution Systems, I suggest:

W.H. Dickinson's, "Method Of Electric Load Data Calcs...", found in,

IEEE Transactions On Industry and General Applications, Vol.IGA-2, No.5, Sept/Oct 1966.

Regards, Phil Corso

 

[don_resqcapt19]

...
As I said before, we do have have non-code based calcs for doing plant service sizing. .... I've NEVER seen a MCC overloaded, but if you add up the connected load, sure it appears that way.

We have code rules that, in my opinion, apply to the MCC feeder and the MCC busing. It appears to me that 430.24 applies to both the MCC feeder and the MCC busing. That being said, I do not see those sections applied to industrial MCCs.

As far as tripping MCCs, I have seen a few where the feeder OCPD had tripped from overload.

 

[Lady Engineer]

We have code rules that, in my opinion, apply to the MCC feeder and the MCC busing. It appears to me that 430.24 applies to both the MCC feeder and the MCC busing. That being said, I do not see those sections applied to industrial MCCs.

As far as tripping MCCs, I have seen a few where the feeder OCPD had tripped from overload.

Then it wasn't a good design, or maybe they replaced the motors with larger sizes, without increasing the service. Usually when I ask for the peak demands for an existing service, the largest I've seen is 60% of the rating.

That 125% applies motor feeders, but after 125% is added to the largest motor, you CAN add the remaining at 100%, but if you think the service will be sized properly by applying 430.22, then good luck. We usually add 125% to the entire building load, because of the continuous loads, but use 100% breakers. I would go with 250%, per table 430.52, then add the remaining motor loads if I weren't sure how to do it.

I personally feel the code doesn't addresses a plant, but there are ways to come up with a good design without it.

 

[don_resqcapt19]

Then it wasn't a good design, or maybe they replaced the motors with larger sizes, without increasing the service. Usually when I ask for the peak demands for an existing service, the largest I've seen is 60% of the rating.

That 125% applies motor feeders, but after 125% is added to the largest motor, you CAN add the remaining at 100%, but if you think the service will be sized properly by applying 430.22, then good luck. We usually add 125% to the entire building load, because of the continuous loads, but use 100% breakers. I would go with 250%, per table 430.52, then add the remaining motor loads if I weren't sure how to do it.

I personally feel the code doesn't addresses a plant, but there are ways to come up with a good design without it.

I don't think I ever commented about service sizes...just MCC and MCC feeder sizes and as I said in one of my first posts, I don't think I have ever seen an industrial MCC that complied with the code rules and most work just fine.

The ones that I have seen trip were when major changes were made to the operating process and the load was increased.

While I would agree that the code does not do a good job with rules for industrial occupancies, I don't think that there is any question that the NEC applies to those occupancies. There is nothing in 90.2(A) that would lead you to think that the NEC does not apply to industrial installations.

 

[LeeB]

Load Survey

Load Survey

NEC 220-87 states that existing load can be calculated, where you would need to make a list of motors and assign loads per 430-22 or what may be easier is a 30-day load survey, taking the highest average reading over a 15-minute period.
We use data loggers made by Dent Instruments which work well, but cost $3K with CT's. Some places rent them or you can call a contractor that performs the survey.
Hope this helps.