• We will be performing upgrades on the forums and server over the weekend. The forums may be unavailable multiple times for up to an hour each. Thank you for your patience and understanding as we work to make the forums even better.

746W/HP vs NEC 430 motor tables - sizing a generator/UPS

malachi constant

Senior Member
Location
Minneapolis
Hi all,

I've long been aware that 746W/HP is a good rule of thumb. Somehow I've never taken the time to compare it to the NEC tables. I have a motor application where a UPS sales vendor thought their equipment would be cost competitive with a generator. I sent them the HP loads and they came back with a UPS that I thought was undersized. In exploring that I discovered they used 746W/HP while I was using the NEC tables to get FLA, then multiplying that by 460 * sqrt (3). We are off by nearly a factor of two.

To be more specific, the application is (10) 7 HP motors.
Approach #1: 10 motors * 7 HP/mtr * .746kW/HP = 52.53kW
Approach #2: Per NEC 430.250, at 460V a 7.5HP motor has 11 FLA. 11 FLA * 480V * sqrt(3) * 10 motors = 91.34kVA

7 HP vs 7.5 HP accounts for some of the discrepancy.
I am but a lowly consulting engineer and mentally default to everything having unity power factor (i.e. in my simplistic world kW = kVA).

I do not think this UPS rep does a lot of motor loads so am hesitant to use their number. I could reach out to the generator rep (who I trust understand motor loads better with respect to generator equipment) but thought I would start here.

FWIW these are fan loads and each will be on a VFD. The reason UPS may be cost competive is the generator (or UPS) will located in a code-classified storm shelter, which is buried within the belly of a building - meaning it needs to survive and operate even if the building around it is destroyed - which makes generator intake/exhaust a challenge. The UPS may or may not have similar ventilation issues, haven't got to that exercise yet, just doing a first look on UPS cost and footprint to see if it is close. But them sizing their UPS about half as small as I expected needs to be figured out first.

That last paragraph you can ignore, but included it for those curious as to why I am putting so many motors on a UPS. My primary question for you all is wanting to reconcile the two kW calculation approaches with respect to sizing a power source (UPS or generator). Thanks in advance!
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
Hi all,

I've long been aware that 746W/HP is a good rule of thumb....
It is not a good rule of thumb for sizing anything.

746W/HP is good for estimating the output of a motor. It does not take into account the efficiency or power factor which are required to go from motor HP output rating to motor input rating.
For estimating input requirements, with a good margin of safety, I often use 1KW/HP.
 

tom baker

First Chief Moderator & NEC Expert
Staff member
Location
Bremerton, Washington
Occupation
Master Electrician
Our POCO uses 1KW per hp for transformer sizing. I have done one small
motor on a UPS and the ups was double the size of the motor kw due to startup amps. You should be able to stagger to statup of the fans with an on delay timer.
 
Last edited:

malachi constant

Senior Member
Location
Minneapolis
Jim/Tom - so the NEC motor tables are used for sizing motor branch circuit conductors / OCPD and to derive panelboard / service calcs. But are overkill for sizing generators/transformers etc - 1kW/HP is a good rule of thumb for those. I believe I follow the logic here, just want to make sure I understand. Thanks!
 

LarryFine

Master Electrician Electric Contractor Richmond VA
Location
Henrico County, VA
Occupation
Electrical Contractor
Don't forget to allow for motor starting current, which lasts longer than most inrush currents.
 

David Castor

Senior Member
Location
Washington, USA
Occupation
Electrical Engineer
1 kVA electrical input per 1 hp mechanical output is a conservative estimating guide for motor load estimating. The NEC FLA tables must be used for sizing motor feeders, but are generally very conservative.
 

dkidd

Senior Member
Location
here
Occupation
PE
Jim/Tom - so the NEC motor tables are used for sizing motor branch circuit conductors / OCPD and to derive panelboard / service calcs. But are overkill for sizing generators/transformers etc - 1kW/HP is a good rule of thumb for those. I believe I follow the logic here, just want to make sure I understand. Thanks!
It should be 1KVA/HP
 

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
How do you guys come up with that 1VA number?
Since T430.250 is non-linear, I'd say there is no fixed kw/hp ratio.
I graphed the VA of the table recently to make my own table for 400V motors.
To get T430.250 VA of each HP motor I take motor voltage from the header X SQRT 3 X table amps
So for a 460V motor I take 460V X 1.732 = 796.7V
Then for 460 1HP I get 796.7V X 2.1A = 1673 VA
1673 VA / 1HP = 1673
1.673 KVA/HP
etc..
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
How do you guys come up with that 1VA number?
Since T430.250 is non-linear, I'd say there is no fixed kw/hp ratio.
I graphed the VA of the table recently to make my own table for 400V motors.
To get T430.250 VA of each HP motor I take motor voltage from the header X SQRT 3 X table amps
So for a 460V motor I take 460V X 1.732 = 796.7V
Then for 460 1HP I get 796.7V X 2.1A = 1673 VA
1673 VA / 1HP = 1673
1.673 KVA/HPt
etc..
What values in T430.250 are based on actual running currents? They are intended to be higher than commercially available general purpose motors. The NEC specifically says to not use these table values when selecting motor running protection.
 

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
The NEC specifically says to not use these table values when selecting motor running protection.
My bad I thought we were talking about a load calculation for a optional standby system & feeder?
And in the case of actual running currents, it still non-linear.
 

topgone

Senior Member
It should be 1KVA/HP
Maybe not. It would be best to do the needed calculations and not use a rule of thumb. For example, I have a 5 HP motor that operates at full load with a PF of 0.84 and an efficiency of 80%. A quick calculation tells me that the electric power needed to power this motor will be:
= 5 X 746W per HP /(0.84 x 0.80) = 5,550.6 VA ~ 5.55 kVA!If we take the ratio of the power needed in kVA to the HP rating, that would be 5.55kVA/5HP = 1.11, more than 1 kVA/ HP. Not conservative at that range of capacity, IMHO.
1 kVA/HP will only hold on larger motors where the power factor is more improved and the efficiency approaches 100! Also, people on the other pond use 1 HP = 735.5 watts, far lesser than the figure that the late Mr. James Watt told people how much his steam engines are rated when compared to the power of a horse (746 W = 1 HP).
 

kwired

Electron manager
Location
NE Nebraska
Occupation
EC
Jim/Tom - so the NEC motor tables are used for sizing motor branch circuit conductors / OCPD and to derive panelboard / service calcs. But are overkill for sizing generators/transformers etc - 1kW/HP is a good rule of thumb for those. I believe I follow the logic here, just want to make sure I understand. Thanks!
NEC full load tables are worst case expected power factor/efficiency and intended for sizing items according to current drawn. Transformers are not kW rated they are kVA rated so the tables actually apply to transformers as well as conductors for the most part.

A generator has to be able to deliver the load, the inefficiency and any additional power factor related current. Use a power factor correction capacitor and you narrow it down to mostly needing to deliver load and the inefficiency, that's before considering whether it can deliver what is needed to start the motor anyway.
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
My bad I thought we were talking about a load calculation for a optional standby system & feeder?
And in the case of actual running currents, it still non-linear.
Yes we are talking about load calculations. The NEC does not provide much assistance in sizing sources, such as generators and transformers. It seems to be focused on conductor sizing and conductor protection.

Yes, motor currents are not linearly related to size.
 

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
The NEC does not provide much assistance in sizing sources, such as generators
Perhaps with a manual transfer switch.
with an ATS in the picture the way I read it is:
702.4(B)(2) -> sends me to article 220
220.50 sends me to -> 430.24
430.24 sends me to 430.6(A)
430.6(A) says to use the tables, not the nameplate.
If I graph the table 430.250 into VA(y axis ) X HP (x axis) it never touches 1000VA
 

JoeStillman

Senior Member
Location
West Chester, PA
Jim/Tom - so the NEC motor tables are used for sizing motor branch circuit conductors / OCPD and to derive panelboard / service calcs. But are overkill for sizing generators/transformers etc - 1kW/HP is a good rule of thumb for those. I believe I follow the logic here, just want to make sure I understand. Thanks!
The chrono-synclastic infundibulum is a zero-loss phenomenon.
 
Top