This topic comes up a zillion times. Everyone wants to know if they can run their 5HP Unisaw or PM66 or Dust Collector on a 12ga wire, what circuit breaker to use, why their product manual lists one amperage for the motor, then recommends a different circuit breaker, etc.
Here's some info on the topic. My caveat, I am not an electrician, and this is not the exact NEC but excerpts. Intended as a guide only, consult your electrician and local codes and local inspector, not responsible for your misuse, etc. etc. That being said, I have read pretty in depth on the topic and in the NEC and on various forums, and the info contained should get you started on the right path/research of your own.
==============================================
Most folks want to just look at the motor nameplate for their 5HP motor, see 19.5 A, then google up "ampacity of wire", find a table from the NEC table 310.16, allowable ampacities of insulated conductors, and read across to find a wire that has an amps listed above 19.5. Here's the table:
From here you can see that 12 gauge wire with 60C and 75C insulation ratings can flow 25A and not heat up above their insulations temperature rating. Folks then look at the 90C column (THHN) and find 30A and think, "wow! I can use THHN and go up to 30A on my 12 ga!". Even the 14 AWG wire will let you flow 20A for 60C and 75C insulation, and go up to 25A for 90C THHN insulation--the wire itself will take it. Sooo tempting hey? But...
HOWEVER, the ampacity table has a footnote for "small diameter conductors" i.e. 14, 12, and 10 ga:
Small conductors [Sec. 240-3(d)]. Unless specifically permitted in Sec. 240-3(e) through Sec. 240-3(g), overcurrent protection shall not exceed 15A for No. 14, 20A for No. 12, 30A for No. 10 copper, and 15A for No. 12, and 25A for No. 10 aluminum and copper-clad aluminum - after taking the ampacity correction factor into account.
People often stop here and just say "there it is, 20A max on 12ga, your 5HP is above that, so must use 10ga." Unfortunately, it's a lot more complicated than that. If you don't want the complication, just go with this general rule, you'll likely be conservative and safe. If you want to see if you CAN push above these small conductor limits, you must see if it is "specifically permitted elsewhere in the NEC", sections 240-3(e) thru 240-3(g) list the exceptions. And we have to apply those ampacity correction factors still!
... so you actually have to correct for ambient temps above 30C or 86F, and for more than 3 current carrying wires bundled in a conduit (which you can read at the top of table), and derate the ampacity of the wire for both of these. That's because the ampacity is based on the wire's insulation ability to not melt at that current flow and resulting wire temperature... and anything that keeps the wire from cooling off like high ambient temps or lots of wires bundled together makes the thing get hotter.
For ambient temps above 30C or 86F, you can read the correction factors for various ambient temps for each wire type at the bottom of the table. Different wire insulation types have different correction factors due to their different abilities to dissipate the wire's heat (insulation properties).
When wires are bundled together, like inside conduit or a raceway, they can't dissipate heat as well and so they get hotter than expected. NEC Table 310-15(b)(2)(a) limits the permissible load by giving derating factors that apply to table ampacities. For example for 4-6 wires the ampacity is only 80% of the table value; 7-10 wires is 70%; more than ten but fewer than 21, 50%, and so on. However, if the raceway is not over 24 in. long (classified as a nipple) the NEC assumes heat will escape from the ends of the raceway, and the enclosed conductors need not have their ampacity derated.
We'll figure out what the amp requirements for our 5HP motor are, then we'll come back to this table, apply temp and bundling corrections, and read what wire size that gives us. Since we don't want to just live with the "12ga= 20A CB, 10ga=30A CB" rule, we're going to look at the exceptions to this basic wire-protection giudeline, and luckily MOTOR CIRCUITS are one of those!
"Small gauge wire 14/12/10 limitations--exceptions" Section 240-3(e) thru 240-3(g) list the exceptions,we're interested in Specific equipment requirements [Sec. 240-3(g)].
Motors [Secs. 430-22(a) and 430-52(c)].
Protecting motor circuits has three separate requirements:
1. A disconnect which must be visible from the motor point of operation, and <= 50ft. You can use a circuit breaker or a plug/receptacle for a 5HP motor, there are actually specifics for more complicated motors.
2. OVERLOAD protection-- protecting the motor and it's windings from damage if it's overloaded, this is usually a thermal cutout in the motor or starter circuit itself.
3. Short circuit protection and ground fault, i.e. the circuit breaker.
Overcurrent and short-circuit protection aren’t the same for motors, unlike most other circuits. Usually we provide overcurrent protection for most circuits by using a circuit breaker that combines overcurrent protection with short-circuit and ground-fault protection--the circuit breaker has to do it all, the NEC has no idea what might be plugged in there. However, this isn't usually the best choice for motors, because motors can have a huge inrush current to get them started, 2-3 times normal operating current, which would trip a breaker. With rare exceptions, the best method for providing overcurrent protection in these cases is to separate the overload protection devices from the short-circuit and ground-fault protection devices. Motor overload protection devices like thermal overload disconnects (which is on the 5HP Baldor motor on Oneida cyclones) protect the motor, the motor control equipment, and the branch-circuit conductors from motor overload and the resultant excessive heating. They don't provide protection against short-circuits or ground-fault currents. That's the job of the branch and feeder breakers, which don't provide motor overload protection. This makes motor calculations different from those used for other types of loads. Why? Because if the motor itself already has more than HALF of the total circuit protections built into IT, the NEC lets you take advantage of this! If the motor doesn't have built in overload protection, you can calculate that part and supply it still separate from the circuit breaker, but we won't have to do that for our motors since we're only going to use one that is thermally protected already... I hope! To do the calculations, you apply NEC article 430.
Overload protection. Motor overload devices are often integrated into the motor starter, and you see this on the motor nameplate by something like "thermally protected overload". But you can use a separate overload device like a dual-element fuse, which is usually located near the motor starter, not the supply breaker. Motors rated more than 1HP without integral thermal protection must have an overload device sized PER THE NAMEPLATE RATING per table 430.32. It takes into account eh motor service factor and the allowed thermal rise of the motor, both on the nameplate.... THIS ISN'T APPLICABLE to most dust collector motors, or your unisaw motor most likely, as they ARE integrally protected, and so the nameplate FLA rating of the motor is not used!
Sizing branch-circuit conductors. (the wire size finally!)
I'll use my Oneida Pro2000 Cyclone Dust Collector for my example, it has a motor nameplate that says 5HP, 19.5 amps, SF 1, continuous duty at 40C.
According to Sec. 430-22, branch circuit conductors serving single motors of continuous duty are sized at no less than 125% of the motor full-load current listed in Tables 430-147 to 430-150, (430-128 for single phase in our case) not the motor nameplate amperes [Sec. 430-6(a)].
(here's the link for 3-phase FLC motor ratings for completeness:
NEC Table 430-150 Full-Load Current - Three-Phase AC Motors )
So according to the NEC, which you must use to calculate overcurrent protectionFull Load Single Phase Motor Amperes for 5HP @ 230V is 28 Amps. When motor service factor is greater than 1, you increase full load amps proportionally, so if service factor is 1.15, increase above amp values by 15%. My Oneida 5HP Baldor is SF 1.0, so NEC says 28 FLA.
------This is the motor current number you must use to do any calculations if you want to exceed the 20A CB on 12ga wire "small conductor" rule!
Continuous Duty or Not?
Boy is that part complicated! The NEC says that if a motor runs under full load for 3 hours or more, then it is "continuous duty", and you have to size all the wires for 125% above it's FLA from table 430-128 above (or another way of saying it is you can only have a motor FLA that is 80% of the wire's ampacity, 125% = inverse of 80%). It also states that "Note: Any motor application shall be considered as continuous duty unless the nature of the apparatus it drives is such that the motor will not operate continuously with load under any condition of use."
If it's not continuous duty, for other types of circuits, you get to use 100% and just make the wire match the load... but motors become even more complicated. You have to look up the duty-cycle descriptions in table 430.22(E). You'd THINK that a motor rated for continuous duty like the Baldor, that you ran at something less than that like 15-30 minutes then off, you'd be able to use the normally calculate wire, or even less, but it's the exact opposite of what you'd think! You're not even allowed to run a continuous duty motor for "periodic" according to the NEC, it doesn't give the motor time to cool down from the current inrush of starting! So, for continuous duty motors run less than continuous, it's basically a 1.4 correction factor to NAMEPLATE current, not FLA.
So, for my 5HP, depending on how I classify it's use:
continuous duty = 125% * 28A = 35A
intermittant or periodic duty = 140% * 19.5A = 27.3A
varying duty = 200% * 19.5A = 38A
Well, I read for 2 hours and 20 motor examples but think the best call is to call it "continuous" as that's what an inspector will assume for any motor circuit, and the note at bottom of 430.22E pushes you that way, so we'll use 35A as the requirement. (However, a case might be made for intermittant duty. ) Remember, you're sizing this wiring beefy to allow for HEAT DISSIPATION, not real expected increased current flows!
When we size wiring, there are two completely separate calculations we have to look at, and then take the most conservative of the two:
1. wire termination criteria for the devices the wire hooks to on either end. Basically everything you screw the wires to is temperature rated, and you can't let the wiring exceed that temp--the device manufactures are allowed to use the wire itself as a "heat dump" and they expect to! You have to use 60C unless the device specifically says 75C on it... but virtually everything and all motors made recently assume 75C. So even if your wire insulation rating will let you flow current until it heats up to 90C, you're capped at 75C because it's screwed onto a circuit breaker than can't take almost-boiling temps! This is all covered in NEC [110.14(C)]. It's similar to the wire-size increase required for motors in that it's there to dissipate heat.
2. The allowed current that wire can carry without heating up above it's insulations temperature rating. All wires of the same gauge will flow the same currents and heat up at the same rates, the reasons for different "ampacities" is that the insulation melts at different temps. So long as you stay below the rated amperage for a given gauge/insulation-rating, that wire can flow that current for a century and never heat up above it's temp rating. The critical chart, probably the most-seen table in the NEC, table 310-16 Allowable Ampacities, is up at the top of the post.
I need to size my wires for 35A. Looking at table 310-16, I see 10AWG 75C wire will do 35A, so that should work fine for wire termination criteria--so long as I'm not using NM cable, which is limited to 60C termination (as Rollie points out below). My sub-panel and receptacle both take 75C connections.
For derating, I'll assume I run a new circuit with 3 or fewer current-carrying wires in the conduit so no bundling derating. If I had to account for this, it doesn't matter what type wire insulation I use, the multiplier is the same for all.
However, in Phoenix my workshop may reach 105-113F, or 41-45C, so I need to derate for temperature. If I use 90C THHN wire, I use a correction factor of .87 but am allowed to use the full table amperage of 40A for this calculation, so I get 34.8A derated. If I used the lesser 75C wire, 35A*.82 = 28.7A. Both of these are well above the ACTUAL current the motor will really run, 19.5A, so I could use either one.
What Circuit Breaker?
Here's where the motor article 430 lets us avoid the "30A breaker on 10AWG wire limit".
The motor Short-Circuit/ Ground-Fault protection device must be sized at no more than the motor full-load current rating listed in Tables 430-147 to 430-150, multiplied by the percentages shown in Table 430-152. According to Table 430-152, inverse time circuit breakers (a normal CB) may be sized up to 250% of the motor FLA, if required for starting. So, since motor FLA from the NEC tables was 28A, I am allowed to protect the circuit for ground fault/short circuit with up to a 28A*250% = 70A breaker! Of course we don't want to do that unless required to allow starting of the motor--in fact, I'll start with a 30A breaker and see how that works (it should as it's decently above the 19.5A I really expect from nameplate), but I'm ALLOWED to go all the way up to 70A because the motor itself on this dedicated circuit provides overcurrent protection to avoid heat buildup in the wiring. And since 70A isn't a normal size, I'm even allowed to step up to the next normal size (240.6) of 80A! I know that goes against the grain of most of what we learn ("the CB protects the wiring!"), but that just highlights how different motor circuits (with integral overload protection) really are!
[ In fact, table 430-152 gives me four choices for branch SC and GF protection:
standard fuse @ 300%, time delay fuse @ 175%, instantaneous breaker @ 800%, and inverse time breaker at 250%; then round up to the next available normal size fuze/CB. These are all max allowed, and you should start with the lowest fuse/breaker you can and then step up to the max if required for motor starting. And remember this is only allowed when the motor itself contains overload protection, or it's on the circuit as a starter or motor controller with overload protection, a good point to re-emphasize as Rollie says below!]
In summary, I ended up with a 10AWG wire and a 30A to 70A breaker on that wire... a loooong roundabout way to arrive at the common "wag" of running 5HP on 10AWG. I think a 5HP tablesaw or bandsaw would be open to playing around with the definitions for continuous use, and might allow you to take a stance for 12AWG as being acceptable due to non-continuous rated motors running periodically where wiring can be .85-.9 of FLA resulting in 25A... but I was unable to find any examples of woodworking motors being treated in this way, myself.
I'm sure I made some minor mistakes, the NEC is very complicated in this area, and again I don't claim to be an electrician. This does replicate methodology I found for 5-6 examples in various NEC-learning books, and the intent is to demonstrate the complexity of the analysis, when this subject comes up again. Hopefully I won't be raked over the coals by an electrician too badly, and I'll be happy to tweak if needed!
Enjoy
Reply With Quote
(It's pretty EZ there though, as the manufacturers do all the work & the info is on the data plate).
