Wire Size For DC Motors

[aaron_c]

So I'm preparing to order stuff so I can wire up my new trailer. And I have a question that I can't find an answer to, which is: How do I size the wire used to power a motor with DC power? I understand that electric motors require more current on startup. In solar systems those motors are usually run off the inverter and the wire running from the battery to the inverter is of course sized for the inverter. No problem, lots of advice on how to do that.

But I've also got two small travel fridges that will be powered via a DC connection. If I understand the situation correctly, they use brushless motors, and I think those motors will require a higher-than-rated current draw when they start up. For the sake of discussion lets say that they normally use as much as 3.7 amps, but at startup they could use 20 amps (that's what the person on amazon's Q&A said for this product, though the startup current seems high to me...). Anyhow, I understand I'd want to size the fuse at 3.7 A X 1.25 = 4.0875...so lets just say 5 amps. And I understand that I should use a slow-blow fuse for this, so that the startup current doesn't blow the fuse. But what I don't know is how I should size the wire. Should the wire be sized for the 3.7 A which is the normal draw? Or should I size it for the peak draw?

Thanks yall!

 

[John Frum]

Just wire and fuse for continuous amperage and voltage drop.

 

[John Frum]

Wire size calculator by voltage

Voltage drop and ampacity for 4/0 to 10 AWG for any voltage. "Chart is based on USCG, ABYC, and UL standards for marine-rated (105°C dry / 75°C wet) wire."

diysolarforum.com

Fuse size should be 1.25 times the continous load or the max for the wire, whichever is smaller.

 

[Hedges]

You'll have a fairly hard time sizing wire for 3.7 Ampacity. I would just use 12 AWG stranded, with whatever fuse the fridge calls for but not over 20A. As you say 5A slow-blow might be fine, but that is minimum, not maximum.

I'd also run a fat wire from the battery (with fuse at battery), then tap off to these circuits. Maybe 6 AWG? Depends on what your total loads are.
I would expect a brushless DC motor to have low start-up current because it soft-starts.

 

[aaron_c]

Thanks everyone, that was quick! And very helpful and to the point.

So I'll size the wire for the continuous amperage. @smoothJoey I have read that electric motors use less voltage at startup while they use more amperage, which I think is what you are referring to when you said I should wire for voltage drop? But I'm confused, I thought that, at least with a 12 V system, I'm basically sizing wires by the amperage and not the voltage at all. Sorry if this is a stupid question, I'm definitely new at this.

Follow up for @Hedges: I'm confused about the use of the "fat wire". Is that a fat wire that goes to the fuse box, because I need to size that for just in case I turn on every DC appliance all at once?

 

[Hedges]

When sizing for voltage drop, consider maximum start-up current, say 20A. Make sure wire resistance won't drop the voltage at motor too far below 12V.

If your battery is in the trunk or on the hitch, run a fat wire to the fuse box. Minimizes IR drop and carries enough current for any or all appliances on at once. Ideally, put a fuse at the battery to protect that wire (whether or not individual circuit fuses add up to more than that fat wire can carry.)

Having the fuse box closer to the several large loads reduces length of thin wires and IR drop to appliances.

For 3.7A to 5A, wire ampacity charts show 22 AWG to 24 AWG, which would be silly. That's LAN cable. I would just get spools of stranded house wire (12 AWG) or worst case automotive hookup wire (16 AWG) which is good for about 13A.

I would use 12 AWG. Maybe you'll put in a fridge that draws more than 50W some day.

 

[John Frum]

@smoothJoey I have read that electric motors use less voltage at startup while they use more amperage, which I think is what you are referring to when you said I should wire for voltage drop?

Nope I'm talking about voltage drop as a function of wire thickness, length and current.
The link I shared should have some explanatory power.
Don't worry about the surge current when sizing the conductors or fusing.

 

[Hedges]

Wire size calculator by voltage

Voltage drop and ampacity for 4/0 to 10 AWG for any voltage. "Chart is based on USCG, ABYC, and UL standards for marine-rated (105°C dry / 75°C wet) wire."

diysolarforum.com

Fuse size should be 1.25 times the continous load or the max for the wire, whichever is smaller.

How about, "Fuse should be no less than 1.25 times the continuous load. Wire should be at least large enough for the fuse."
Otherwise, if the wire is too small, you put in the smaller fuse less than 1.25 times continuous load and nuisance trip causes all your food to spoil.

Also, undersize wire can have too much IR drop at startup so a motor never starts, and the motor overheats.
His application has such small loads (3.7A continuous) it is cheap enough to oversize everything.

 

[aaron_c]

Once again, this is just great. Thanks for the link (which you shared earlier but I missed) @smoothJoey, I've bookmarked it. I didn't realize that smallish or undersized wires could cause voltage to drop.

And thanks in particular for this quote @Hedges

Fuse should be no less than 1.25 times the continuous load. Wire should be at least large enough for the fuse.

I was just wondering about this today...though it does lead to yet another question: When you say "large enough for the fuse" are you suggesting that I size the fuse by the wire size rather than the appliance? I understand that would be an easy to way to do it, and I'd avoid undersizing the fuse and having problems with it popping when it shouldn't. But if I just go with 12 AWG wire for all theses DC appliances and I give them all 40 A Maxi fuses (which I believe is what this chart indicates I should use?) then would I be putting my lower-amp appliances at risk? Is there some scenario where my system would put 12 V / 30 A though my 12 V / 4 A computer charger and fry my computer and/or charger? Or are fuses really there to protect the wire, not the appliances?

 

[John Frum]

Once again, this is just great. Thanks for the link (which you shared earlier but I missed) @smoothJoey, I've bookmarked it. I didn't realize that smallish or undersized wires could cause voltage to drop.

And thanks in particular for this quote @Hedges

I was just wondering about this today...though it does lead to yet another question: When you say "large enough for the fuse" are you suggesting that I size the fuse by the wire size rather than the appliance? I understand that would be an easy to way to do it, and I'd avoid undersizing the fuse and having problems with it popping when it shouldn't. But if I just go with 12 AWG wire for all theses DC appliances and I give them all 40 A Maxi fuses (which I believe is what this chart indicates I should use?) then would I be putting my lower-amp appliances at risk? Is there some scenario where my system would put 12 V / 30 A though my 12 V / 4 A computer charger and fry my computer and/or charger? Or are fuses really there to protect the wire, not the appliances?

Fuses protect the wire not the appliance.
https://diysolarforum.com/resources/wire-and-fuse-selection-quick-reference-blue-sea-systems.34/

 

[Hedges]

Here's a link to NEC ampacity. It also has good stuff like derating for higher ambient temperature and for multiple wires bundled together.

Note that for the smaller conductors e.g. 12 AWG which has ampacity of 30A, "**" references a note that says no larger than 20A fuse/breaker.
Those circuits typically have multiple outlets and get overloaded often. A 20A breaker will carry 30A for about 10 minutes, and the 12 AWG wire isn't damaged. Overload currents like that if they exceeded ampacity would bake out the insulation after maybe 100 times.

The BlueSea chart you linked shows a 40A Maxi fuse on 12 AWG, with notes 105 degree insulation and not in the engine room. You certainly don't have to use that large a fuse. Looks like the minimum was 5A for your refrigerator. House wire is rated 90 degree, so couldn't support 40A. If you're not in a hot location it could support 30A for this non-building application. But I'd use 20A (or less to possibly protect the appliance as well.) Figure the appliance has thinner wire inside, and you don't want that wire catching fire either.

When I said Wire "large enough for the fuse", I meant pick a fuse large enough for 125% of the load, e.g. 5A in your case. Then pick a wire which can carry 5A or more. You don't want to have a wire that can only carry 4A and then either use a 5A fuse (fire danger) or a 4A fuse (and it might blow due to current draw of the refrigerator.)

The only way you will put 12V/30A through your computer charger that is supposed to be 12V/4A is if the charger draws too much (or maybe your voltage goes too high due to a failed regulator/alternator). But the 30A fuse would let it happen if the charger got a short. A smaller fuse would blow sooner. The charger may have its own fuse, maybe not. The oversize fuze won't fry a computer unless the computer wants to fry.

 

[aaron_c]

@Hedges

So much good info. OK, that's good to know about minor overload currents, and particularly about minor overloads to the fuses. Is that why you seemed unconcerned about a slow-blow fuse vs a regular fuse? Should I interpret that to mean that I don't need a slow-blow fuse for the inverter, even if I anticipate that the 2000 w inverter will occasionally output as much as 4000 w?

Also: The engine room thing is a heat thing. That makes sense!

OK, I think I'm getting the hang of this. So the wire determines the maximum fuse size and the appliance determines the minimum fuse size. And if there's a big discrepancy there, I aim towards the lower end of the range.

I found a link to the NEC ampacity data you mention. Another good reason to use oversized cable, since bundling cables together (and I plan on doing that) will--if the appliances are all on at the same time--effectively reduce the ampacity. I guess because of heat? Also there's at least some chance that my trailer will occasionally be subject to temperatures above 40 degrees C. So I'll prepare for that with large enough wiring. And I suppose, large enough fuses. So depending on my wire and how heat-resistant it is, I might need to use fuses that are at least 1.4 X the rated amp draw of my appliances, since I don't want a fuse frying just because it was too hot. Just as long as the wire can support a fuse that large.

 

[Hedges]

I might prefer a fast-blow fuse sized for 4000W rather than slow-blow sized for 2000W, has a chance to protect electronics. Only if wire ampacity can handle that. There are "semiconductor fuses" meant to blow fast and protect electronics. Special fuses with good performance at high prices, not something you'll be fitting in an automotive fuse box. You should have a suitable fuse right at the battery. I spent about $180 on two 350A fuses and two fuseholders from Blue Sea. The breakers in my inverters are supposed to trip first, but a cable short to chassis will blow the fuses.

See if you can find a time/current chart for your fuses. Quality fuses could handle 2x rating for some seconds, don't know about yours. Slow-blow might carry it for minutes.

NEC ampacity: Yes, heat. Bundle of more than 3 means heat from center conductors also passes through insulation of outer conductors. The derating isn't applied if only crammed together for less than 2' before branching out (like a short conduit between two boxes) because it can conduct lengthwise through the copper to where bundle is smaller, then escape to air.

If you design everything according to NEC which covers AC circuits in residential/commercial, that should give margin for your mobile application.

2000W at 12V? That's 200A (NEC would call for 2/0 cable) but if continuous duty then 1.25 oversize would be 250A fuse, 4/0 cable. You may decide that is only a few minutes at a time, use a smaller fuse and/or 105 degree 2/0 cable. This is why we get 24V or 48V inverters for solar power, but if you're charging off the vehicle while driving the 12V is probably it.

It would be good to have the inverter real close to the batteries (but not exposed to outgassing). Keep the cables unbundled and vented, "in free air" so ampacity rating is higher.

 

[aaron_c]

One more question!

So I understand that heat reduces the ampacity of wires--different amounts depending on the temperature rating of the wires in question. But I realized that I don't understand how this interacts with wire length.

For example I've got a few wires that will travel maybe 25 feet (50 foot round trip). These wires are long enough that I have to increase the wire gauge substantially just to make it from my battery to my appliance (and back). This means that even a generously-sized fuse (for example a 15 amp fuse on a wire for a 5 amp device) is well under the maximum current that the wire could safely carry. Yes, if the wire were used to carry this maximum current it would lose a lot of that electricity as heat, but not enough to cause fires or whatever because the heat loss would be spread out over the length of the wire. If I understand correctly.

So here's my dillema as I pick wire: I've got some wire that's rated for -40 - 80 degrees Centigrade. I take that to mean it's close to the 75 degree rating on the NEC ampacity charts so I estimate from the chart that in temperatures of 45 degrees centigrade it might lose 25% of it's ampacity. But the wire is being used for a long (50 foot) run. Which means that the loss in ampacity only matters if high ambient temperatures an the resulting loss in ampacity caused the wire's resistance to increase and thus increased the current or voltage lost to heat. If the problem is, instead, that the high ambient temperatures meant that the wire would melt more easily (and that was why the ampacity was lowered; to prevent this) then a wire sized for a long run wouldn't really be impacted by this since the fuse (which would generally be sized at about half of the wire's stated ampacity) would still fry before the wire had a chance to.

In other words, I'm wondering whether, for the sake of wire sizing, I need to pretend that all of my loads are 25-40% greater than they are in order to compensate for potentially high temperatures (recall that my trailer won't have air conditioning). Or whether I can ignore wire length in this check and just make sure that the raw wire ampacity will still be sufficient to run what I want in high ambient temperatures.

Hopefully that made sense

 

[GXMnow]

At just 12 volts, a small voltage drop can be a problem and in a long wire run, it can be a big problem. So the minimum wire size is what will not get hot enough to melt. But at 200 feet, a wire could be totally safe on temperature yet be dropping 4 volts, leaving your load at 8 volts, which just won't work.

So if you have to step up to a larger wire to reduce the drop, you should still fuse for the load on the line. 5 amps on a #12 wire is just fine. but if it was very long, you may need to go to a #10, but it is still just a 5 amp line. If the motor shorts out in the fridge, the internal wires will still be kept from melting with a 5 amp fuse. At 12 volts, you should size the wires to keep the voltage drop to less than 1 volt. This is why it is good to have high current devices like the inverter close to the battery. Output of the inverter is 120 volt, so a 1 volt drop is nothing, and the current for 1200 watts is only 10 amps not 100 amps.

There is no danger of using too large of a wire. It is just heavier and costs more. Going up a gauge when building it might be cheaper than having to change it later if you do need more power for a new device later on.

There is also no danger using a smaller fuse. Set the fuse high enough that it won't blow when things are working right. The 1.25 times the constant load is a good minimum. Then go up to the next standard available size. For a short wire run, the wire needs to be able to handle the fuse rating all day long. That is the basic "Ampacity" of the wire. As the run get's longer, the wire needs to get larger to reduce the drop in voltage, but you should still leave the fuse sized for the design as if it was a short run on the original wire size. The maximum current is now limited my voltage drop, not the wire failing from heat, so we still want to protect the circuit for the design current.

On one job I worked on, they installed a lighted sign over 700 feet from the building. They had to run something like #3 AWG cable for just 30 amps at 120 volt. They had to make a splice down to #6 at the end as the #3 would not fit in the lugs on a 30 amp breaker. This is fine as the 1 foot of #6 can easily handle the 30 amps and the drop over that 1 foot is nothing. On another job, they used 12 volt light strips as wall decorations. The first panel was dim, the second was real dim, the third and beyond were not lit at all. The panels were just 60 watts each total, so the electrician used #14 wire which would have been fine for a row of eight 60 watt light bulbs. But at 12 volts, this 480 watts is actually trying to pull 40 amps. That #14 wire didn't have a chance. By the 4th panel, we measured under 3 volts. They ended up using two parallel #10 wires for both leads and you could still see the panels were each a bit dimmer down the wall. That was all they could fit since they only ran a single 1/2 conduit down the wall. They would have to rip open the drywall to do it any better. At least the 2 x #10 can take the 40 amps without overheating.

Long runs and low voltage are always a problem. Only thick wires can fix it.

 

[aaron_c]

@GXMnow Thanks!

OK, so I'm looking at one of those common wire gauge / amperage / wire length charts. The charts generally tell me what thickness of wire I need if I want to avoid more than a 3% voltage drop. So since I'm using these charts to size my wires, and since these charts are concerned with the drop in voltage rather than a loss in amperage, I should be fine using the wire size these charts indicate and not increasing the size further unless it's a short run where the wire amp rating is just enough to satisfy the device's power requirement and thus when the wire gets hot and loses amperage, it becomes unable to supply enough current.

In other words, ambient temperature impacts a wire's ability to carry current (because it will more quickly melt when the ambient temperature is hot?) but it doesn't impact the voltage. Whereas wire length impacts voltage but doesn't change how much current that wire can safely carry.

Right?

 

[Hedges]

They ended up using two parallel #10 wires for both leads and you could still see the panels were each a bit dimmer down the wall. That was all they could fit since they only ran a single 1/2 conduit down the wall. They would have to rip open the drywall to do it any better. At least the 2 x #10 can take the 40 amps without overheating.

The dimming between panels could be fixed with 2 panels on one circuit, greatly improved with more than 2 panels, by feeding positive to first panel and daisy-chaining to the rest, negative to last panel, daisy-chaining back to the first. Just like people do with parallel batteries. But now you've got 3 x #10 for one circuit, and 3 x #10 for the other, so 6 x #10 in the conduit.

A chart I have says only 5 x #10 of one type wire, but 6 x #10 for another. Not 100% clear which conduit type.

Aaron_c:
Right, length of wire doesn't figure into ampacity derating, only into voltage drop.
Twice as long a wire does generate twice as much heat, but still same number of watts per foot, each watt still has the same thermal path to escape, same temperature rise.

 

[aaron_c]

When I size fuses for these wires should I assume that the fuses will also lose ampacity in the heat?

For example I've got a water pump which claims that although it's a 24 V device, it'll need 4 amps. Seems like a lot but whatever. The wire run (there and back) will be 40 feet, so according to the wire size chart I need 10 gauge or maybe 12 gauge. I'll go with 10 to be safe.

10 gauge wire can handle a maximum of 30 amps according to the chart I'm looking at. Or maybe it's that a 30 amp fuse is the largest fuse I can use on that wire safely. In any case, I wouldn't want more than a 30 amp fuse on the wire, but that's obviously much more than I need anyhow since I'll only ever be sending a maximum of 4 amps through the wire if all goes according to plan. I could get a 5 amp fuse, but here's my concern: If the 5-amp fuse is impacted by heat the same way that wires are, that 5-amp fuse would become a 4- or even 3- amp fuse in extreme temperatures, which would mean that it would blow when I tried to use my water pump. So since my rig might be subject to high temperatures at times, it seems like I should size my fuses accordingly--at least when the wire is large enough to accommodate that. Right?

 

[Hedges]

When I size fuses for these wires should I assume that the fuses will also lose ampacity in the heat?

Yes. You can find charts on fuse/breaker trip time vs. current, also some references to heat.
I even saw something (from SquareD or Schneider, I think) about providing airflow to keep a breaker from tripping too soon.

10 awg has ampacity 40A for bundle of up to 3 current carrying wires, but max 30A fuse to be safe. Can stick with 30A unless derating for bundle or ambient drops ampacity below 30A

A 5A fuse is 25% over your max 4A. So long as it isn't fast-blow and can handle a start-up surge, that would be fine under most conditions. High ambient temperature might call for more.
At 80' round trip and 4A in 10 awg the IR drop is probably fine, but do the math if only for the exercise.

You can probably find data for ambient and and maximum, "junction" or derating for fuses and figure out how much current they can handle at your elevated temperature. If nothing else, assume linear derating of power ( proportional to I^2) from 25C to 100C.

Some fuses are dual-element, handle moderate overload longer but still blow fast for large overload.

100W isn't a lot for a pump, all depends on how much work it is doing.
A thermal shutdown is nice for a motor to protect it. A stalled motor (e.g. my A/C compressor on a hot day) will draw more current and eventually trip breaker. I'd like to also have a thermal shutdown on the motor before windings get hot enough to be damaged. I recently bought some automatic and manual thermostats to put on a transformer for fan control and disconnect.