Wire Guages and amp limits.

[jimw1960]

Zombie, Amen, brother. Precisely the point.

 

[zombiess]

Zombie, Amen, brother. Precisely the point.

We went 7 pages and no one proved it with math so I figured I would, I needed the mental exercise anyways.

General rule of thumb, for high current runs, use the largest wire you can comfortably fit. I used 8awg and 10awg on my bike to minimize voltage drops. My current peak is a paltry 65 amps during block time.

 

[Spacey]

Explains it just fine...cheers for taking the time to write it.

 

[wesnewell]

You just don't learn, do you? It is not just voltage drop, but heat management you need to worry about at 50 amps and higher. The video showed that heat generated in the wire is proportional to the current. It was not a dead short, it was a controlled current supply. You are trying to say that wire could handle the current if something else was in the system. That is just plain wrong!!!! Amps is amps. wire has resitance. resistance causes heat in proportion to the square of the current and if you don't understand what that means you should just hold your tongue and admit you don't know.

The video is what it is. It proves that a 5' 12 awg wire heats up considerably with 130A pumped directly across it. It would heat up a lot faster with only 2 feet. Now make that wire 1000' long and do the same test. Think you'll see smoke there? Not likely, as the energy would be dispersed over a much larger area that would dissipate the heat better. How many feet of wire in a hub motors windings?
Why is everyone dodging the questions of wire sizes using 50A controllers in ebike applications, which was the original subject? Do you think 12 awg wire is not large enough for this? I'm using 16 awg in my setup and the wires never even get warm. Of course I don't short them across the battery either.

 

[jimw1960]

I made a short video of a dead short test on an A123 cell with 8ga jumper cables. After the video, I back calculated total current at 166 amps. The battery did just fine, but the wires were getting hot and starting to smell, so I ended the test after 3 min.

http://www.youtube.com/watch?v=lWk24MbYEBg

[youtube]lWk24MbYEBg[/youtube]

 

[jimw1960]

The video is what it is. It proves that a 5' 12 awg wire heats up considerably with 130A pumped directly across it. It would heat up a lot faster with only 2 feet. Now make that wire 1000' long and do the same test. Think you'll see smoke there? Not likely, ....

You still don't get it, the heat dissipation is independent of length, as zombie just showed mathematically. I'm done responding to you.

 

[auraslip]

LFP, so what are the merits of making the piss weak wire that comes off of these motors shorter?

Well, shorter wire does mean less power wasted. Even if it heats the same. My original point 5 pages back was that the shorter the wire the less power wasted. On an electric vehicle, power is limited, so you should do anything you can to reduce losses.

Also, by reducing the length of the weak wires and using thicker wire as close to the motor as possible, the heat from the thin wire will transfer to the thicker wires and dissipate quicker and easier.

 

[zombiess]

The video is what it is. It proves that a 5' 12 awg wire heats up considerably with 130A pumped directly across it. It would heat up a lot faster with only 2 feet. Now make that wire 1000' long and do the same test. Think you'll see smoke there? Not likely, as the energy would be dispersed over a much larger area that would dissipate the heat better. How many feet of wire in a hub motors windings?
Why is everyone dodging the questions of wire sizes using 50A controllers in ebike applications, which was the original subject? Do you think 12 awg wire is not large enough for this? I'm using 16 awg in my setup and the wires never even get warm. Of course I don't short them across the battery either.

Please read my post with the math, it should help you understand the concept you are confused about.

When you say 50A controllers are you talking about delivering 50A continuous right? If you want to deliver 50A continuous I would suggest 8 gauge wire since it's relatively small, flexible, pretty easy to work with and cheap. Most people only peak at 50amps on their setups, but it really depends on the motor and terrain. The general rule is use as large a conductor as reasonable and keep the lengths as short as possible.

 

[wesnewell]

The video is what it is. It proves that a 5' 12 awg wire heats up considerably with 130A pumped directly across it. It would heat up a lot faster with only 2 feet. Now make that wire 1000' long and do the same test. Think you'll see smoke there? Not likely, as the energy would be dispersed over a much larger area that would dissipate the heat better. How many feet of wire in a hub motors windings?
Why is everyone dodging the questions of wire sizes using 50A controllers in ebike applications, which was the original subject? Do you think 12 awg wire is not large enough for this? I'm using 16 awg in my setup and the wires never even get warm. Of course I don't short them across the battery either.

Please read my post with the math, it should help you understand the concept you are confused about.

When you say 50A controllers are you talking about delivering 50A continuous right? If you want to deliver 50A continuous I would suggest 8 gauge wire since it's relatively small, flexible, pretty easy to work with and cheap. Most people only peak at 50amps on their setups, but it really depends on the motor and terrain. The general rule is use as large a conductor as reasonable and keep the lengths as short as possible.

I'm really not confused about any of this, but I do have one question. I've got a 14s lipo battery and I've got 3' (6' total) 14 awg from battery to controller. My controller maxes out at 31A. Phase wires to motor are 6' 16 awg. What would I gain by replacing the 14 awg wire with 8 awg? I estimate about .5v. That's not really much of an incentive. Going from 12 awg to 8 awg with a 50A controller would be about the same .5v gain. Stll not much of an incentive. Any comments?

 

[auraslip]

I'm really not confused about any of this, but I do have one question. I've got a 14s lipo battery and I've got 3' (6' total) 14 awg from battery to controller. My controller maxes out at 31A. Phase wires to motor are 6' 16 awg. What would I gain by replacing the 14 awg wire with 8 awg? I estimate about .5v. That's not really much of an incentive. Going from 12 awg to 8 awg with a 50A controller would be about the same .5v gain. Stll not much of an incentive. Any comments?

According to my spreadsheet - power loss @ 31a with 3 feet of battery to controller and 6 feet of controller to motor is
wire gauge minimum ---- max (2.5x phase multiplication)
16gauage 55.080676w --- 155.406193w
14gauage 34.64405w --- 97.7457125w
12gauge 21.782026w --- 61.4564305w
10gauge 13.696172w --- 38.642771w
8gauge 8.132943w --- 22.94651775w


The battery current with 14s is 1643 watts... so with 16 gauge your losing at MOST almost 10% of the power your battery is putting out and 3.3% at a minimum.

At 50a?

Battery current 2650 watts
Worst case scenario Best case scenario
Power loss: 404.2825 power loss: 143.29
15% 5%

Upgrading to 8 gauge wire is the absolutely cheapest way to upgrade the performance of your bike.
Next to going on a diet of course

 

[liveforphysics]

He is trolling us guys. Nobody could be so stupid (unless he is a master-electrician).


He is implying a wire can somehow see what is causing the current in it (like a motor or light or whatever), and that it makes a difference in the heat the wire gets from passing that amount of current. Like the motor calls up the wire under load, and tells it, hey, I'm causing this current in you, so its cool, don't get warm.

Or if the wire is long, the down-stream section of wire emails the upstrean first 5ft section of wire and tells it to behave differently under a given amount of current.

We explained to him how the heating works. We explained it makes ZERO difference in what causes the current (shorted across a constant current supply or in series with any load or whatever), or how long the wire is.

He is still acting like he doesn't get it.

He is just trolling us guys. No human can be this dumb.

 

[wojtek]

i have a practical question about building battery packs. I have 20 packs of turnigy 22.2v5ah packs with 8awg wires. With that i will be building a 44v50ah pack.

So 10x 22.2v5ah connected parallel will give me 22.2v25ah.
To do that i need to connect 10x 8awg wires into one single wire - prob 4awg or so. [the setup is astro 3220 with HV160]

What is the best, cleanest way to connect 10x 8awg wires together into single wire? What is your experiences and best practice for that?

 

[gtadmin]

...What is the best, cleanest way to connect 10x 8awg wires together into single wire? What is your experiences and best practice for that?

Copper busbars. Solder your matching connectors to the copper and plug you battery discharge leads onto them. One output connector for each busbar. (Well that's how I'm doing it)

 

[wojtek]

...What is the best, cleanest way to connect 10x 8awg wires together into single wire? What is your experiences and best practice for that?

Copper busbars. Solder your matching connectors to the copper and plug you battery discharge leads onto them. One output connector for each busbar. (Well that's how I'm doing it)

something like that?
http://bluesea.com/viewresource/74

 

[wesnewell]

He is trolling us guys. Nobody could be so stupid (unless he is a master-electrician).

Not really. I'm just trying to get to a practical answer as related to the original question. 12 awg with 50A controller. I still contend that it is more than adequate for several reasons. What I'd really like to know is how long at 50A it would take to melt the insulation in a free air environment. Voltage drop compared to using 8 awg is minimal and not really a concern. With a 50A controller, it's unlikely that anyone would be drawing 50A continuously to begin with, but taking that as worse case, using a 50AH battery pack, continuous would never exceed 1 hour. I think a more practical example would be using a 25AH pack making the time 30 min.. Next, lets consider the size of wire used by the controller. Unless the controller uses 10 awg wire or larger, it makes no sense to use anything larger than 12 awg to the controller unless you have your battery pack on a trailer or something where voltage drop would be excessive. Next concern is the motor itself. What will happen to it after 30 minutes of 50A continuous to it. With a 48v system, that's 2400W into what size motor, with what size wire windings. Is 2400W for 30 minutes straight going to burn up the motor. I don't know, but I think it would burn up a lot of them. Maybe this doesn't matter to a lot of people, but it does to me. All this makes perfect sense to me why 12awg is adequate for a 50A controller on an ebike.

 

[Alan B]

Luke's test showed that 50A heats up 12 gauge to near insulation melting in some number of seconds. So if you climb a hill that can easily happen. Using it in phase wires is even worse since the current at low speed and high load can be higher than battery current.

If you are going to run 12 gauge stick to 25A controllers, or avoid hills.

 

[wesnewell]

You need to watch the video again. He used 130A and it took about 50 seconds, not just a few, to get a little smoke. He then let the wire cool to only 47C from what I can tell and then applied 50A. I never saw any more smoke but he only left it there for about a minute. FWIW, I've been using 14/16 awg for months with a 31A controller and never had the wires get even warm. There are reasons for that. One being airflow over the wires help cooling them, but mostly I don't go wot 100% of the time. And I would wager no one else does either. Besides, my battery pack is only 5ah 50V , about 250W, which means I can only go a few minutes at 31A What really pisses me off is how people imply using smaller wire will have catastrophic consequences when it's clearly not the case. There are a multitude of other factors to consider when choosing a wire size. And bigger is not always better.

 

[liveforphysics]

I will run the test again. This time on my just-unboxed super bad-ass Amrel 8-50 supply with Kelvin outputs. This power supply is so sick I'm just drooling looking at it, waiting for something to smoke with it.

 

[wesnewell]

I will run the test again. This time on my just-unboxed super bad-ass Amrel 8-50 supply with Kelvin outputs. This power supply is so sick I'm just drooling looking at it, waiting for something to smoke with it.

How about running it at 50A with a fan blowing on the wire. Unless you have a large fan, there's no way to simulate the airflow it would get on a bike going 30mph, but it would be more realistic results. Thanks.

 

[Alan B]

All right! Another test!! It would be useful to know how long a 12G lead can take 50A.

In Luke's previous second 50A experiment the wire temperature was rising, so 50A is enough to heat the wire to "soft insulation" stage where he stopped. We don't know the time since it did not start cool, but you can actually calculate it from the temperature rise rate that is observable in the video. It is not too long, probably order of a few minutes at most.

Note that it is also dependent on the temperature capability of the wire's insulation and the airflow. A high temperature insulation out in really good airflow will be somewhat better. If you use water cooled cables you can push them even farther!

A lot of folks have their controllers set for 50 amps or more, but they just use it for acceleration and the current drops off quickly. They never climb 50 amp hills slowly for several minutes (if the speed is high the current drops down). And their systems work fine, right up until they tackle that big hill that slows them way down and they keep putting 3kw into the motor when it is only doing 1kw of work and producing 2kw of heat.

So the question is, do you want to design your system for the average case, and have it fail when you hit that worst case one day. As dogman found out, just having low air in his front tire pushed him into the failure area, as did sand and rough steep climbs. Others have found out that towing a kid trailer up a hill put them into failure. Yet others have found out that climbing a dirt hill that is rough slowed them down long enough to fail items in their system, or climbing a very long paved grade. Note in many cases it was the motor that failed, putting 6 gauge wire into the controller won't protect the motor and phase leads. To do that you have to set reasonable limits and upgrade parts to hold those limits.

 

[Alan B]

The cooling from moving air has a lot of benefit at fairly low airspeed, and doesn't increase much at higher velocity, so a fan should be plenty good. Also 30 mph is only 44 fps so it is not really a high air speed, and around in close to the bike it is even less. Lots of folks put their wires in battery boxes or in tightly bundled cables, so even a still air measurement is good for many situations.

 

[liveforphysics]

I'm just sitting here loving on my Amrel 8-50. Took it all apart all ready, it's so damn nice inside. It makes the Sorenson look like a cheapo power supply. I think it could survive 300amps for a few minutes (but it's regulated to max at 50amps of course).

If any of you guys have power-supply fever, let me suggest Amrel. I've got 3 of there supplies now, and I'm just in love.

If you've got some $$$ burning a hole in your pocket, and wanting a power supply, these mofo's make the best stuff I've ever come across so far.


http://www.amrelpower.com/AMRELPowerProducts/linear_power_supply.html


These supplies are soooo sick. I will put it back together soon and do the 50amp 12awg test.


Also, this is the table for wire ampacity in open air that was included in the manual for the power supply. Pretty relevant to this thread, as it's a wiring guide for open-air applications. Notice they list 10awg at 21amps ending up with a 30degC temp rise in free air.




CURRENT CARRYING CAPACITY: As a minimum, load wiring must have a current capacity greater than
the output current rating of the power supply. This ensures that the wiring will not be damaged even if the
load is shorted. Table 2.1 shows the maximum current rating, based on 450 A/cm², for various gauges of wire
rate for 105degC operations. Operate at the maximum current rating results in an approximately 30degC
temperature rise for a wire operating in free air. Where load wiring must operate in areas with elevated
ambient temperatures or bundled with other wiring, use larger gauges or wiring rated for higher temperature.
TABLE 2.1 CURRENT CARRYING CAPACITY FOR LOAD WIRING
WIRE SIZE (AWG) MAXIMUM CURRENT (A)
2/0 303
1/0 247
1 192
2 155
4 97
6 61
8 36
10 21

 

[liveforphysics]

In Luke's previous second 50A experiment the wire temperature was rising, so 50A is enough to heat the wire to "soft insulation" stage where he stopped.

I would have kept going, but it was all ready over-boiling point (100degC), and it was gooing insulation, and making my office stinky. And that was just with the 50amp run for a minute or two after it had mostly cooled down from the 130amp test.

In my shop, I won't have to stop, I've got good ventilation, and fires are always welcome.

 

[zombiess]

Besides pimping your new super sexy power supply that will bring all the nerds to your yard, the test is pretty much semantics and entertainment (dooo iit!).

Moral of the story, bigger is better for high current. Smaller wires are fine for short higher current transients. If your wires get really warm, they are too small and you are wasting power.

 

[gtadmin]

...What is the best, cleanest way to connect 10x 8awg wires together into single wire? What is your experiences and best practice for that?

Copper busbars. Solder your matching connectors to the copper and plug you battery discharge leads onto them. One output connector for each busbar. (Well that's how I'm doing it)

something like that?
http://bluesea.com/viewresource/74

Sorta ... but I was thinking like AussieJester's or my unfinished ones