Motor Wire Length

[Captglvr]

Hi All,
im doing an experiment of using a drone motor to power a FT plane. ( SunnySky V4008 600kv for a SCOUT XL)

I found a cross bracket from another manufacturer, so I can mount it on a firewall, but the motor comes no bullet points installed and with motor wires 23” long!

I’ve heard that the length of the wires to the ESC is important, that you shouldn’t use extension wires between the motor and ESC.

How do I decide where to cut and install the bullet points? Is the length different on different motors? Is there some formula for each motor?

Thanks for any help!

 

[JasonK]

length of the wires mostly doesn't matter. as far as efficiency, shorter is better, but shouldn't matter a whole lot.

 

[Merv]

...I’ve heard that the length of the wires to the ESC is important, that you shouldn’t use extension wires between the motor and ESC...

I’ve heard that also. Then I’ve heard others say it doesn’t matter, that it’s the distance between the ESC and battery that counts. When necessary, I’ve always added wire between the battery and ESC, just one less wire.

I’d make the wires whatever length needed to make it work.

 

[JasonK]

I’ve heard that the length of the wires to the ESC is important, that you shouldn’t use extension wires between the motor and ESC.

if that gets to long, the back EMF/etc that the ESC uses to track the motor position could get to noisy and/or you could induce to much noise into your other electronics. So yah, there could be a point were it gets to long, but other then that, generally shouldn't matter.

 

[Captglvr]

Excellent! Thanks All!

 

[quorneng]

Captgivr
An ESC is damaged by excessive battery wire length.
You may have noticed the capcitors on the battery side of the ESC under the shrink wrap. They are there to protect the delicate MOSFETs in the ESC from voltage spikes created by it switching the battery on and off a few hundred times a second. Excessively long battery wires causes the capacitors to be over worked. They degrade which then results in one or more MOSFETs failing so the motor stops.
There are similar pulses and volrage spikes in the motor wires but there are no delicate electronics in the motor so it has no effect.
Ovioulsy the longer the wiring the greater is the total voltage drop but if the battery has to be a long way from the motor it is always better to extend the motor wires rather than the battery wires.

I have a couple of models (they are EDFs!) where the motor is about a metre from the battery but the ESC to battery wires have been left 'as supplied'.
No ESC problems over a few years use.

 

[Captglvr]

Thanks for the detailed info!

 

[Mr NCT]

Captgivr
An ESC is damaged by excessive battery wire length.
You may have noticed the capcitors on the battery side of the ESC under the shrink wrap. They are there to protect the delicate MOSFETs in the ESC from voltage spikes created by it switching the battery on and off a few hundred times a second. Excessively long battery wires causes the capacitors to be over worked. They degrade which then results in one or more MOSFETs failing so the motor stops.
There are similar pulses and volrage spikes in the motor wires but there are no delicate electronics in the motor so it has no effect.
Ovioulsy the longer the wiring the greater is the total voltage drop but if the battery has to be a long way from the motor it is always better to extend the motor wires rather than the battery wires.

I have a couple of models (they are EDFs!) where the motor is about a metre from the battery but the ESC to battery wires have been left 'as supplied'.
No ESC problems over a few years use.

This idea, length of battery wires vs length of motor wires, seemed pretty interesting to me. I saw the same conclusion in this month's AMA magazine that longer battery wires increased pulses that needed to be handled by the capacitors in the ESC and therefore would cause the caps to meet an untimely demise. The AMA article also mentioned the length of the battery wire adding extra resistance. So I thought I'd test. BTW, I'm an electrical engineer who works with electricity/electronics in the manufacturing field on a daily basis. I have a lot of real world experience with inverters which is basically what ESCs are.

This is not a comprehensive, complete test of every motor, battery, cable & ESC combination. So please don't flame me for not being exhaustive. It's only an indication of what happens on a set up that I fly regularly - my real world conditions. Your mileage may vary.

I used 16ga. battery wire, an FT C pack motor swinging a 10x4.5 prop, a 35A FT ESC and a 2200 3s battery. The battery wire on the ESC is 3" long and the battery has a 4" wire. I did my tests at the end of the ESC wire. I have a 36" battery extension that has a resistance of 0.075 ohms per wire for a total of 0.15 ohms. I didn't use a watt/amp meter, though I did use a calibrated Fluke 289 meter and a LeCroy WaveAce 102 scope to measure the two things I was interested in, the resistance of the wire and the pulse generated.

The first test I did was battery to ESC with no extension so a total wire length of 7". I cranked the motor up to full speed and saw a 0.6 volt spike that changed frequency as the motor changed speed.

For the second test I inserted the 36" extension so now a total of 43" of battery wire. This time I saw the same spike but measuring 0.8 volt.

So, in my set up there was a difference. A difference that I'm going to worry about? Probably not. Personally if I need an extension I'll still extend the battery leads. Extending the motor wires leads to a whole new set of problems like inductive resistance - there's a reason power transmission wires are so precisely spaced from each other.

Just my 2 cents.

 

[JasonK]

The AMA article also mentioned the length of the battery wire adding extra resistance.

This is correct... however one thing to note, any increase of wire length is going to add extra resistance and my experiance has been the wires going from the battery to ESC tend to be thicker then from the ESC to the motor, so [outside of quality of wire differences] there will be more increase in resistance [looking at the full circuit battery -> ESC -> motor] by increasing the length of the ESC to motor wires over the leads to the ESC.

The first test I did was battery to ESC with no extension so a total wire length of 7". I cranked the motor up to full speed and saw a 0.6 volt spike that changed frequency as the motor changed speed.

For the second test I inserted the 36" extension so now a total of 43" of battery wire. This time I saw the same spike but measuring 0.8 volt.

Any chance you could test doing the same thing, but between the ESC and motor and see what type of behavior you get? (my first thought would be input side fluctuation to be closer to the 7" feed as you would still have the 7" feed )

Would also be curious to see what the output side spikes look like with the 2 different wire lengths.

 

[quorneng]

Mr NCT
Indeed any increase in wire length will increase resistance and thus the total power loss.
This does does not alter the fact that increases battery wire length will also increase the risk of damage to the ESC. From what I understand it is not the voltage of the spike but the energy absorbed by the capacitor to contain it and the heat created that damages the capacitor over time. It comes down to a judgement as to what level of risk of an inflight power failure is acceptable to you.

There are ways of adding additional protection to the ESC from battery wire length voltage spikes by increasing the capabilities or numbers of the capacitors.

 

[JasonK]

the level of voltage spike varying with the wire length has to do with the transition line effect of the wires between the battery and ESC (and the response speed of the battery). Generally your going to have some sort of low pass filter effect from the length of wire, so you would get more localized noise right at the ESC with longer power leads to the ESC, but I would expect you to get less noise at something else (IE if your feeding your FPV gear) connected nearer to the battery.

It is all trade offs, that cap right at the ESC is there to help smooth out the noise from the ESC and to handle the constantly changing load needs from the ESC/motor combo.

I would also be curious if the ESC -> motor would be happy with really long wires as the ESC is monitoring back EMF on the non-powered wire to know when to switch which pair of wires (and possibly polarity) to power.

 

[Mr NCT]

Mr NCT
Indeed any increase in wire length will increase resistance and thus the total power loss.
This does does not alter the fact that increases battery wire length will also increase the risk of damage to the ESC. From what I understand it is not the voltage of the spike but the energy absorbed by the capacitor to contain it and the heat created that damages the capacitor over time. It comes down to a judgement as to what level of risk of an inflight power failure is acceptable to you.

There are ways of adding additional protection to the ESC from battery wire length voltage spikes by increasing the capabilities or numbers of the capacitors.

Guess I don't understand the problem. Seems that if there's only a 0.2 volt change (1.7% change at 12vdc) in what the ESC has to deal with that seems minimal to me. Also don't get how a 0.2v increase drastically increases the energy the cap has to deal with.
Be happy to see the results of your tests.

 

[JasonK]

Guess I don't understand the problem. Seems that if there's only a 0.2 volt change (1.7% change at 12vdc) in what the ESC has to deal with that seems minimal to me. Also don't get how a 0.2v increase drastically increases the energy the cap has to deal with.
Be happy to see the results of your tests.

sounds like your generally in the same thought process as me... the "damage" risk to the ESC seems minimal and if I was really worried, I would just put some more power filtering on the input side of the ESC (as the output side of the ESC doesn't have that filtering to smooth things out and the back EMF/etc from the motor - I would expect the output side to have more noise to deal with)

 

[Mr NCT]

sounds like your generally in the same thought process as me... the "damage" risk to the ESC seems minimal and if I was really worried, I would just put some more power filtering on the input side of the ESC (as the output side of the ESC doesn't have that filtering to smooth things out and the back EMF/etc from the motor - I would expect the output side to have more noise to deal with)

It amazes me how easy it is to go down the rabbit hole on something like this. Maybe we should name it: EIRHS (excessive information rabbit hole syndrome). I'm going to leave my testing where it is. It convinced me that while it might be a real thing this pulse effect really is no danger to any of my set ups. My planes have way more to fear from my flying skills (or lack thereof) than any 0.2 volt increase in spike.

 

[quorneng]

I can only suggest you look at the solution to extended ESC to battery wire length suggested by Castle Creations.
They are a serious top end (= expensive) manufacturer who specifically suggest & provide using extra capacitor packs for over 18".

Ignore the rabbit at your peril.

 

[Mr NCT]

I can only suggest you look at the solution to extended ESC to battery wire length suggested by Castle Creations.
They are a serious top end (= expensive) manufacturer who specifically suggest & provide using extra capacitor packs for over 18".

Ignore the rabbit at your peril.

Look, do what you like. I was very up front with my test & data. Under what conditions did they see failures? What motors? Batteries? ESCs? I don't know and apparently you don't either. I can see this being an issue at voltages and current way above what I'm running but my test shows that it's a nonissue in my world. If you want to wave your arms and predict dire consequences if I trust my tests instead of listening to "the experts" - fine. Just don't wave your arms at me.
BTW, true fact, everyone who eats broccoli will eventually die. Ignore that at your peril!

 

[quorneng]

NCT
It is not the size of the voltage spike but the frequency that is of concern. The energy in each pulse has to be absorbed by the capacitor(s). It is dissipated in the form of heat which can damage a capacitor.
No one has stated that a single voltage spike kills everything instantly but it can do over a period of use from heat build up. There is no way of knowing when the capacitor will fail as there are so many variables at play.

All I can say is I have experienced a number of ESC failures in small motor installations that after seeking advice about the issue were only mitigated by a redesign of the wiring layout to keep the battery to ESC wire length as short as possible and preferably no longer than using the tails supplied.

 

[JasonK]

It is not the size of the voltage spike but the frequency that is of concern. .

The PWM switching frequency of ESCs is in the kHz range. IIRC BSHELI-S ESCs can be set between 24kHz to 48kHz [some up to 96kHz], I am not sure were a fixed wing ESC might be for that. That frequency doesn't change based on the length of the wires.

 

[quorneng]

That frequency doesn't change based on the length of the wires.

Agreed but the size of the pulse does, as NCT's tests showed, hence the energy that has to be absorbed by the capacitor and the heat generated.

 

[bwarz]

Under normal circumstances (meaning within its rated voltage) a capacitor will not heat up. Main factors of heating of a good cap are over voltage as well as ambient temps for what we are applying. Overheating causes the electrolyte to dry out, thus the failure. 'absorbing energy' is what a cap is designed to do - it doesn't cuase it to overheat. 48kHz for a cap is not really that high (detrimental) of a frequency anyways. A capacitor wants to keep the voltage steady, thus why its used to filter AC. In a simple AC rectifier, rated properly, the cap will never heat if you don't push it close to its rated voltage. Otherwise switching power supplies would melt down all the time as their switching frequencies can be well above the 200kHz range if not above the MHz range. Capacitors when rated properly (ie not exceeding the rated voltage) don't fail.

So we are back to this - not sure why the adamant argument after testing results by someone, but at rated voltages as tested, the spikes will never overheat or otherwise stress the capacitor. When noting that Castle says to do it - they are playing a game of CYA for the guy that really pushes the ESC to the limit. I'm sure they wouldn't say "if you are using this with a 3S battery blah blah blah"... They need to assume someone is going to abuse the heck out of the ESC, running it hot hot hot at abs max voltage, and thus suggest in the absolute worst case scenario.

So NCTs tests, the 0.2V on his setup would most assuredly never exceed the caps voltage rating even if running at max S rating. I myself have several twins with the battery wires extended anywhere between 10-18" with no issues (never tested as NCT did, but also don't abuse the ESCs). NCT showed that we don't need to instill such fear into everyone (or fly at your own risk and be sure you don't care if your plane crashes and burns) if they lengthen the battery wires by 10 inches.