ESC overheating.

[Keith Miles 2]

I was reading a thread recently where it was suggested that ESCs run hotter at low throttle than they do at high throttle and, therefore, are more likely to fail in the hands of a beginner undergoing training.

Is this another one of those received “myths” or is there, in fact, a technical explanation, preferably (with all due respect to other parties) from an electronics engineer.

I am a retired electrician with limited knowledge of electronics but this is my view on the matter, based on that limited knowledge.

Firstly,

1. All equipment should be fit for purpose and of decent quality.

2. All components i.e. battery, ESC, motor and prop should be correctly matched.

3. All components should be properly and correctly installed.

So, it seems to me that provided all of the above conditions are met there should be no need to worry about the possible failure of an ESC, or anything else, due to maintaining any continuous throttle setting be it low, high or anywhere in between!

Secondly, and I’m happy to be re-educated here, it is my understanding that all electrical circuits generate heat, to a greater or lesser degree, and the materials used are designed accordingly to dissipate it. The primary factor here, being the current although there are others.

It has been suggested, in the “continuous low throttle” case that the ESC is “working harder” at low throttle than high throttle and this is what puzzles me.

An ESC, as I understand it, is a switching device which sends pulses to the motor to control speed. The greater the pulses the greater the speed, the lesser the pulses the lesser the speed. The switching process itself generates heat.

A tiny computer CPU is, similarly, a switch and despite operating on much smaller voltages and currents has to be mounted on a heat sink with thermal paste together with a large fan to cool it!

It seems to me, therefore, that when the ESC is switching at a lower frequency the switching process, at least, will create less heat not more, likewise the proportional reduction in motor current.

So, if low throttle does mean more heat what is causing it other than reduced airflow?

Even then, I would refer back to the three points made above.

 

Edited By Keith Miles 2 on 08/01/2020 11:40:18

[Nigel R]

Zero throttle involve no switching and no power, so lets forget that one.

Full throttle involves switching only to achieve commutation. I cannot be certain but I have this dim recollection that the switching occurs at points where the current flow would naturally be low. I stand ready to be corrected on that point. If I am remembering right, this therefore represents very little wasted heat.

In between involves wasted heat during the extra switching required to achieve the PWM control of power. Current flow is likely to be at its highest at the point of switching. Maximum wasted heat.

Either way, any reasonable install will have sufficient cooling and it won't be a problem.

[Keith Miles 2]

Cheers, Nigel.

Yes, and I can understand likely brief transients when changing throttle settings.

The notion, however that the ESC itself is “working harder” at lower constant power settings thereby creating more heat seems a bit back to front to me and certainly counter intuitive!

Something was said along the lines of the ESC having to dump excess power as if the ESC had current to spare which, of course, isn’t the case because the motor determines the value of the current draw from the battery and through the ESC.

If there is a technical explanation that counters my instincts, I would be most interested to receive it!

I also recently discovered, during a conversation, that some people (even experienced flyers) seem to have been taking (and might still be taking) the ESC current rating as the target value when testing or changing props as opposed to the maximum current rating of the motor and where the motor value is not known or clearly stated and is being assumed to be the same as the ESC!

As I was saying to someone the other day, there are far more complexities and more traps, even dangers, for the unwary compared to IC power and, therefore much greater understanding required unless you are a member of the “just buy it and fly it” brigade!

😊

 

Edited By Keith Miles 2 on 08/01/2020 13:10:56

[Romeo Whisky]

One point to note is that there are two primary types of brushless speed controller - linear and switch-mode.

Most (though not all) larger (40A+) amp ESCs these days are switch-mode; most smaller ones (<=25A) are still linear. Those in between may be either and it's worth buying switch-mode for preference.

Linear ESCs run hotter than switch-mode and usually incorporate a larger heat-sink. It is also true that all ESCs can get warm in use, so most incorporate a thermal cut-out, and you REALLY want to avoid that happening in flight so you need to ensure adequate air-flow over the ESC when the model is in flight.

It is also true that ESCs have to dissipate more heat when running at half throttle, and this is one reason why it important to use an ESC with plently of ampage headroom (ie. suggest 25% bigger than the maximum likely amps draw at full throttle).

Hope that helps.

[Alan Gorham_]

Linear and switch mode just refers to the design of the BEC circuit fitted to the ESC and is nothing to do with the motor control FETs.

[Nigel R]

" technical explanation"

As said, to the best of my knowledge, its simply that you are switching during periods of high current flow. The resistance of the FETs is why the waste heat comes from the ESC.

The switching at full throttle, if I remember right, happens at points of low current (or it may be at zero voltage across the FET, I may be mixing it up). But, I believe it is done at point where volts x amps = zero, so, no waste heat is produced.

[Nigel R]

Theoretically, you could have linear control of a brushless ESC. It would add pointless complexity to the circuit though, probably another stage of FET, and require massive heatsinking, probably to the point of rendering the thing utterly nonviable.

The ability to PWM control power comes only at the cost of additional software complexity in the ESC's CPU, where control of the brushless commutation is already being done.

[FlyinBrian]

I don't know the answer as to whether more heat is generated at reduced throttle but as the majority of a models flight is at less than WOT it would seem that ESCs are designed to cope with this,

[Nigel R]

Indeed so. Ensure they receive reasonable airflow, problem solved.

[Alan Gorham_]

Posted by Nigel R on 08/01/2020 12:49:52:

Theoretically, you could have linear control of a brushless ESC. It would add pointless complexity to the circuit though, probably another stage of FET, and require massive heatsinking, probably to the point of rendering the thing utterly nonviable.

The ability to PWM control power comes only at the cost of additional software complexity in the ESC's CPU, where control of the brushless commutation is already being done.

That is not strictly true for every ESC as there are dedicated BLDC commutation IC's available that allow workload to be taken away from the Microcontroller.

[Dickw]

Posted by Keith Miles 2 on 08/01/2020 11:35:42:

I was reading a thread recently where it was suggested that ESCs run hotter at low throttle than they do at high throttle and, therefore, are more likely to fail in the hands of a beginner undergoing training.

Is this another one of those received “myths” or is there, in fact, a technical explanation, preferably (with all due respect to other parties) from an electronics engineer.

....................................

It has been suggested, in the “continuous low throttle” case that the ESC is “working harder” at low throttle than high throttle and this is what puzzles me.

An ESC, as I understand it, is a switching device which sends pulses to the motor to control speed. The greater the pulses the greater the speed, the lesser the pulses the lesser the speed. The switching process itself generates heat.

..............................

It seems to me, therefore, that when the ESC is switching at a lower frequency the switching process, at least, will create less heat not more, likewise the proportional reduction in motor current.

So, if low throttle does mean more heat what is causing it other than reduced airflow?

............................

Edited By Keith Miles 2 on 08/01/2020 11:40:18

Our motors generate a “back emf” when running, and the higher the rpm the higher the back emf.

The voltage available to drive current through the motor is the difference between the supply voltage and the back emf.

Therefore, at any given supply voltage the current will be higher at low rpm than at high rpm.

To throttle the motor the ESC sends shorter pulses with longer gaps between to the motor – so a normal meter, which effectively “averages” the current, will show a reduced current BUT

the instantaneous current in those short pulse will actually be higher than full throttle current – that may or may not be a problem for the FETs.

Dick

[Keith Miles 2]

Dick and Nigel,

Many thanks for that. Much of it is along the lines of what I was thinking together with some additional info.

I understand back emf but I hadn’t, I confess, given it much thought in terms of the effect on current. Got it now!

I think that we can agree then, all that said, with the three key points made in my original post about component quality, matching and installation!

Having to worry about throttle settings when flying seems a bit ridiculous to me and not very practical!

And for all the reasons given, the idea that slow flying beginners are more susceptible to ESC failures seems to me to be ignoring where the problem really lies and failing to resolve it is hardly helpful to the student either!

Edited By Keith Miles 2 on 08/01/2020 13:31:04

[Chris Walby]

work is work, if you are driving a prop at increasingly higher speeds (within reason) then it must by the laws of physics consuming more power.

The twist to this is how stressed and efficient the ESC is at various speed settings. As Dick has eluded to, at lower speeds the instantaneous current through the FET's might be higher thus placing more stress on the FET device + it will be less efficient and produce proportionally more heat. This might explain why ESC's tend to fail (if not overloaded) at lower speed settings and why they get warmer.

At WOT and increased airspeed there will be more air passing over the ESC (assuming its ventilated) to cool the additional heat generated hence the combined effect of lower air flow and lower ESC efficiency places the ESC under greater stress when at low throttle settings. Does that sound logical Spock?

[Nigel R]

Posted by Alan Gorham_ on 08/01/2020 12:55:54:

That is not strictly true for every ESC as there are dedicated BLDC commutation IC's available that allow workload to be taken away from the Microcontroller.

What sort of ESCs are using separate ICs for the commutation?

Just intrigued, as it seems a relatively well solved problem in ESC software.

It would seem tough to avoid having a cheap and cheerful CPU, as it can mop up the other tasks of servo control input decoding and dealing with storing/applying the various settings within the ESC.

[Keith Miles 2]

Do we agree, then, that ESC technology is not inherently flawed and prone to failure in normal use and within the design limitations of a given ESC?

[Denis Watkins]

This is why we go up with a higher W value ESC

Perhaps why Century props went down a gear

And why we should throttle up and down in flight, into turns and manoeuvres

Initial leccy model builds, in the olden days, had the ESC on the outside of the model

From Brushed and heavy packs of old, we have come a long way surely

[Jesus Cardin]

The only ESC I have burned in my life was checking the turning direction of a motor without a propeller and at low speed.

It was with a proven set and the ESC smoked at once, without a second to cut!

Edited By Jesus Cardin on 08/01/2020 16:21:01

[Keith Miles 2]

Posted by Denis Watkins on 08/01/2020 16:06:59:

This is why we go up with a higher W value ESC

Perhaps why Century props went down a gear

And why we should throttle up and down in flight, into turns and manoeuvres

Initial leccy model builds, in the olden days, had the ESC on the outside of the model

From Brushed and heavy packs of old, we have come a long way surely

Agree on points one and two.

Throttling up and down is, surely, a flying issue not an ESC overheating issue?

And, yes, we have come a long way which is why I don’t buy the argument that you need to fly in a particular way to avoid an ESC failure.

Do electric cars and full size aircraft use similar technology?

If, so, and we accept the “cautionary throttle operation” theory, I see a big problem there!

Edited By Keith Miles 2 on 08/01/2020 20:50:49

[PatMc]

Posted by Keith Miles 2 on 08/01/2020 20:36:37:

Posted by Denis Watkins on 08/01/2020 16:06:59:

This is why we go up with a higher W value ESC

Perhaps why Century props went down a gear

And why we should throttle up and down in flight, into turns and manoeuvres

Initial leccy model builds, in the olden days, had the ESC on the outside of the model

From Brushed and heavy packs of old, we have come a long way surely

Agree on points one and two.

Throttling up and down is, surely, a flying issue not an ESC overheating issue?

And, yes, we have come a long way which is why I don’t buy the argument that you need to fly in a particular way to avoid an ESC failure.

Do electric cars and full size aircraft use similar technology?

If, so, and we accept the “cautionary throttle operation” theory, I see a big problem there!

Edited By Keith Miles 2 on 08/01/2020 20:50:49

ESC's are not limited by Watts but by current & voltage discretely not by the combination.

[Martin McIntosh]

I bought a YEP 120A esc and it went up in smoke straight away at just 20A. The replacement one seems fine. Also one of the 55A ones on a twin died on take off, second flight only, too late to save the model.

I suppose that the majority of my models are now electric but give me i/c any day. If they blow up they are easily repaired, not an expensive total wipe out.

By the way, hardly any of my esc`s have any cooling airflow whatsoever due to restrictions imposed by the models (Lanc. and Vulcan for instance, all buried in the fus.), but I do go for rather over rated ones for the applications.

[Simon Chaddock]

Keith

Don't forget it is not unknown for ESC manufacturers to 'gild the lily' a bit as far as the specification max Amps.

Hence the advice to size the ESC to at least 20% over the maximum possible actual amps, say fresh batteries and a panic snap opening to full throttle?

It only take a fraction of a second of serious over load to 'kill' electronics.

[Dickw]

Actually not all ESCs are that delicate.

The attached pic shows a 150 amp ESC and the results of hitting 380 amp peaks in a series of short (2 - 3 seconds) motor runs on a very hot day. It was OK with 370 amp peaks earlier in the morning when it was cooler, but once the batteries warmed up and the current increased at the same time as cooling decreased ...........!

I now use a 135 amp ESC at 200 amp peaks in the same way - it helps if you can monitor ESC temp (at least in initial flights) to make sure all is well.

Dick

[Keith Miles 2]

Posted by PatMc on 08/01/2020 21:29:51:

Posted by Keith Miles 2 on 08/01/2020 20:36:37:

Posted by Denis Watkins on 08/01/2020 16:06:59:

This is why we go up with a higher W value ESC

Perhaps why Century props went down a gear

And why we should throttle up and down in flight, into turns and manoeuvres

Initial leccy model builds, in the olden days, had the ESC on the outside of the model

From Brushed and heavy packs of old, we have come a long way surely

Agree on points one and two.

Throttling up and down is, surely, a flying issue not an ESC overheating issue?

And, yes, we have come a long way which is why I don’t buy the argument that you need to fly in a particular way to avoid an ESC failure.

Do electric cars and full size aircraft use similar technology?

If, so, and we accept the “cautionary throttle operation” theory, I see a big problem there!

Edited By Keith Miles 2 on 08/01/2020 20:50:49

ESC's are not limited by Watts but by current & voltage discretely not by the combination.

To whom does that last line belong, it ain’t mine?

And yes, I am fully aware of how ESCs are rated.

By current. In Amps!

And current x voltage IS watts.

I’m an electrician!

Looks like a posting mix-up here!

Very confusing!

 

Edited By Keith Miles 2 on 08/01/2020 23:39:49

[Keith Miles 2]

Posted by Simon Chaddock on 08/01/2020 22:08:04:

Keith

Don't forget it is not unknown for ESC manufacturers to 'gild the lily' a bit as far as the specification max Amps.

Hence the advice to size the ESC to at least 20% over the maximum possible actual amps, say fresh batteries and a panic snap opening to full throttle?

It only take a fraction of a second of serious over load to 'kill' electronics.

Yes, indeed, and it doesn’t just apply to ESCs!

On the “headroom” point I have raised elsewhere my recent discovery that some people may have been under the impression, (especially those with pre-supplied combos with unmarked motors or “buy it and fly it” models) that the ESC rating is the target current!

Took me quite a while to convince one very experienced flyer otherwise!

Makes you think how many more have the same misunderstanding!

Edited By Keith Miles 2 on 08/01/2020 23:48:57

[Keith Miles 2]

DickW,

May I ask the purpose of your incendiary experiment?