"You're KILLING your ESC by running that low"

[xspare]

I keep seeing statements like this thread's title in the forums here. The counterpoint also comes up from time to time.

I have a decent (if fading) understanding of efficiencies and switching losses in CMOS and DC-DC converters circuits, and the claim seems like folklore to me. However, most folklore has at least some basis in reality, so...

Can anyone (EEs or otherwise) shed some light on where this claim is coming from? There's no need for wild ass guesses or intelligent speculation on this question - the answer is more likely to come from the critical mass of experience here at Helifreak.

Cheers.

[Danal Estes]

Throw a test rig together (NOT just a heli strapped down, please!) and aim an infrared thermometer at the ESC. Run it at various percentages.

Tell us what you find.

[nevercrash]

I don't completely understand the claim either, this is what I believe they are talking about.
If you are running the ESC at low throttle, it probably means you are also over gearing to get the head speed you want. This would cause more current (more heat in the ESC) than would otherwise be needed if proper gearing was used.
The ESC is most likely using 3 power MOSFETs to pulse the motor windings with square waves. These devices are driven in either complete saturation or cutoff, so there is little heat dissipation across those particular devices. When I look at the CC ESC I have, the heat sink is not even on the 3 power transistors, so without a schematic I can't really tell what components are producing all the heat. I'm interested in CC's design, but I just don't know how they were made yet. Interesting though, my CC ESC is always cold as ice no matter how hard the flight. What are other people doing to make it so warm because I've never seen it.

[brtmac]

Go listen to the Power Systems Corner segments of all of the Inside Heli and All Things That Fly podcosts. Several of them discuss the details of how an ESC works and why they can burn up if you run them at partial throttle.

In short, it comes down to efficiency. At full, or near full, throttle an ESC is pretty close to 100% efficient. As the throttle decreases the efficiency also decreases which results in more heat in the ESC. The explanation for why the efficiency decreases is explained in those Power System Corner segments.

One thing that most people fail to mention is that running an ESC at partial throttle is okay if you are also running a partial load. This is why a plane buzzing around the field at half throttle doesn't burn up it's ESC. The load it is pulling is also greatly reduced and so the heat generated in the ESC is able to be radiated off. But, when you do something like go from a 3s pack to a 4s pack on a heli without changing gearing or motor KV and adjust your throttle curve down to keep the headspeed reasonable you aren't decreasing the load. Instead, you are moving the same load down into a less efficient realm of the ESC, so you are generating a lot more heat. Think of it like lugging down an engine.

[brtmac]

Originally Posted by nevercrash

When I look at the CC ESC I have, the heat sink is not even on the 3 power transistors, so without a schematic I can't really tell what components are producing all the heat. I'm interested in CC's design, but I just don't know how they were made yet. Interesting though, my CC ESC is always cold as ice no matter how hard the flight. What are other people doing to make it so warm because I've never seen it.

I believe that the heatsink on most ESCs is on the BEC. This is especially true if it's a linear BEC, which all current CC ESCs have I believe. A linear voltage regulator burns off excess voltage as heat by design. The more current you draw the more heat it has to burn off to keep the voltage regulated. Also, the higher the input voltage the more heat it has to generate.

If you look at a ESC for a RC car/truck you'll find that they generally have huge heatsinks, often with fans, and those heatsinks are attached to the FETs. This is because they tend to run at partial throttle a lot so they need to get rid of a lot of heat. If you put a big heatsink on a heli ESC and made sure it had good airflow you should be able to run a lot more load at partial throttle without burning things up.

++Brett;

[neonbjb]

Since an ESC at lower throttle effectively applies power on less pulses to the motor, those pulses that are firing draw more current (since they have to do more work, as a motor has less resistance at full speed than it does at a low speed). The extra heat and smaller efficiency that you see people claim has to do with this; more resistance over the motor and more heat produced as a side effect. Since the net amount of energy consumed is equal to the amount of energy produced, if you have two motors turning at 50,000rpm, but one is producing two times more heat (a form of energy), then that second motor is wasting more power. Of course, in the case of less throttle, you'd have something more along the lines of one motor at 50,000rpm and the other at 40,000 with twice the heat, and then the equation becomes much harder to solve as to which is consuming more power. I generally believe though that if you run a motor at 90% and 100%, the 100% is probably consuming less power with most escs.

[Danal Estes]

Waveforms are not even close to square waves. See:

http://www.aerodesign.de/peter/2001/...l#Anker1591256

Look at the "part load" waves. That will begin to give you a glimmer of why the ESC heats more at partial.

[ping]

Aside from the heat though it doesn't lessen the longevity of the components in the ESC. Heat will always be produced in any ESC and as long as your within acceptable tolerances it doesn't matter where your at on the throttle curve.

Aircraft are a good example of this. Heli will tend to run a constant a motor at a constant RPM usually in the upper 80 percentile of your available power. A plank on the other hand especially a 3D plank will only run full throttle in short bursts. The majority of a flight will tend to be 1/4-1/2 throttle range.

[brtmac]

Originally Posted by ping

Aside from the heat though it doesn't lessen the longevity of the components in the ESC. Heat will always be produced in any ESC and as long as your within acceptable tolerances it doesn't matter where your at on the throttle curve.

Aircraft are a good example of this. Heli will tend to run a constant a motor at a constant RPM usually in the upper 80 percentile of your available power. A plank on the other hand especially a 3D plank will only run full throttle in short bursts. The majority of a flight will tend to be 1/4-1/2 throttle range.

The difference is that on a fixed pitch plane, if the power draw at max throttle doesn't burn up the ESC you should be safe running at lower throttle settings because the power draw will be considerably lower as well.

On a heli your power draw isn't controlled by your throttle alone. It's also affected by the collective. Reducing the throttle just reduces headspeed. You can still draw a large amount of power by increasing collective. This is especially true if you've done something like set up your heli so that 100% throttle would result in way too much power (ie. putting a 4s battery on a heli set up for 3s). When you reduce the throttle to control the headspeed you aren't reducing the power draw. Instead you are moving the power draw down into a less efficient realm for the ESC where you will generate a lot more heat for the same amount of power used, which is very different from what is happening when you fly a fixed pitch plane around on 1/2 or 1/4 throttle.

Another place this can bite you is if you have a plane which is over proped and draws too much power at full throttle. You can't fly it around at 1/2 or 3/4 throttle and expect your ESC to live, even though you are staying below it's max amp rating. That max amp rating is based on full throttle. At 3/4 throttle your max amp draw is probably less than 3/4 of the rated amps of the esc.

[Pinecone]

And any amount of extra heat, reduces the life of the components.

The cooler, the longer the life.

[JackdaWack]

I know that as a battery looses power, the internal resistance goes up, and the current goes down. Effectively if the current goes down that means the resistance goes up, not sure if that can be translated into a ESC type application. But if your run less current to the motor then your resistance int he ESC goes up and more power is turned into heat, and also the motor runs less efficiently as well, So you may be running less power but much more wasted power is turned into heat. The problem is most flight ESC's are designed with the intentions of high airflow, so they have small heat sinks, i would think if u weren't running atleast +90% you would want to keep the ESC in a good area that gets good airflow.... not under a tray.

[OnTheSnap]

.....

[JimLerch]

I'll want to add something, there are two situations that seemed to be getting balled up into one answer. Those situations would be:

1. A properly selected battery, motor kv, gear ratio, and rotor system
2. An alternate situation where one or more of the above are not optimal.

If your in situation 1, and you want a lower head speed, odds are your not going to hurt anything running at 60% throttle.

However, if your in situation 2, and have to run at 60% throttle to prevent an over speed, then yea, your going to end up destroying your ESC.

[OnTheSnap]

Hi Jim,
I agree that with a well designed ESC matched to you power system, you're not going to hurt anything at any throttle setting. However, at lower throttle settings , you will be generating lots of heat/wasted energy. It's a fact that heat decreases the life time of components. More heat equals a shorter life for the ESC.

[J-S-Q]

I was wondering about this very question since I started running my TRex 250 (stock setup) on a 70% throttle curve for tame indoor flying. The ESC does get a little hotter than I'd like but from what JimLerch is saying above, it should still be ok in the long run.

[OnTheSnap]

.....

[DominicD]

Alrighty let me clear up a few things

- The heatsink is primarily for the esc not the bec
- The 3 phases of the esc output have their own bridge rectifier as each phase has to allow current to pass both direction (nothing there is no return path for the current in a delta or Y configuration
- The esc is the least effecient at 50% NOTTTTTTTTTTT at low throttle points. The reason for the reduced effeciency is the increased ripple required to maintain 50% average voltage, this added ripple only serves to heat up the motor. The inductance of the motor results in a low-pass filter which filters out both the ripple and the switching frequency of the esc
- The switching frequency is independent of the throttle position so it will pass thro the linea region the same amount of time at 100% throttle and 1% throttle. The amount of time the MOS devices are in the linear region is so small it can be neglected. We are only talking about 13-18Khz here folks
- The reason people are finding their esc running hotter at lower throttle setting is due to the reduced HS. Without the built up energy in the blades as a result of a higher HS will cause the pilot to cause larger peak currents during manuvers.

[brtmac]

Originally Posted by HFG

Alrighty let me clear up a few things

- The switching frequency is independent of the throttle position so it will pass thro the linea region the same amount of time at 100% throttle and 1% throttle. The amount of time the MOS devices are in the linear region is so small it can be neglected. We are only talking about 13-18Khz here folks

If the switching frequency is the same regardless of throttle position what's actually controlling the amount of power going to the motor? My understanding is that at full throttle the FET stays open for all of the time that it's phase is energized. As you reduce the throttle it starts switching off for some of the ticks and on for the rest. So, it looks something like this (1 for on, 0 for off).

100% - 1111111111111111111111111111111111111111
90% - 111111110111111111011111111101111111110
80% - 1111011110111101111011110111101111011110
50% - 101010101010101010101010101010101010101010
10% - 1000000000100000000010000000001000000000

each of those digits represents 1 tick of the switching frequency (eg. 16khz). So, while it's true that the switching frequency is constant, the rate that the FETs actually switch on and off is not constant. At 100% it's on for the entire phase. At near 80+% it switches on and off, but the time between is fairly long so the overall switch rate is fairly low. At 50% it's switching on and off rapidly. The power loss happens during the transition between off and on. While it's in that transition state (ie. linear state) it's losing power. The time to go from off to on is very short so there is very little power loss, but the time to go from on to off is not so short, so a decent amount of power is lost as heat.

[DominicD]

Duty Cycle.

https://www.helifreak.com/showthread.php?t=64727

[OnTheSnap]

OK, thinking about this more, here is another stab at this.

A spinning motor acts as a generator, and produces "back EMF". At slower RPMs, there is less back EMF (a lower voltage produced by the motor). The ESC outputs pulses at the operating voltage (battery voltage), and at lower motor RPMs, the voltage difference between the ESC output pulses and the motor back EMF is maximized. This means that more current flows for each ESC output pulse to the motor then at higher RPMs.

The ESC output to the MOSFET is a PWM signal (pulse width modulated. AKA duty cycle controlled signal). At lower RPMs, the on interval is shorter (or a low duty cycle). So we end up with lots of short pulse with much higher instantaneous currents because of the lower back EMF (what matters is the difference between the back EMF voltage and the ESC output voltage which is the same as the battery being used). The higher currents for the short on intervals puts more stress on the MOSFETs and is a lot harder on the ESC. This produces more heat and current in the MOSFETs and maximizes the stress on the ESC. This can reduce the operational life time of the ESC or if thermal/electrical limits are exceeded it could lead to a failure.

In contrast running at high RPMs (a high duty cycle with the FETs almost always on) results in a very high back EMF voltage from the motor (closer to the battery voltage) and results is much lower currents, less heat and more efficient operation.