Thread Tools
Sep 11, 2011, 07:42 PM
Registered User
Thread OP
Discussion

ESC efficiency questions


I came across this interesting test that Rusty did regarding esc's and temp.

https://www.rcgroups.com/forums/show...p#post10192943

If I understand it correctly he demonstrates an ESC is less efficient as it warms up. He was no where near overtemp on the esc's they were just warm. He shows an rpm increase of 250 rpm just putting the esc in propwash. That is 5% more rpm for the same power.

Do all esc's behave this way? Less efficient at higher temps even though you are no where near the over temp limit.

My second question is related to throttle setting and efficiency. I have heard that an esc is most efficient above 80% throttle. But in a multicopter you do the majority of your flying at 50% throttle.

Does anyone know how much efficiency drops at say a 50% throttle setting vs a 90% throttle setting?

Does esc amp rating matter in regards to efficiency? For example if a drive pulls 20 amps is there any difference in losses when using a 40 amp esc vs an 80 amp esc? I expect they will both be at the same throttle setting but at different points in their ultimate capacity.
Sign up now
to remove ads between posts
Sep 11, 2011, 10:18 PM
Registered User
Wintr's Avatar
All electronic devices perform less efficiently as they heat up, so, yes, keeping an ESC cool is always a good idea.

An ESC works by chopping the current to the motor into little pieces, and energy is lost during the switching. Efficiency should be highest when the switching is the least, which is full throttle, but at full throttle, the current begins to heat the transistor junction, so it may be highest at less than full throttle.

A higher current ESC will heat up less at the same current level than a lower current one will, because the junction resistance will be lower; lower resistance = less heat. The down side of using a much larger than necessary ESC is size and weight, which may take away the small gain in efficiency.
Sep 11, 2011, 11:20 PM
Wake up, feel pulse, be happy!
Piece's Avatar
Quote:
All electronic devices perform less efficiently as they heat up
So you mean I should keep my variable-heat soldering iron as cool as possible?
Latest blog entry: Jeti ESC resto-mod
Sep 12, 2011, 01:11 AM
Registered User
Quote:
Originally Posted by TP16
So you mean I should keep my variable-heat soldering iron as cool as possible?
Of course, when you are not actually soldering. I set mine to preheat just under the temperature that rapidly discolors the tip. When I'm really ready to solder, particuliarly if it's going to be a difficult job, I'll raise the temperature, solder, then lower the temperature until I'm ready for the next joint.
Sep 12, 2011, 09:53 AM
sneu's Avatar
Quote:
Originally Posted by milo12

Does anyone know how much efficiency drops at say a 50% throttle setting vs a 90% throttle setting?
If you throttle to 50% RPM expect to see a 30-40% reduction in over all efficiency over 100%. 90% RPM efficiency will be off by 10% or so. The heat loss is shared between the motor and controller.

Steve Neu
Sep 12, 2011, 10:05 AM
Registered User
Thread OP
Thanks Steve.

Those are some big losses. I am sure the multicopter folks would jump at an esc designed for maximum efficiency at 50% throttle. Is that theoretically possible or do the basic electronics prevent that type of design? Are there any alternative esc designs out there?

I have seen some mention that the Jeti esc's were more efficient at partial throttle. I can not find any mention of this on the Jeti site. Can any one confirm if this is true or just a rumor?
Sep 12, 2011, 11:59 AM
Registered User
Quote:
Originally Posted by milo12
Those are some big losses.
At 50% throttle, a propeller load is only 12.5% of full throttle load. (Power is proportional to RPM cubed)

Although the efficiency of the ESC may decrease compared to full throttle, I fail to see how the decrease would have a significant effect on total flight time.
Quote:
Originally Posted by sneu
The heat loss is shared between the motor and controller.
If the heat loss is shared between the ESC and the motor, doesn't that imply that they both have the same losses?

I find it difficult to believe that a 100W system at 70% efficiency with a total power loss of 30W has a 15W ESC loss.

I would expect that most modern ESCs have efficiencies much greater than 90% at all throttle settings.
Last edited by Martyn McKinney; Sep 12, 2011 at 12:13 PM.
Sep 12, 2011, 12:28 PM
sneu's Avatar
I said it was split--I did not define the ratios. Here are a couple of dyno plots that are an example of what can be seen. The CC40 is a typical controller/motor used today--the other is a example of a more exotic controller that is not found in hobby use.

Steve Neu
Sep 12, 2011, 12:42 PM
Registered User
If I am interpreting those plots correctly, they are indicating SYSTEM efficiencies, not ESC efficiencies, which is the topic of this thread.

The thread below has some real life power and temperature data used to determine the ESC efficiency.

https://www.rcgroups.com/forums/show...ght=11w&page=2

I found Post #66 to have information pertinent to this discussion.

https://www.rcgroups.com/forums/show...8&postcount=66

To summarize this post's conclusions, if you had a system drawing 70W at full throttle, the speed controller would be dissipating about 1.5W.

If you increased the voltage from 2S TO 3S you would have to throttle back to 40% to get the same power out. In this case, the speed controller dissipation jumps to about 5W.

At partial throttle (40%), this is a decrease in SYSTEM efficiency of (5-1.5)/70=5% due to increased inefficiencies in the ESC.

The motor will also likely display an increase in losses, but the SYSTEM losses (due to the ESC) will increase by only 5% when using the ESC at partial throttle.
Last edited by Martyn McKinney; Sep 12, 2011 at 01:51 PM.
Sep 12, 2011, 02:38 PM
sneu's Avatar
Quote:
Originally Posted by Martyn McKinney
If I am interpreting those plots correctly, they are indicating SYSTEM efficiencies, not ESC efficiencies, which is the topic of this thread.
Might be splitting hairs but given the fact that a motor useless without a controller and controller won't do much good without a motor considering them as a system is rather relevant......

Steve Neu
Sep 12, 2011, 03:02 PM
Registered User
Thread OP
Yes I agree the system is what matters most.

Those AC Propulsion folks really know their stuff. That is an amazing improvement. Can you tell us what they are doing different? Maybe someday that technology will trickle down to us.

I am asking these questions because multicopters have a unique set of conditions that no helicopter or airplane has.

I have 6 motors and 6 esc's. Other large vehicles have 8 motors and 8 esc's. Either way that is a lot of esc's, wiring, connectors and motors to haul around. So a small loss can be multiplied by 6 or more and that happens every flight.

I would like to see a manufacturer design multicopter specific hardware. It would be a huge advantage if all 6 esc's were on one board with a heat sink, one power in and 18 motor outs. That would save significant weight and if the esc architecture was optimized for partial throttle it would be a home run.
Sep 12, 2011, 03:13 PM
Registered User
Sorry, I thought the title of the thread was "ESC efficiency questions".
Sep 12, 2011, 03:16 PM
sneu's Avatar
Quote:
Originally Posted by milo12
Yes I agree the system is what matters most.

Those AC Propulsion folks really know their stuff. That is an amazing improvement. Can you tell us what they are doing different? Maybe someday that technology will trickle down to us.
Alan designed the 9 phase controller for his solar plane--the typical model plane controllers/motors have very poor "turn down ratio" efficiency. Might be 90+% at 1000 watts but 50% at 100 watts--clearly not suitable for a long duration plane that needs 1000 watts at takeoff, but spends most of the flight at cruise of 100 watts. The 9 phase controller was a solution to the problem. The motor sees only a clean sine wave of varying voltage--it does not have to deal with the PWM from the controller--that is handled with some special high frequency transformers. The result is much improved part throttle efficiency.

Things have progressed since then and there is a new design that has a good chance of finding its way in to the hobby world. It has the advantages of the earlier 9 phase but is a simpler design--about the same foot print as the hobby controllers we have now.

Steve Neu
Last edited by sneu; Sep 12, 2011 at 03:22 PM.
Sep 12, 2011, 07:49 PM
Registered User
Wintr's Avatar
Quote:
Originally Posted by TP16
So you mean I should keep my variable-heat soldering iron as cool as possible?
Funny. Misleading.

While the heating element of your iron is an electrical device, it is not an electronic device. My comment stands.
Sep 12, 2011, 08:08 PM
sneu's Avatar
Quote:
Originally Posted by milo12
Thanks Steve.

Those are some big losses. I am sure the multicopter folks would jump at an esc designed for maximum efficiency at 50% throttle. Is that theoretically possible or do the basic electronics prevent that type of design? Are there any alternative esc designs out there?

I have seen some mention that the Jeti esc's were more efficient at partial throttle. I can not find any mention of this on the Jeti site. Can any one confirm if this is true or just a rumor?
The usual hobby controllers --Jeti included all are affected to about the same degree when throttled back. Some might be slightly better than others but it would be a small percentage difference.

The real improvement comes when yo get to a sine wave type controller which simple in concept but more difficult in practice.

Steve Neu


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Charging questions-efficiency discostu956 Batteries and Chargers 4 Oct 15, 2009 04:44 AM
Measuring amps, ESC efficiency caseih Power Systems 1 Sep 28, 2005 12:25 PM
ESC Efficiency at Low RPM? JRuggiero Power Systems 47 May 07, 2004 12:37 AM
Efficiency of Old Jomar SM-4 ESC Philipp Power Systems 3 Mar 26, 2002 07:25 AM
Brushless Motors Efficiency Questions kelvin Power Systems 6 Mar 09, 2002 09:11 AM