HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Apr 17, 2005, 12:36 PM
Rocket Scientist
dtknowles's Avatar
Slidell,LA,USA
Joined Sep 2001
203 Posts
Part Throttle Motor Performance

Is it a good approximation when modelling part throttle motor performance to use a lower supply voltage to simulate the effect of reduced throttle position? For example: If I wanted to know what the efficiency of my Aveox 1406-2Y motor is at three quarters and one half throttle stick position is with my current gear drive and prop if I just look at the efficiency with 10 nicad's (3 cell Lipoly) full throttle, 8 nicad's slightly reduced throttle and 6 nicads significantly reduced throttle. The Aveox virtual test stand says 86% full throttle, 83% slightly reduced throttle and 77% significantly reduced throttle. Power out is 266, 154, and 72 watts output respectively. Taken with a grain of salt this looks really good and I need performance somewhat like this as I fly at reduced throttle a great deal. On a slightly different note, I assume that this does not account for gearbox losses and my actual perfromance will be lower. This system is flying currently and beside the fact that the cheap gearbox is not up to this task I am happy with the limited testing so far.

What about the modeling, good approach, data useless, better way???

Tim
dtknowles is offline Find More Posts by dtknowles
Reply With Quote
Sign up now
to remove ads between posts
Old Apr 17, 2005, 12:57 PM
Registered User
vintage1's Avatar
East Anglia, UK
Joined Sep 2002
29,687 Posts
Efficiency at reduced throttle - especially on a brushless is a very very moot point.

The way teh controller chops the waveform makes for hotter controllers and sometimes less efficient motors than simple theory suggests.

I have no answers here: Just a huge caveat.
vintage1 is offline Find More Posts by vintage1
Reply With Quote
Old Apr 17, 2005, 02:19 PM
Dance the skies...
Tom Frank's Avatar
United States, MA, Walpole
Joined Dec 2003
16,599 Posts
Quote:
Originally Posted by dtknowles
Is it a good approximation when modelling part throttle motor performance to use a lower supply voltage to simulate the effect of reduced throttle position? For example: If I wanted to know what the efficiency of my Aveox 1406-2Y motor is at three quarters and one half throttle stick position is with my current gear drive and prop if I just look at the efficiency with 10 nicad's (3 cell Lipoly) full throttle, 8 nicad's slightly reduced throttle and 6 nicads significantly reduced throttle. The Aveox virtual test stand says 86% full throttle, 83% slightly reduced throttle and 77% significantly reduced throttle. Power out is 266, 154, and 72 watts output respectively. Taken with a grain of salt this looks really good and I need performance somewhat like this as I fly at reduced throttle a great deal. On a slightly different note, I assume that this does not account for gearbox losses and my actual perfromance will be lower. This system is flying currently and beside the fact that the cheap gearbox is not up to this task I am happy with the limited testing so far.

What about the modeling, good approach, data useless, better way???

Tim
Tim,
You've touched on one of my "pet peeves" about motor testing, i.e. all we ever see is WOT figures, when I suspect the majority of us only fly there for very brief periods. The way I am looking at this is (hopefully to tested by Dr. Kiwi for a Hacker A20-22L very shortly) to compare the power required to spin the prop at the lower voltage, and note the rpm. Since rpm can be directly related to thrust (by test data), the lower voltage tests will yield the power required to spin the prop at that rpm, and I can relate that to my flight performance needs. So what I want to know is how many amps will my system draw when run at higher volatge but at part throttle?

It should take the same amount power (assuming motor efficiency is constant) for a pack of higher voltage to turn the same prop at the same rpm. But since the pack has higher voltage, it will produce this same rpm at part throttle and lower average amp draw. For brushless motors, the voltage to the motor is the same at part throttle (just in short pulses... not sure how this applies to brushed motors), so I can estimate how many amps it would be drawing at the reduced throttle setting from the basic power equation (P=IV). From that I estimate run time for a specific battery pack.

The big IF is the assumption that motor efficiency is constant, which we know it isn't, but there is a "plateau" on top of the motor efficiency curve where the efficiency isn't varying much (usually used to define a motors "max efficiency" amp range by the manufacturers). So if the calculated amp draw is still within this range, I suspect the error of this method will be relatively small.

As mentioned, Dr. Kiwi was going to run a test to measure amp draw at part throttle setting that can be compared to his other WOT data at various voltages for one specific motor and prop. Once that result is in, the accuracy of this approach can be tested. One benefit of this approach is that running two test points at the same rpm, one at lower voltage WOT and one at higher voltage and part throttle, the variations caused by the prop (and gearbox, if you use one) fall out of the picture, leaving only the motor peformance to impact the results. If this test were duplicated with same prop and rpm for a different motor, it would also allow some meaningful relative comparison of the two motors (something I'm struggling with right now to help select a new motor).
Tom Frank is online now Find More Posts by Tom Frank
Reply With Quote
Old Apr 17, 2005, 03:14 PM
100% electric since 1990
twest's Avatar
United States, MA, Ware
Joined Jul 2001
1,392 Posts
Steve Neu did an article a few months back on part-throttle efficiency, and to make a long story short, it don't look good. Efficiency drops pretty steeply with reduced throttle, and brushless motors suffer moreso than brushed.

The best thing to do is prop for full-throttle power that is only slightly more than you need most of the time. Don't overprop, then plan to just throttle back to half throttle (i.e. hover at half throttle). If you have way more power than you need at full throttle, try a smaller prop. Then again, I always want more power, so I tend to throttle back myself.

ESC brand has some influence on part throttle efficiency, and kontronic esc's tend to fair best (of those tested for the article a year back or so).

BTW, what gearbox are you using? I have some suggestions if you want them.

As for modeling, electricalc lets you "throttle down" in simulation, and it does report lower efficiencies, but not as low as Nue measured them.
twest is offline Find More Posts by twest
Reply With Quote
Old Apr 17, 2005, 07:14 PM
Registered User
vintage1's Avatar
East Anglia, UK
Joined Sep 2002
29,687 Posts
I am fairly sure my cansd INCREASE efficiency at part throttle. Its the only way I can account for the very low discharge I seem to get on the packs.

Thats particular to can motors though - a decent motor will tend to be optimised for best efficiency at maximum power: You can't do that with a can really as the RPM would explode the thing if you tried!
vintage1 is offline Find More Posts by vintage1
Reply With Quote
Old Apr 17, 2005, 09:44 PM
Southern Pride
everydayflyer's Avatar
Haralson County GA. USA
Joined Oct 2004
34,361 Posts
Here is a paste from an old post of mine. Not really efficency related but watts required to turn x RPMs.

I am posting data for APC SF 10X3.8 prop only because I have the most data for this one.I will list prop RPMs and number of watts required.
3,200-10
4,100-28
4,400-33
4,800-48
5,000-61
5,300-58
6,300-114
These figures were obtained while using both two and 3 cell packs and several different motors and packs of various C rating. They are real numbers obtained by real testing not a software program.In theory it should take 8 times more power(watts) to double the prop speed(RPMs) these numbers indicate that it took 11 times more to almost double it, however different electric motors operate at effeciences at different power settings. Not all battery packs are able to maintain their nominal voltage or supply the amps. the motor is wanting.There are many variables out there in the real world.

Charles
everydayflyer is offline Find More Posts by everydayflyer
Reply With Quote
Old Apr 17, 2005, 09:47 PM
Registered User
Dr Kiwi's Avatar
Chattanooga, Tennessee, United States
Joined May 2003
27,193 Posts
Tom, I'll see what I can do for you on Tuesday!

After doing a Thrust v RPM plot with an 11x3.8 APC SF on the 22L today, it may be Tuesday by the time the motor cools down from the 86C to which I subjected it!

Cheers, Phil
Dr Kiwi is offline Find More Posts by Dr Kiwi
Reply With Quote
Old Apr 17, 2005, 11:02 PM
Dance the skies...
Tom Frank's Avatar
United States, MA, Walpole
Joined Dec 2003
16,599 Posts
Quote:
Originally Posted by Dr Kiwi
Tom, I'll see what I can do for you on Tuesday!

After doing a Thrust v RPM plot with an 11x3.8 APC SF on the 22L today, it may be Tuesday by the time the motor cools down from the 86C to which I subjected it!

Cheers, Phil
That will be fine, Phil. I saw the postings on the HAcker vs the AXI 2212/26, and I'm just as interested in those results since I already have that AXI and am considering the Hacker A20-22L as an upgrade (if, in fact, it really is).
Tom Frank is online now Find More Posts by Tom Frank
Reply With Quote
Old Apr 17, 2005, 11:08 PM
Rocket Scientist
dtknowles's Avatar
Slidell,LA,USA
Joined Sep 2001
203 Posts
I had higher expectations for the group

6 posts and not a single strait answer to my question. I clearly understood that the efficiency of this combination will decrease with decreased throttle. I am operating this motor at a lower power output than its peak efficiency point. Maybe you need to consider this: If your full throttle amp draw is greater than the peak efficiency amp draw then reducing the throttle will increase your efficiency. This is what happens with most speed 400 combinations and many other can motors as we are way overdriving them. Many brushless motors have a peak efficiency amp draw that is much greater than we would tend to operate them at, so throttle back and your efficiency gets worse.

I want to know how bad it might get to help me select throttle settings and if needed change up props and gear ratios. If the efficiency gets too bad I may opt for a different motor maybe even go back to an Astro Cobalt.

I wanted to know is the data I presented reasonable, it looks too good to be true. Is the modeling approach flawed and if so how to change it.

Twest posted that Electricalc does part throttle efficiency but did not indicate how they modeled it. It might be of interest if they have a good approach, I might even be interested in getting a copy then. No point in buying a program that isnít better than what I am already doing.

Everyday Flyerís table is obviously flawed because it indicates it takes more power for 5000 RPM than 5300 RPM but at least he understood that it was so because of the way it was generated. Using power in not power out to create a prop curve is going to only produce rough results.

Twest also indicated it might be better to prop for full power only slightly more than needed for most of the flight. This will work for some. It is more complicated than that. I was looking to prop peak amp draw only slightly greater than the continuous rating of my LiPoly pack 6C 4000mah or 24 amps and I wanted to keep full throttle under 7C and fly most of the flight at around 4C for long flights and easy on the pack. Take off and climb is good at full throttle and cruise is fine at pretty low part throttle but if the efficiency was bad I would be better off at a higher throttle setting. Of course I throttle back up for maneuvers and way down for touch and goes.

Does anyone know how to calculate part throttle efficiency?

Tim
dtknowles is offline Find More Posts by dtknowles
Reply With Quote
Old Apr 17, 2005, 11:39 PM
Registered User
sneu's Avatar
United States, CA, San Diego
Joined Dec 1999
7,277 Posts
Quote:
Originally Posted by dtknowles
6 posts and not a single strait answer to my question. I clearly understood that the efficiency of this combination will decrease with decreased throttle. I am operating this motor at a lower power output than its peak efficiency point. Maybe you need to consider this: If your full throttle amp draw is greater than the peak efficiency amp draw then reducing the throttle will increase your efficiency. This is what happens with most speed 400 combinations and many other can motors as we are way overdriving them. Many brushless motors have a peak efficiency amp draw that is much greater than we would tend to operate them at, so throttle back and your efficiency gets worse.

I want to know how bad it might get to help me select throttle settings and if needed change up props and gear ratios. If the efficiency gets too bad I may opt for a different motor maybe even go back to an Astro Cobalt.

I wanted to know is the data I presented reasonable, it looks too good to be true. Is the modeling approach flawed and if so how to change it.

Twest posted that Electricalc does part throttle efficiency but did not indicate how they modeled it. It might be of interest if they have a good approach, I might even be interested in getting a copy then. No point in buying a program that isnít better than what I am already doing.

Everyday Flyerís table is obviously flawed because it indicates it takes more power for 5000 RPM than 5300 RPM but at least he understood that it was so because of the way it was generated. Using power in not power out to create a prop curve is going to only produce rough results.

Twest also indicated it might be better to prop for full power only slightly more than needed for most of the flight. This will work for some. It is more complicated than that. I was looking to prop peak amp draw only slightly greater than the continuous rating of my LiPoly pack 6C 4000mah or 24 amps and I wanted to keep full throttle under 7C and fly most of the flight at around 4C for long flights and easy on the pack. Take off and climb is good at full throttle and cruise is fine at pretty low part throttle but if the efficiency was bad I would be better off at a higher throttle setting. Of course I throttle back up for maneuvers and way down for touch and goes.

Does anyone know how to calculate part throttle efficiency?

Tim
A little to demanding??

The short answer is that you really have to measure the setup that you have with real test equipment to get anything that is useful. What happens depends on motor and controller design and software.

I also ran some tests on a quality brushed motor at part throttle and you may well be better off in terms of efficiency with it over the common low cost brushless motor.

Steve
sneu is offline Find More Posts by sneu
Reply With Quote
Old Apr 18, 2005, 09:19 PM
Rocket Scientist
dtknowles's Avatar
Slidell,LA,USA
Joined Sep 2001
203 Posts
Quote:
Originally Posted by sneu
A little to demanding??

The short answer is that you really have to measure the setup that you have with real test equipment to get anything that is useful. What happens depends on motor and controller design and software.

I also ran some tests on a quality brushed motor at part throttle and you may well be better off in terms of efficiency with it over the common low cost brushless motor.

Steve
I was kind of peeved that my message was hijacked. I apologize if I was rude or abrupt. I suspect that if they would have stayed away you would have read my original post and noticed that I was not using an inexpensive brushless motor. It is a Aveox 1406-2Y with a sensored controller. I also proposed an approach that used the Aveox calc program and reduced cell counts to simulate part throttle operation. This approach indicated very good efficiency down to less than half of full throttle power which caused me to be skeptical. I donít have a dyno but I could use the prop tables from Bob Boucherís motor handbook and a Whatt-Meter and a tach and get approximate efficiencies. Testing is always more trouble and more accurate than simulations.

I expect that I am correct in timing the motor with very little advance if I expect to fly a lot at part throttle. Even full throttle on my setup is not a lot of load for this motor (3 Lipoly cells, 8 x 6 APC Electric Prop, 3-1 gearing). It is on a 3 pound high wing aileron trainer. Pulls less than 30 amps at full throttle with a fresh charge.

I pulled off the Astro Cobalt 05 as the brushes wore out and needed to be replaced. I had just removed the Aveox from my E-Streak where it was direct drive but the Kv was kind of high (too little prop - 7 x 4) for that application and I am now trying lower Kv motor and a 7 x 6 prop in the E-Streak.

Thanks Steve

Tim
dtknowles is offline Find More Posts by dtknowles
Reply With Quote
Old Apr 19, 2005, 05:43 AM
Registered User
vintage1's Avatar
East Anglia, UK
Joined Sep 2002
29,687 Posts
Quote:
Originally Posted by dtknowles

Does anyone know how to calculate part throttle efficiency?

Tim
Yes, on a brushed, assuming its effectively running at a lower v9oltage and current.

And you know the idle current and resistance.

Losses are (idle current times applied voltage) plus (current squared times resistance.)

On a cheapo can motor to get any decent power at all they are run way over best efficiency: That comes at part throttle - usually about half throttle.

On a decent motor idle current will be higher and resistance lower. If you get into the part of the curve where idle losses dominate (very possible in a good brushless) the losses will go down as the voltage goes down, but power goes down as the square of the voltage, so efficiency can drop like a stone

However with brushless there are timing issues and controller losses to consider as well.

Steve Neu is the only man with a dynamometer that can measure true output watts versus input watts. The rest is just educated guesswork.

If you want good part throttle efficiency you need to go towards a high current lower voltage than you would otherwise use. So teh best effeiciency current is at half throtlle or so, and the overall voltage is not so high that teh idle losses cook the motor irresepective of loading.
vintage1 is offline Find More Posts by vintage1
Reply With Quote
Old Apr 19, 2005, 10:23 AM
Dieselized User
gkamysz's Avatar
Chicagoland
Joined Feb 2000
7,147 Posts
http://www.rcgroups.com/forums/showt...=part+throttle

There is no good way to model part throttle efficiency. There are simply far more variables than we have defined.

Greg
gkamysz is offline Find More Posts by gkamysz
Reply With Quote
Old Apr 19, 2005, 07:29 PM
Rocket Scientist
dtknowles's Avatar
Slidell,LA,USA
Joined Sep 2001
203 Posts
Thanks Greg for the link. I guess I should have searched before I asked. A big part of the answer I was looking for was there. I can fly around for the whole pack at half power and not worry about burning up the motor or speed control. The two flights I have made with this setup kind of confirm that. Nothing got hot even though I used extensive part throttle operation.

In both threads, the linked one and this, it seems like the posters were looking for more precision or refinements for small gains. I guess I was not clear I was looking for a ballpark number to know if I was going to be stressing my equipment.

I might still do the rough efficiency checks I discribed if I am stuck in the shop and can't go fly ;-).

Thanks

Tim
dtknowles is offline Find More Posts by dtknowles
Reply With Quote
Old Apr 20, 2005, 09:28 AM
Dieselized User
gkamysz's Avatar
Chicagoland
Joined Feb 2000
7,147 Posts
More than likely, if you are within the rated current of the ESC at full throttle, you are safe. Some controllers are borderline even in this case. But staying with somehitng well known will prevent problems.

I searched for a simple test I did around the time of that previous thread. I found that I lost about 10% efficiency by running at part throttle.

The test was something like this.
Measure I and RPM at full throttle at 7.0V.
Measure I at same RPM part throttle at 10.0V
It took ~10% more power to run at the same RPM at 10.0V and part throttle. I can't remember which motor, but I think it was a Himax outrunner and a PH25 ESC.

Greg
gkamysz is offline Find More Posts by gkamysz
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Question Motor efficiency at part throttle question shaneyee Power Systems 9 May 04, 2006 04:37 PM
Rave A look at g/W and part throttle motor performance landru Power Systems 9 Apr 21, 2005 03:42 PM
need help interpreting motor performance tests joey221 Power Systems 7 Apr 13, 2002 10:50 AM
TR389 (M100) motor performance? KatManDEW Parkflyers 0 Oct 21, 2001 09:29 AM
LS motor performance probs tderse Parkflyers 8 Apr 21, 2001 03:40 AM