This thread is privately moderated by Dr Kiwi, who may elect to delete unwanted replies.
Feb 21, 2011, 02:07 AM
We want... Information!
Hastings, New Zealand
Joined Jan 2001
5,213 Posts
Quote:
 Originally Posted by A Useless Geek If they want us to buy their product the least they can do is provide some numbers that they must have generated themselves even before building a prop.
I think you will find that most prop manufacturers have not comprehensively tested all of their props, and I doubt that they have much (if any) test data to hand.

Another problem with this idea is that manufacturers could have different testing methods and/or calibration issues, so their results may not be comparable. Far better would be to collate data from independent testers (who have verifiable testing methods) and compensate for calibration issues by comparing their results for identical props.

Quote:
 I am hoping the prop OEMs can give us standardized numbers like thrust per RPM
'thrust per rpm' is not a constant. Thrust is (theoretically) proportional to rpm^2. Plastic props bend under load, so their exponents may not be exactly 2. To properly describe a prop's performance at different rpm you need a 'thrust-vs-rpm' constant and a 'thrust-vs-rpm exponent' constant.

Quote:
 Originally Posted by rgoble Heres an example graph of an APC SF 9x4.7 showing thrust and g/W vs RPM.
Could you fit a curve of thrust vs. rpm to the averages of all those points, then show the derived constants? The curve's equation would be of the form 'thrust = a*rpm^b' (where constant a = thrust.rpm.factor, and b = thrust.rpm.exponent).

As your graph shows, 'g/W' is almost worthless. The reasons for this are firstly that 'W' is the input power to the motor, not output power - so the result is dependent on the efficiency of the motor, and secondly that 'g/W' is also not a constant!

To get the true picture of a prop's thrust vs power you need to measure motor output power with a dyno (which few people have). This is a pity, because thrust vs power is a better measure of prop performance than thrust vs rpm. However, thrust vs input power is useful for comparing the performance of different motors with the same prop.
 Feb 21, 2011, 07:24 AM Jack USA, ME, Ellsworth Joined May 2008 17,619 Posts On prop RPM limits, when I got interested in that and looked for it, I found that some prop makers did provide maximum RPM limitations and others did not. For example, GWS, one of the most widely prop lines, makes no recommendations. But there are some that do and there is also a lot of ancedotal evidence and experience that can be drawn on. Below are the notes I have collected on prop limits from various RC forums: GWS Reduction Series or RS (marked EP) props = 40,000-50,000 RPM / Diameter (inches) GWS 6" to 10" Direct Drive Series (marked HD) = 80,000-100,000 RPM / Diameter (inches) GWS 5" or smaller Direct Drive/HD = much tested/used at 15,000-20,000, max safe RPM 80,000 RPM / Diameter (inches) or more APC Glow/Speed 400 Electric = 190,000 RPM / Diameter (inches) APC Thin/Folding Electric = 145,000 RPM / Diameter (inches) APC Slow Fly = 65,000 RPM / Diameter (inches) APC Racing (8.75 N,W,8.8,Series 40 Pylon) = 225,000 RPM / Diameter (inches) Master Airscrew (Windsor) = 165,000 RPM / Diameter (inches) A discussion on the Wattflyer forum on prop RPM limits: http://www2.wattflyer.com/forums/sho...91&postcount=2 //// A well known on this forum and very experienced propeller tester's opinion on prop limits: There is no official list since GWS refuses to address the question. Since a certain lawyer tells me that if I recommend anything, I can be held liable if someone gets hurt, I can take no responsibility for these numbers below! But I have tested a LOT of props and I'm giving you what I think are reasonable limits. The GWS RS series, especially the larger higher pitched ones (9-10-11x4.7, 11x7 etc), are very flimsy and I reckon they are good for only about 40,000-50,000rpm/diameter. The GWS HD series are better and most of them can cope with 80,000-100,000rpm/diameter. The tiny (<5") HD/DD props can cope with 15,000-20,000rpm...(call that 60,000-80,000rpm/diameter)... maybe a bit more. //// http://www.apcprop.com/v/html/rpm_limits.html APC Suggested RPM Limits 1. Glow Engine and Speed 400 Electric Props Maximum RPM=190,000/prop diameter (inches) (For example, a 10x6 glow engine prop should be limited to 19,000 RPM) 2. Thin Electrics and Folding Electric Props Maximum RPM=145,000/prop diameter (inches) 3. Slow Flyer props Maximum RPM=65,000/prop diameter (inches) 4. Racing Props 8.75 N,W and 8.8 series 40 Pylon props Maximum RPM=225,000/Prop diameter (inches) Note: In December 2010 it was noted that limits for Thin Electrics and Folding Electric props has been dropped from 190,000/prop diameter to 145,000/prop diameter. Those had formerly been included in category 1 with the Glow Engine and Speed 400 Electric Props and covered by the 190,000/prop diameter limit. //// http://masterairscrew.com/images/selecting_a_prop.pdf Master Airscrew limits: RPM Operating Limit = 165,000 divided by Diameter in Inches. For example, a 10” diameter prop has an operating limit of 16,500 RPM, well above the requirement of a .40 engine. //// As for the other props not mentioned above (Great Planes, E-Flite, Hoye (?), etc.) I simply compare the material in the blades and roots at the hub and think in terms of the limits I am using for the similar props above that have known limits. And when I spin a prop up in static testing or when preparing to launch, I listen to what is going on. I have not found a prop yet that will not give you an audible warning when you start nearing the limits. The bottom line? Keep yourself out of the swing line of all props at all times, that was drilled into me in the mid 50's when I was a 12 year old kid flying a Cox .049 on U-Control and it is still valid advice today. Jack
Feb 21, 2011, 09:28 AM
Curiouser and curiouser
Rochester, NY, USA
Joined Oct 2005
2,066 Posts
Quote:
 Originally Posted by Bruce Abbott ................... 'thrust per rpm' is not a constant. Thrust is (theoretically) proportional to rpm^2. Plastic props bend under load, so their exponents may not be exactly 2. To properly describe a prop's performance at different rpm you need a 'thrust-vs-rpm' constant and a 'thrust-vs-rpm exponent' constant. Could you fit a curve of thrust vs. rpm to the averages of all those points, then show the derived constants? The curve's equation would be of the form 'thrust = a*rpm^b' (where constant a = thrust.rpm.factor, and b = thrust.rpm.exponent). ...........
I have been very interested in thrust vs. rpm plots for a long time because if you have a valid plot all you need is a tachometer to find thrust - don't need a test stand, you can just do it on the plane.

Dr Kiwi has SCADS of this data and plots published.
The only variable is air density - so altitude and temp are the real variables.
Those other items, like flexing blade are a constant for any model of prop.
So if you have a plot for a specific manufacturer's prop of a given size and type you are home free.
In my opinion, the only way to go is plots from measured data - plots from theoretical info has never worked well for me.
For my a repository for just props that includes some stuff like hub date would be useful - however, I have purchased some props (GWS) that were identical EXCEPT for the hubs .

I will look for some of my links to his data and post it here.

Walt

EDIT:
Here is a place to start:
It has some of the plots.
Unfortunately his web site was deleted.
He keeps an enormous spreadsheet of his data that includes lots of info - motor data with prop data embedded in that.
At one time he sent me a copy and I sorted it to get the prop data and curves that I wanted.
Maybe he will offer to send out his spreadsheet - LOL - I just noticed that he IS the moderator of THIS thread.
Last edited by Kokopeli; Feb 21, 2011 at 09:49 AM.
 Feb 21, 2011, 09:52 AM Registered User Dayton Montgomery, Ohio, United States Joined Sep 2004 259 Posts Bruce, I will see what I can come up with tonight when I get home from work. wjbite, There is some data from some other on flybrushless, but I would say that at least 99% of it is from Dr Kiwi. Robert
 Feb 21, 2011, 09:58 AM Curiouser and curiouser Rochester, NY, USA Joined Oct 2005 2,066 Posts Yeah, I think he (Dr. Kiwi) moved his efforts to the flybrushless site. Maybe this discussion would be a better fit to the languishing THRUST vs. RPM thread: http://www.rcgroups.com/forums/showthread.php?t=973950 But looking at the flybrushless site it is so strongly oriented to motors that the prop data is REALLY buried. So I think we still need a prop data base - probably based largely on Phil's data for consistancy. BTW here is the flybrushless link: http://www.flybrushless.com/ Walt
Feb 21, 2011, 04:53 PM
Registered User
Antony (France)
Joined Sep 2003
3,120 Posts
Quote:
 Originally Posted by wjbite Yeah, I think he (Dr. Kiwi) moved his efforts to the flybrushless site. Maybe this discussion would be a better fit to the languishing THRUST vs. RPM thread: http://www.rcgroups.com/forums/showthread.php?t=973950 But looking at the flybrushless site it is so strongly oriented to motors that the prop data is REALLY buried. So I think we still need a prop data base - probably based largely on Phil's data for consistancy. BTW here is the flybrushless link: http://www.flybrushless.com/ Walt
Hi Walt
There are several prop calculators with embedded formulas "thrust vs rpm"
In Scorpion Calc there is such a calculator (with exponent 2) for a prop library
(291 props). Some data are based on Dr Kiwi results.
Louis
 Feb 21, 2011, 06:02 PM Curiouser and curiouser Rochester, NY, USA Joined Oct 2005 2,066 Posts Fordan: I am sure that that calculator is very useful for lots of people. However, my needs are much simpler. Just thrust vs. RPM is all that I need and having too wade through the complications of combining motor data with prop data is a complication that I don't need. My main use is to measure how fast my rig will spin the prop and find out how much thrust that will probably generate. Since prop thrust vs. RPM is almost (except for air density) a fixed curve any side trips through motor data is not useful especially since there seem to be big variables in the manuf. tolerances. My simple mind just wants to measure RPM and look at a curve for my prop to pick out the thrust. I would like to see a data base with JUST that info. Thanks, Walt
Feb 21, 2011, 10:06 PM
Registered User
Dayton Montgomery, Ohio, United States
Joined Sep 2004
259 Posts
Quote:
 Originally Posted by Bruce Abbott ... Could you fit a curve of thrust vs. rpm to the averages of all those points, then show the derived constants? The curve's equation would be of the form 'thrust = a*rpm^b' (where constant a = thrust.rpm.factor, and b = thrust.rpm.exponent).
Bruce,

I plotted the data in excel and added a trend line and it came up with

thrust = 0.000005*rpm^2.0835

Once I can wrap my head around how excel spit out that value I think I can add a new calculator thats exactly what wjbite is asking for.

Robert

# Images

Feb 22, 2011, 01:03 AM
Registered User
Dayton Montgomery, Ohio, United States
Joined Sep 2004
259 Posts
After dusting off the old math books and re-learning stuff I forgot ages ago I was able to figure out how to use the power law to come up with the best fit lines.

Attached are 2 screen shots of graphs for two different props. It will take me a few more days to get this to a point where everybody can use it on the site.

Robert

# Images

Feb 22, 2011, 06:33 AM
Registered User
Antony (France)
Joined Sep 2003
3,120 Posts
Quote:
 Originally Posted by wjbite Fordan: I am sure that that calculator is very useful for lots of people. However, my needs are much simpler. Just thrust vs. RPM is all that I need and having too wade through the complications of combining motor data with prop data is a complication that I don't need. My main use is to measure how fast my rig will spin the prop and find out how much thrust that will probably generate. Since prop thrust vs. RPM is almost (except for air density) a fixed curve any side trips through motor data is not useful especially since there seem to be big variables in the manuf. tolerances. My simple mind just wants to measure RPM and look at a curve for my prop to pick out the thrust. I would like to see a data base with JUST that info. Thanks, Walt
Hi Walt
The menu "propellers" in Scorpion Calc is completely independent of motor, battery, current ...
You enter (or just scroll rpm by the mouse) you get the static thrust in live
+ power absorbed by the prop
Louis
Feb 22, 2011, 09:30 AM
Curiouser and curiouser
Rochester, NY, USA
Joined Oct 2005
2,066 Posts
Quote:
 Originally Posted by Bruce Abbott ............ Could you fit a curve of thrust vs. rpm to the averages of all those points, then show the derived constants? The curve's equation would be of the form 'thrust = a*rpm^b' (where constant a = thrust.rpm.factor, and b = thrust.rpm.exponent).
I think that b = 2 works best.
I use a plotting program called GRACE to plot my measured points and using QUADRIC (second order) REGRESSION fits the data points well.

Walt
Feb 22, 2011, 09:33 AM
Curiouser and curiouser
Rochester, NY, USA
Joined Oct 2005
2,066 Posts
Quote:
 Originally Posted by Fourdan Hi Walt The menu "propellers" in Scorpion Calc is completely independent of motor, battery, current ... You enter (or just scroll rpm by the mouse) you get the static thrust in live + power absorbed by the prop Louis
Thanks Louis:
I will shut down my Linux machine and start up my Windows (eek) machine and give it a spin.
Walt
Feb 22, 2011, 10:56 PM
just Some Useless Geek
Chicagoland
Joined Oct 2008
2,549 Posts
Quote:
 Originally Posted by wjbite ...if you have a valid plot all you need is a tachometer to find thrust - don't need a test stand, you can just do it on the plane.
Sort of my point, eh? If the prop manufacturers were serious they'd have thrust/RPM numbers, measured at 760 mm, 75 (or some number) percent relative humidity, 25 degrees C. They would also give a max RPM number, but as we've already discussed, the prop OEMs are a bunch of cowards who can't be trusted to stand behind their product.

This is really stupid. I remember back in the '70s when audio equipment OEMs were making all kinds of wild claims about their power amplifier output before the FTC forced them to specify power as continuous sine wave output, low to high frequency spec, and amount of total harmonic and intermodulation distortion. Suddenly those 500 watt per channel amplifiers for home audio were putting out 35 watts per channel. Heh.

I am not immediately interested in power per thrust. That will come much later, once we've established what the prop itself will produce. Knowing what thrust props generate gives us a baseline for comparing motors on an apples-to-apples basis. Saying that motor X123 draws 10A while getting prop PDQ456 to spin at 10,000 RPM but motor Z789 only draws 8A while making prop ABC098 turn at the same 10,000 RPM tells me nothing unless I know exactly what each prop is doing.

Having all this specialized thrust measuring equipment is totally unnecessary if we can measure RPM directly -- which is much easier and cheaper to do, and totally reliable to boot. Being able to compare one power setup to another one can be done on the plane if we can measure voltage, current, and RPM directly, then plot thrust to power consumed.

Eh?
Feb 23, 2011, 07:39 AM
Jack
USA, ME, Ellsworth
Joined May 2008
17,619 Posts
Quote:
 Originally Posted by rgoble After dusting off the old math books and re-learning stuff I forgot ages ago I was able to figure out how to use the power law to come up with the best fit lines. Attached are 2 screen shots of graphs for two different props. It will take me a few more days to get this to a point where everybody can use it on the site. Robert
Robert,

I'm not very smart about the math and process that results in those graphs but, for me, those are the kinds of graphs that I would really like to see when I was contemplating motor and prop choices.

One question I have is about the larger number of points in the APC curve and the fact that the points deviate less from the line than the points in the GWS curve.

Is that a reflection of having more samples for the APC prop and there being less variation in the RPM as the prop was tested? Or is it combination of both of those things?

I think in the end, when I am deciding prop choices, one of those two props will be chosen for use based entirely on my "seat of the pants" impressions from flight testing.

I would tend to choose props for flight testing based on curves like those. And then finalize a choice with flight testing.

Jack
 Feb 23, 2011, 01:22 PM Registered User Dayton Montgomery, Ohio, United States Joined Sep 2004 259 Posts Jack, Thanks for the feed back letting me know if those graphs would be useful to you. I'm working on a way to include them on the site now. I'm not expert either but as for the differences between the two graphs I would say its a combination of several factors. Having more sample points should allow for a better estimation, and it lets you more easily see how close that is to the test results. The reason for the bigger deviation for the GWS Prop could be cause by any number of factors. For example it could be a result of differences in manufacturing. If all of the test with the APC props could were done using the same 1-2 props you would expect a tighter grouping of the numbers. Where as the GWS prop tests could of been done using a different prop for each test. Robert