Author Topic: Propeller Calculations---more than you want to know!-Pointy head alert! (Read 4471 times)

Alan Hahn · « **on:** December 06, 2007, 06:54:22 PM »

I have been playing around with a software program called "Prop Calc", http://www.drivecalc.de/PropCalc/index.html
I have some questions about the calculations into the developer, so I am not sure of what physics is actually in these calculations, but I was curious about how a particular propeller might vary about some known points. Since Prop Calc will calculate how thrust and power for a fixed rpm (read governor mode) vary as a function of airspeed (our typical maneuvering about a fixed level flight airspeed), it seemed useful to change some prop parameters just to see how thrust might vary. In the cases here I was interested in a 54 mph (24m/s) airspeed that my E-Nobler Arf flies at, and what happens when I pull the nose up.
As a bit more background here is a graphic on how my airspeed varies when I go into the reverse wingover. The airspeed measurement from my EagleTree data recorder is a little suspect in absolute magnitude (it says I'm flying at an average ~46mph, whereas lap speed and line radius from the same plot tells me I am at 54 mph. However what I notice, is that right after the snap into the wingover, the plane decelerates quite fast (almost a free fall) until it reaches roughly 1/2 way up the hemisphere. After that the plane begins to accelerate. This is all with a constant rpm mode ESC (and the rpm really is staying constant according to the readback). In the first plot, Altitude is the red trace (feet), the green is airspeed (mph), the light blue is a calculated acceleration from the speed trace (in m/s^2),and the violet trace is power supplied from the battery. Sorry for the mixture of english and metric units. Free Fall acceleration is -9.8 m/s^2 for reference. Also recall that I am unsure about the absolute magnitudes of airpseed (and also acceleration) and the altitude. Haven't completely got those instruments calibrated or installed correctly. The altitude is primarily useful as a marker for when the maneuver starts. The wingover starts at ~77s on the horizontal axis. You can see the deceleration peaks just before this time, must be when I snap that corner. The peak value is ~-15m/s^2 (slows down). You can see that when the altitude trace (red) is a little less than half way up, the acceleration is =0 and soom after the plane is accelerating (gaining back lost airspeed). The acceleration is coming of course because at a fixed propeller rpm, as the airspeed drops, the prop gets a bigger bite into the air, and the thrust increases (also power needed increases too).

So what I am interested in is for a fixed type of prop, how can I optimize the thrust as the airspeed drops, but at the same time not go overboard and waste a lot of my precious battery power thrashing the air during the straight and level part of the flight. The faster thrust would increase as the airspeed drops, the quicker the acceleration would move back into the positive range (and the plane would accelerate.

The next entry will talk about the prop calculations.

Alan Hahn · « **Reply #1 on:** December 06, 2007, 07:15:01 PM »

Ok, here are the prop calculations from Prop calc. To start with, they don't have my prop (APC 11x5.5 E prop) in their database, so I chose what I thought wouldn't be too far off--a 2 blade Aeronaut 10.5x6 prop (I have use a Aeronaut 10x6 prop on the E-SuperClown). The variations I tried were, a 3 blade version of this prop, a version of the 2 blade where the pitch angle is increase by 4 degrees, and a version where the 2 blades are simply scaled up to 11.5 inches. This includes scaling the chord too.

What I did was to find the thrust (~4.5N --about a pound in weight) of the original 2 blade prop at 24 m/s (my 54 mph) and with a rpm of 9000---roughly equal to what I fly the E-Noble at. Then for each variant, I adjusted the rpm until that variant's thrust more or less equalled the original value at the same 24 m/s (which by the way gives me ~5.0s laps on 62 foot radius flying circle (includes my arm length). Then everything is plotted as airspeed is varied. The steeper traces indicate a thrust which increases faster as airspeed drops, and thus provides more acceleration to the plane. Put another way that prop will load the motor more than the other props (which is what I want of course). At the same time I don't want the power needed at 24m/s to be much higher than the original prop, because I pay (literally) for that by a heavier and more expensive higher capacity battery.

So what I show here is that diameter seems to trump (if I want power to not change much right at 24m/s). At the same time I was a little surprised that the higher pitch prop loses. I had thought that if I had the power available that there would be no difference (the higher pitch runs a lower rpm). Anyway I am still absorbing this, and still have some questions about the calculations (I am surprised how the 3 blade prop needs more power---I don't disagree necessarily, but I would like to understand how the program puts a multiblade inefficiency in--I am actually making some other calculations to see what the program gives, and I am a little confused by what it is telling me. But that is another story. Anyway, here is a plot of the above calculation for your entertainment.

Oh, almost forgot, I also added a calculation which gives the thrust and power when the rpm is increased by 200 rpm (to 9200), and indicate the thrust increase for a 1000 rpm increase (to 10000 rpm) to give you an idea how thrust at a fixed airspeed depends on rpm. I also show you the thrust needed at 24m/s to accelerate my ENobler (44oz) by 1g ( which of course would just compensate gravity when the nose is pointed vertically).

Maybe more than you really want to see.

Thrust is the lower group of traces, power the upper traces

Dean Pappas · « **Reply #2 on:** December 07, 2007, 09:01:13 AM »

Hi Alan,

AHA! To quote you, "So what I show here is that diameter seems to trump ".

That's right, assuming you have a constant RPM (or even close) then the bigger the prop diameter and the lighter the airplane the more constant the actual airspeed will be. Pushing a more massive column of air (and I'm picking the term mass carefully here) through a smaller change in velocity in order to get the same level flight thrust produces a system with a bigger thrust versus change in airspeed slope. We would like that slope to be dead vertical, but that isn't practical.

On another note, a few days ago, on another thread, someone asked if more prop and motor rotational inertia would help carry an airplane through corners. It might, but it is far more important for the RPM governor to sense a small change quickly and add more voltage and current. That makes the added power come on more quickly. Minimizing the inertia would probably be best.

Your droop in your airspeed readings seems to almost lead the altitude changes, which surprises me. Perhaps you are getting errors in the reading due to the changing angle of attack at the pitot? I once saw a full-scale system that put the pitot in an "arrow on a gimbal" to eliminate the tilt-induced error.

I need to go dig up the basic prop power and thrust relationships ... I'll post to this thread later.

take care,
Dean

Alan Hahn · « **Reply #3 on:** December 07, 2007, 02:07:10 PM »

Dean,
I am very suspicious of the airspeed and altimeter readings, mainly because they are not quite on in simple straight and level flight. So for now I use them with some caution--more to indicate relative changes. I had notice that the blip precedes the altimeter change, but I have also some indications that the altimeter is sensitive to some type of orientation. I note that my outside loops don't show the bottom altitude as cleanly as the inside loops. Now I know I tend to miss the outside bottoms more than the inside, but this is a big effect. I am guessing that the belly pan (where I have placed the actual altimeter and airspeed sensors) is being pressurized a little on outside maneuvers. I need to seal it off better I think. So my data isn't as clean as I would like, and I am not positive how to get it setup right. It may be that CL aerobatics is tough on things like Pitot tubes and altimeters, because as you point out, the rapid attitude changes can do some strange things to the pickups.

I did hear back from the PropCalc developer, and he does state the multiblade inefficiency is NOT coming from some "unclean" air factor. As he pints out, as long as the plane is in reasonable motion, each blade is cutting into clean air. Instead he gives other reasons for the effect, but I haven't completely absorbed his explanations yet---and would probably butcher his explanation if I tried to give it right now.

But as everyone knows, diameter is king as long as the torque is there in the engine/motor AND the prop can clear the ground on takeoff. That's where the rub is for me. I guess I could make the ENobler into the E-Nobler_Stork with long legs, but a multiblade prop still may pay off in getting more thrust faster as airspeed falls. Unless I can figure out a way to have a sensor unambiguously determine when exactly to bump the throttle setpoint by a few hundred rpm and how long to hold it after I bang (more gently!) that corner. That's the other way to get that extra thrust.

Of course everyone should know that I find this stuff enjoyable, which is why I am wasting my time for you!

Igor Burger · « **Reply #4 on:** December 07, 2007, 03:12:29 PM »

Alan, I think the POWER is that what makes the slope better, not the diameter. You can change number of blades up, width of blades, diameter of blades or lower pitch, all will lead you to better slope, and at the same time it will cost you more power. The only trick which can change, is airfoil shape which will move max efficiency of pprop little bit up or downt.

Alan Hahn · « **Reply #5 on:** December 07, 2007, 06:07:56 PM »

Igor,

I agree with you.

No doubt, what I want (probably everyone does) is a propeller which "slips" along at low power in straight and level flight, then when you pull the nose up and airspeed begins to slow, you want that prop to bite into the air to provide the thrust to fight gravity and any airspeed loss in the increased drag when you pull up (or down). That clearly implies your motor needs extra power to give that thrust. My feeling or hope is that these power "blips" are good, yet don't dominate the total power needed for the entire pattern, which I basically think is a to a large extent dominated by the drag (and the total time spent) in level flight.

However in the real world, for most of us, our access to propellers is only what is out there in the market place. There are a lot of 2 blades, but fewer 3 blades. On my E-planes, I am at the diameter limit (10" for Super Clown and 11" for the Nobler), unless I lengthen the landing gear or increase the wheel diameter. I think multiblades (or wider blades perhaps) is the only way to go. I am thinking about how to cut up one of the APC 11" thin electrics to make a 3 blade version. Maybe the smoothness of an electric motor will allow me to make this thing safely. I have all winter to figure it out until reality collides with theory next spring!

Dean Pappas · « **Reply #6 on:** December 11, 2007, 08:54:07 AM »

Hi Alan,
to quote, "I guess I could make the ENobler into the E-Nobler_Stork with long legs, but a multiblade prop still may pay off in getting more thrust faster as airspeed falls. "

I had to practically beat Hunt senseless on this one ... make friends with the stork. Prop development will soon take the form of how much diameter we can gain with blade shape and airfoil changes, combined with how much static current we can tolerate (albeit briefly) and how much GP is induced at big diameters. I'm betting on light, thin airfoils, and medium blade widths.

have fun!,
Dean

Rudy Taube · « **Reply #7 on:** December 11, 2007, 11:16:26 PM »

Hi Dean,

I'll bet the "beatings" were severe. Bob probably has: Prop=10" dia, tattooed on his forehead after 10,000+ flights with them!

I agree, "love the stork" (for now). I think the easiest way to get there is to use one of Randy Smith's CF TALL LG. I fly off rough grass so I just put it on backwards (it normally slopes to the rear), this works out great, and gives plenty of room for my 12"x6" APC-E prop. On my E P-40 I just bent up some HD Looong LG from 5/32" MW. Heavy, Stork like, but it works.

Hi Alan,

Please keep us posted with your multi blade experiments. FWIW you may want to try making a 4 blade by simply bolting on 2 two blade props with some sort of "key" to keep them at 90 degrees. The keyed hub can still be thicker than just one (for more strength?) and still fit on the shaft. This would allow some interesting E props to be used in the 9" to 10" size? I sure hope you show us the way on this. I think we may be on our own on this for awhile, my guess is that APC and Bolly do not have ECL on their "speed dial".

Regards,

Dean Pappas · « **Reply #8 on:** December 12, 2007, 09:47:24 PM »

Rudy,
How about titanium wire? Worked on Pattern stuff, replacing 5/32" with 4.5 mm dia. titanium.

later,
Dean

Rudy Taube · « **Reply #9 on:** December 13, 2007, 01:29:49 AM »

Quote from: Dean Pappas on December 12, 2007, 09:47:24 PM

Rudy,
How about titanium wire? Worked on Pattern stuff, replacing 5/32" with 4.5 mm dia. titanium.

later,
Dean

Hi Dean,

Do you have a source for 4 mm (same size as 5/32") Titanium? My 5/32" gear added 1.85 oz over the 1 1/2" shorter, thinner, 1/8" stock gear. Tit. Price? $20/ft. or ??

I was thinking of cutting a large CF gear in half and screwing ea half into the wing mounts on my E P-40, but your Tit. LG might be lighter and look better.

Thanks,

Dean Pappas · « **Reply #10 on:** December 13, 2007, 09:39:05 AM »

How about Small parts Inc?
http://www.smallparts.com/products/descriptions/tiw.cfm

You are interested in retro-fitting an existing LG installation, but for the case where you have a clean sheet of paper ...

Try finding almost anything written about pattern guy Nat Penton. He has been using carbon fiber struts with a small bit of wire JB welded into the end for Pattern plane landing gear, at 9 ~ 10 lbs. Should work for us.
Also there is a Germnan guy named Michael Ramel (the guy with the contra-rotating gearbox too) who sells retracts with carbon strutes. He rubber mounts the retract mechanism to preserve the relatively stiff but light carbon strut. Neat idea!

later,
Dean

Mark Scarborough · « **Reply #11 on:** December 13, 2007, 12:39:02 PM »

Ok so how is this stuff, this Titanium, to work? bending radius , amount of effort to bend for forming? My 109 is going to hav erather long storky legs so I am interested in this.

Dean Pappas · « **Reply #12 on:** December 13, 2007, 05:27:19 PM »

Hi Gang,
It's a pain but it works. You probably ought to spens some time on google and, or the building forum asking if there are any Stunt junkies out there who really know about working titanium. I used it once without problems, but I assume it was dumb luck.
Dean

Rudy Taube · « **Reply #13 on:** December 13, 2007, 11:38:13 PM »

Hi Dean,

Thanks for your help. Good idea.

Sorry Alan, I did not mean to get your prop thread off track. At least it has a little to do with props. ;-) We are all going to have to deal with longer LG until we get some 3 or 4 blades that match our ECL needs.

I asked the Titanium LG ? on the Open forum.

Alan Hahn · « **Reply #14 on:** December 15, 2007, 09:02:27 AM »

Rudy,
No problem---since we are now about to embark on ECL Storks!

By the way, when I flew my E-Super Clown at the Sig CL contest last summer, I did talk to a Carrier guy (forgot to get his name) who also suggested doubling up two ACP E props to make a 4 blader. The only problem he (and I) had is that the prop adapter for the Scorpion motor is quite stubby--I can't even get a normal spinner on it and still be able to tighten the prop nut. They have recently come out with a longer threaded stub, but I am not sure that even it will fit two props on simultaneously.

By the way, I look at ECL development now as going two ways--and this observation isn't a criticism or either course. The "Classical" (my definition of course!) approach follows the standard modern CL approach, that is to throw the equivalent of Cubic inches (or cc's) at the issue. This direction then involves building quite light airframes to make up the weight difference. The other approach (at least one other approach) is to try and minimize needed power to keep the overall weight down. I think the older (50-60's) planes with their thinner airfoils may fit the smaller power needs better. As I point out, I try to keep the cruise power down, but hope to keep the vertical power available when needed. That's why I have been playing around with this prop stuff, so at least I have a better idea of what to try out. Now I need to apparently talk to Glenn Lee (a fellow club member) to get some titanium bending ideas! He is also a retired engineer from Fermilab, so we are (or were co-workers).
I am guessing the Cubic inch approach is likely to be a sure thing, whereas the latter (the way I am going) may or may not be the way to go. Anyway my two cents.

Dean Pappas · « **Reply #15 on:** December 15, 2007, 09:12:34 AM »

Hey Gang,
Yes, I like the idea of designing for better horsepower use. Think of just about anything Werwage wrote about the USA 1 or the like. My whole "thing" is to maximize the peak power that gets used as compared to the level lap power. Diameter is king, and multiple blades are a detour that will work just well enough to distract you for a long time. I know, I'm being a PITA

later Compadres!
Dean

Mark Scarborough · « **Reply #16 on:** December 15, 2007, 11:57:25 AM »

Hey Rudy, solution for the long gear legs, , build yoru planes with low wings like the P-40, then the gear legs can be about an inch shorter

walterbro · « **Reply #17 on:** December 28, 2007, 09:21:45 AM »

Hi GUYS - This is my first POST. I agree that dia is" king" except if you don't consider
disc area. I have been using 2 motors and props from 9.2" to 9.6" dia with aTP 4000
4s2p battery in my GEMINI . I have used MOTOCALC for the last 4 years to help me compare disc area and flight power. my conclusion is 2- 9.6'dia props are botter than
a 12" dia in several ways. 9.6 squared x 2 = 184-- 12 sq. = 144-- 13sq = 169.
This is only theory but my flight power has proved to me that it is real . It also avoids long gear and multi blades. NO I did not use HI GOV mode yet! My new twin will give it a try , with counter rotating props.
Walt Brownell

Igor Burger · « **Reply #18 on:** December 28, 2007, 09:31:26 AM »

No it is not area, area will give you only power of 2, reality is it needs power of 5. Means 10% bigger prop will give you 1.1^5 = 1.6 x more thrust and will need 1.6 x more power.

Rudy Taube · « **Reply #19 on:** December 28, 2007, 10:40:15 PM »

Hi Walter,

I'm glad you made your post. You were the inspiration for Will Moore's Big ECL twin bomber You proved to all of us how well a twin ECL plane could fly. :-)

It is interesting that you are getting such good performance with the smaller 9.6" dia. props. What RPM are you turning them? We are turning our 12 x 6 APC-E props around 9,100 RPM with good results. Will's counter rotating props have worked very well, and we think they help. They sure don't hurt anything. ;-)

I think one of the big benefits to the twin ECL setup is the same benefit we get on FS twins; the accelerated air over the wings increases lift. This means a lower AOA, which should mean less power req. for the same AS. This is just what we need in CL. Our speeds are somewhat fixed around 55 MPH, so any increase in lift is a solid PLUS. This benefit may be why you can use such small dia. props and still get good performance? Will's two 12x6 props pull his big 74 oz plane with authority, even using the smaller Hacker motors.

Igor's point is a very good one. We get a very BIG (exponential?) increase in power by increasing prop dia. An increase in Prop Disc area is important, and will help, but the real benefit comes from increasing the dia. of the disc. .... For example:

A 9" dia prop has an area of 64 sq. in. Two have 128 sq. in. of area. BUT, at 9K RPM, the tip speed for each prop is only 240 MPH. ..... A 13" prop has a similar disc area, 132 sq. in. BUT the tip speed is 348 MPH. That is more than 100 MPH faster. Since most of our thrust comes form the outside 20% of the prop, our increase in area occurs where it is really producing a lot more thrust because of the rapid increase in prop speed out there. ..... I think this is what Igor was referring to in his post?

You are right about the LG issue. The smaller props do make it easier. I'm willing to have "Stork" gear for all the extra power the larger props give us, but it does look a little dorky compared to the sleek, low slung look of the wing mounted lower gear models. There are other issues that may limit how big a dia we can go to. The torque and P-factor problems are magnified as we grow our props. I use 28" props on some of my planes and often have to use R RD, so maybe we will have to start using Rabe RDs on our BIG prop ECL planes in the future?

Again, thank you for being an ECL pioneer and showing us the way! :-)

Regards,

walterbro · « **Reply #20 on:** December 29, 2007, 08:21:36 AM »

HI Rudy

Thanks for your input. I have used a MOTOCALC approach to sizing and comparing
motor and battery choices for the last 3 years. I have found that outrunner motors are not handled perfectly by MOTOCALC, however it is a very useful tool for comparing
design tradeoffs. I just compared my axi 2826/10 with a 12"x 6" APCe and my standard
battery TP 4000 4s2p to my GEMINI 2818-900 Torque(2) motors with a 9.6" x 5.3" prop at the same power input 41.1 amps. These are only theoretical calcs but they show some interesting comparisons.
AXI 2826/10 12x6 prop, 41.1a ,12.9v, 529watts input. 8962RPM, 68.6 oz Static thrust, 50.9 pitch speed, 5:51 Time, 1912 FPM rate of climb.
2818-900 Torque , 9.6x5.3 prop, 41.1amps.12.9v,530watts in 10593 RPM, 84.4 static thrust, 53.2 pitch speed ,5:50 time, 2198 fpm rate of climb.
This shows me that with the battery I am using that it is more efficient to use 2 smaller props, for my pattern. This has proven out to be true at the 2007 Nats where
I flew my twin in 10 - 14 MPH winds and could not handle those winds in my Classic
NAKKE with my 2826/10.

Regards
Walt Brownell

Igor Burger · « **Reply #21 on:** December 29, 2007, 07:45:18 PM »

Yes Rudy ... what I mean is that the 11" prop instead of 10" prop will be (11/10)^5 more powerfull (will need 1.6 x more power and will make 1.6 x more thrust)

here you can find more:
http://www.mh-aerotools.de/airfoils/propuls3.htm
http://www.mh-aerotools.de/airfoils/propuls2.htm

tom hampshire · « **Reply #22 on:** January 03, 2008, 07:48:56 PM »

Rudy - May I suggest the use of 1/8 wire with a spreader? Run a piece of 1/16 or so between the gear legs, say an inch or two below the fuse bottom. Wrap and solder the ends to the main legs. Stiffens it just enough to let you get by with the thinner wire. Tom H.

Rudy Taube · « **Reply #23 on:** January 03, 2008, 08:58:01 PM »

Thanks Tom, that's an excellent idea for our planes with the LG in the fuselage.

Dean Pappas · « **Reply #24 on:** January 04, 2008, 09:00:29 AM »

Quote from: Igor Burger on December 29, 2007, 07:45:18 PM

Yes Rudy ... what I mean is that the 11" prop instead of 10" prop will be (11/10)^5 more powerfull (will need 1.6 x more power and will make 1.6 x more thrust)

here you can find more:
http://www.mh-aerotools.de/airfoils/propuls3.htm
http://www.mh-aerotools.de/airfoils/propuls2.htm

Happy New Year Gang,
I just dug up the Prof. Larrabee reference and the exponents of both the RPM and diameter terms are different for power and thrust.

P = Cp * n^3 * D^5

and

T = Ct * n^2 * D^4

later,
Dean

Igor Burger · « **Reply #25 on:** January 04, 2008, 09:07:01 AM »

It is hard to believe, it is at THE SAME condition? It looks like thet second formula miss pitch or not static or what. ... if not, all 3D r/c flyers with large models will cry after reading

Alan Hahn · « **Reply #26 on:** January 04, 2008, 09:11:57 AM »

I think what one should also remember is that a larger diameter prop need not (necessarily) require more power to provide the SAME thrust as a smaller diameter prop. This mainly pertains to the "non-exerting" part of the flight. What the larger diameter prop apparently provides (if I believe the graphs above) is more thrust when the airspeed begins to drop (when the rpm is kept constant). During this phase the power input of course goes up, but I think it is worth it there!

Igor Burger · « **Reply #27 on:** January 04, 2008, 09:15:55 AM »

Alan, those equations from Martin Hepperle page are for STATIC thrust and power.

Alan Hahn · « **Reply #28 on:** January 04, 2008, 09:51:58 AM »

Igor,

I may have jumped into the your and Dean's conversation without fully realizing what was being discussed, but just to be clear about my concerns, I am not worrying about static thrust (airspeed=0).

Alan

Dean Pappas · « **Reply #29 on:** January 04, 2008, 08:52:31 PM »

Hi Gang,
Yeah, you have to love it: all these manufacturers (both wet and electric) quote static thrust like it matters. I don't fly my stooge! Okay, for 3-D stuff it does, but for virtually everything else, it's just marketting-ese.

The two equations given are for "everything else being equal" in flight conditions. That is, at a given airspeed and pitch far from stalled on the ground, a change of diameter produces 5th power changes in power input and 4th power changes in thrust. Dig the RPM exponents too!

So when at the third time twisting Bob's arm (the first two yielded small increases in diameter for the 3-blader) the change to a 12-6 resulted in a very small reduction in the total power consumed for a full schedule. That was just because the efficiency was improved. Similarly, a reduction in RPM and almost identical increase in pitch when combined with the right diameter increase will yield another small efficiency gain. No, I don't remember the exponents for pitch in both equations. I gotta go look at the textbook. My guess is that it's 1 for the thrust equation and 2 for the power equation.

Definitely a pointy-head alert!

Dean

Author Topic: Propeller Calculations---more than you want to know!-Pointy head alert! (Read 4471 times)

Alan Hahn

Alan Hahn

Alan Hahn

Alan Hahn

Alan Hahn

Alan Hahn

Alan Hahn