RoofTop Scale testing

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

berridos

Well-Known Member
Joined
Oct 10, 2009
Messages
1,146
Location
madrid
While this thread could be understood as an RC model thread it still is the cornerstone of proper research before starting a real sized prototype and testing on the rooftop at the end is massivly cheaper and faster than producing a ton of rc models with tiny modifications.
As i am planning to start the construction of a delta i need to research the effect of several devices regarding lift (specially vortex lift), drag and stall speed as well as precise identification of the neutral point (not an easy task in tailless planes). Only stability is still a matter that can only be tested with rc-models.
I had a sleepless night designing the rooftop structure and can conclude that it is really a one weekend task, maybe two)). This morning i checked the bottom of my car and there is plenty of attachment points to the chassis.
I consider it vital to fly in clean air and therefor the structure would reach from the roof to the front wheels of the car (the front structure would be attached to the bottom chassis of the car).
Ingredients are linear guideways, levered arm-structure to change the aoa, pitot tube the works precisly at lower speeds, gopros to record the measurements all at the same time and 3 scales with remote reading (preferbly cilinders that measure pressure in both directions-never saw that type of scales, but i am sure the chinese have them)
The absolute measurements are not vital as i will compare to a benchmark design (so only relative improvements are critical).
In my case i want to build a 2m span prototype as a reflection of the 4,5m original design. The protoype will have three atttachment points, One at the center of lift and to others to check lift (in front and behind at an equal distance).

The point of this thread would be the discussion of scale effects that have to be taken into consideration.
In case of rcmodels, wingloading could be one of the parameters to match in order to replicate the behaviour of the 1:1 plane. Wing Loading, WCL & Conversion Calculator
But what is relevant on the rooftop?
-If i set the prototype at a certain AOA and drag it horizontally on the rooftop, can i paint a realistic liftcurve by iterating this exercise?
-Do you think that the stall could be identified on the rooftop by a sudden loss of lift at the lever arm on the scales? If for example on the rooftop that sudden loss of lift happens at 50kmh , do i need to scale this speed up to match the stall of the realsized plane?
-Another goal of the test would be to test the location of a belly flap that matches exactly the center of lift and therfor doesnt change the pitch attitude when deployed. That test would happen in free air. However could this location be wrong as in the real world the flap will be blown by the propwash of the tractor engine?
I have a dozen more test in mind.

I wonder if you could contribute other effects that could distort the direct reading of such tests and maybe shed some light on how to handle those effects.
 

berridos

Well-Known Member
Joined
Oct 10, 2009
Messages
1,146
Location
madrid
Ok, talking about viscosity is pretty theoretical. How would you translate speed readings under consideration of the reynolds number
Personally in this case i dont want to achieve laminar flow. but reynolds also affetcs other measurements. Would be a linear factor enough to compensate for reynolds numbers?
 

Vigilant1

Well-Known Member
Joined
Jan 24, 2011
Messages
7,162
Location
US
Ok, talking about viscosity is pretty theoretical. How would you translate speed readings under consideration of the reynolds number
Personally in this case i dont want to achieve laminar flow. but reynolds also affetcs other measurements. Would be a linear factor enough to compensate for reynolds numbers?
Sorry, I'd be way out of my league if I claimed to have anything authoritative to offer. My >guess< is that viscosity might be pretty important when considering vortex lift, and I'm under the impression you want to do that. You might also do a search here using appropriate terms ("scale model" "Reynolds" etc) as I recall we've touched on this before.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
16,184
Location
Memphis, TN
You just asked a text book of questions.

I assume you have a standard flying RC of your model before you test. You can ballast and power load it to scale. Once it flys, you can stick it on your roof top and measure. a 2m development plane will be much heavier than a standard RC plane that size. It will be very challenging to fly.

Gathering information is only relative to the information you try to find. What information is throwaway and good? Flying model on a roof only gives numbers of the model. You are inferring that you can transfer the information. That is where the text book comes in.
 

Victor Bravo

Well-Known Member
Joined
Jul 30, 2014
Messages
10,066
Location
KWHP, Los Angeles CA, USA
Make a pylon, pole, or pyramid instead of a rooftop "rack" or "bed". That way you can use a spherical bearing to mount the model, and you will be able to test it at all angles of pitch and yaw. It will also allow you to more easily raise the model away form the vehicle to be in "clean" air.

If you are clever, you can design the rooftop mount so that it clamps or attaches to a roof rack or rails on top of an SUV type car.
 

berridos

Well-Known Member
Joined
Oct 10, 2009
Messages
1,146
Location
madrid
No SUV at hand but as you state, the structure i have in mind allows the plane to fly in clean air on a pylon mounted in front of the car.
Sperical bearing is a good idea. The measuring devices arent perfectioned yet.
I thought about first building the test aerodyanmic model (heavy roof top model with lots of rails and possible adjustments) to see wich modifications are trash and wich could add value. At a later stage, build an rc prototype with the succesfull devices to test if the thing has a gentle behaviour compared to the original.
If i have time i will sketch the structure.
With leverarms i can augment the sensibility of the readings the model produces.
I really dont want that model to fly. Just stop the car and rearranging the configuration and angles a hundred times and recording every pass.
For example testing several samples of sharp leading edges at high aoa for lift and stall. Specially testing if the whole leading edge needs to be sparp or only the inboard sections
Also doing some minor test on planforms.
Wish there was a way to quantifiy the spanwise lift distribution.
 
Last edited:

Hephaestus

Well-Known Member
Joined
Jun 25, 2014
Messages
2,364
Location
YMM
You might want to check the RC DTF/FFF/bluecore threads over on rcgroups.

No offense but you can typically bang out a scale-ish model in a few hours with a little practice.

Compared to the rooftop mount - you'll be days of work creating that alone to get half as much information at the end of the day.
 

berridos

Well-Known Member
Joined
Oct 10, 2009
Messages
1,146
Location
madrid
I dont share that view. Every RC model looks cool and flies awesome, but its a totallty subjective exercise.
comparing the original design and tweaking the modified one step by step on a rooftop is way more scientific and objective. I have so much trash in my garden that i will build that structure in less time and money than any rc model. I used to build rc models when i was a kid (about a dozen and never learned to fly them) there was a guy at the airport that always crashed them for me)))
Anyways once sinthesized a good prototype i well let it test as an rcmodel.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,593
Location
Saline Michigan
Open any basic textbook on fluids, and you will find a section on similitude that goes on to tell us about several terms using a variety of dimensionless numbers. Reynolds, Froude, Weber, Prantl, and Mach come to mind. For airplanes Reynolds and Mach are significant. Reynolds Number is the ratio of dynamic to viscous effects. Mach Number largely shows up with compressiblity effects, shock waves, etc, but wing lift curves and wave drag are also affected by them.

Much of airfoil behaviour requires the Reynolds number to be similar between model and the real thing, and Re is linear with both speed and chord of the wing, so if your model is half scale you must run it at twice the velocity to hold Re, but now your Mach number twice as high as it should be. In a 50 mph plane, that means testing at 100 mph and the Mach number still pretty small, where sonic effects is not too bad. In a 200 mph airplane, 400 mph is about impossible, and the Mach number means a bunch of other issues.

So, you can run a test out front of your car, put on load sensors for lift and drag, and then go fly the model at car speeds. If your speeds are low and your scale is low, Re and Ma will be low. Include your Re and Ma numbers in your data and go look for how your data would be effected by changing to the real thing's Re and Ma numbers.

Have fun, but know that details of stall behaviour and lift/drag relationships will have some level of uncertainty to them.

Billski
 

berridos

Well-Known Member
Joined
Oct 10, 2009
Messages
1,146
Location
madrid
Thanks. i need to dig deeper into theory and do some number crunching in excel, but those 100mph shocked me.
My goal is to lower the approach and stall speed behaviour of the original design (stall speed 60kts) and therefor i will test several potentially vortex lift increasing devices on only one prototype and keep the other one unmodified.
If the original plane had gross weight of 300kg and the scale plane had one third the wingsurface, once the model produces 100kg of lift minus the model weight (whichever speed it takes, hopefullt not 100mph) wouldnt that be a suitable regime to compare apples to apples in a dimensionless exercise? The target reference would be wingloading.
I am wondering if i really need 100mph or some similar excessive speed to match wingloading and be able to test anything in a useful way.
I would need to go to the salt lakes in nevada.
 

Vigilant1

Well-Known Member
Joined
Jan 24, 2011
Messages
7,162
Location
US
Just an idea: the full scale plane you are planning to build has a span of just 4.5 meters (about 14.5 ft). Would it be practical to make a big full-scale aerodynamically accurate model out of solid XPS foam (2 lb cu ft)? Cover it with a thin sheet of fiberglass and epoxy. It would probably be too cumbersome to put it on a cartop, but it could easily go on a tripod on a flatbed trailer a few feet above the car. For the drive to test site, tip it with one wingtip down.
Okay, maybe crazy. But there would be no scale effects, there would be plenty of room to mount pressure taps/manometer tubes, and I personally would have a lot more confidence in the resultant info on stall behaviour, drag, and stability. It takes advantage of the uniquely short wingspan of this type of plane.
 

berridos

Well-Known Member
Joined
Oct 10, 2009
Messages
1,146
Location
madrid
Definitly an option to think about. My CNC Eps cutter is designed for that. However that beast would fly in ground effect. But at least the outcome would be trustworthy.
Maybe there is some vehicle i could rent for that ocasion.
 

jedi

Well-Known Member
Joined
Aug 8, 2009
Messages
3,049
Location
Sahuarita Arizona, Renton Washington, USA
FYI Northwing mounts the wing inverted so as to not lift the truck. Speed is a prooblem for structural load tests.

The Wright brothers took a hint from George Spratt and measured L/D rather than lift and drag separately. They found it much easier to get accurate and useful data. Model needs to be well ahead of the vehicle to avoid the bow wave.
 

Vigilant1

Well-Known Member
Joined
Jan 24, 2011
Messages
7,162
Location
US
However that beast would fly in ground effect.
True, at least to some degree. You could just make 1/2 of an airplane (cut vertically along the longitudinal axis) and point the wingtip toward the sky. Awkward, but put a rail down the middle and move the axis of rotation to wherever it should be. No more ground effect.

ETA: I like jedi's ”mount the plane upside down" idea better.

Maybe there is some vehicle i could rent for that ocasion.
I suppose the mount could go in the bed of a large pickup truck. A flatbed trailer allows you to keep everything set up for days/weeks and you can still use the vehicle for daily driving. A light duty trailer is inexpensive and can be resold easily. Or keep it and you'll always have friends...who need a sofa moved somewhere. :)
 
Last edited:
Top