Wind Tunnel

[fasterthanspeed]

Ive been thinking of building a wind tunnel quite some time now. Im planning on building one this weekend. I found a smoke machine Im planning on using and it will be made out of cardboard to keep down on the cost as cardboard is free to my disposal. I will just be using a simple cooling fan that I use during the summer to keep cool in the garage. Any thoughts about this?
Also, an off topic subject. I was wondering if any of you knew where i could get some non pourus foam blocks as the florists foam i use to create a mold for the body is very pourus, and itd be easier to use a less pourus foam.

[wqqdster]

I have often though about building one myself and probably will before testing my car at serious speed. I remember a website on the subject and one of the points was that the fans had to pull the air through rather than push it. The fan blades twist the air not allowing straight air flow over the body.

Besides smoke, another way to ‘see’ the air flow is to tape very fine threads on the body so the loose end of the thread is to the rear of the body. When the air flow is laminar (not lifting) the threads will be flat on the body and will point in the direction of the flow. If the threads are flopping about and lifting off the body you have a turbulence and a low pressure area there....usually lift!

Is one room fan going to move enough air? I was going to use three of them connected by cardboard ducting to the end of the tunnel. Air resistance and other forces are not linear but exponential so a pocket of seemingly insignificant lift becomes a blow over at 100 per!
It takes like four times the power to move something at 160 as it does at 80 so if you don’t have enough air flow you may not get reliable data. I’m sure someone here with more recent college engineering work can give you the particulars on the formulas.

I was also going to attach a small spring scale to the front of the car so I could see what effect spoilers and fairings had on drag. You could also use a set of sedan scales to see how much down force you’re actually generating and if its up front or out back. There is a ton of info you can glean from a well set up wind tunnel.

...and isn't it the porosity that makes foam.......um.....foam?

Good luck! Post some pics!

Gary O
www.shoeboxtops.com

[NotWalkinBlind]

I would recommend putting two scales under the car... one under the front axle and one under the rear. That way, you can measure downforce/lift separately.

You need to build the top and sides of the tunnel... the part where your car sits... with glass or plexi so you can see it... look in the yellow pages for places that sell plastic for signs... they usually have a scrap bin. I have bought clear 1/8" and 1/4" Lexan scrap from these places before.

If the tunnel area was, say 12" square in cross section, with cardboard, you could flare it out and attach it to the outer edges of a simple 20-24" box fan like you see at Lowe's, Home Depot, etc. for $15-20... I would think that on high, it would suck air through that 12" space at a fairly high rate.

[NotWalkinBlind]

I decided to unleash my mad Photoshopping skills to show y'all the box fan cardboard & plexiglass windtunnel I described... enjoy... :D

http://img.photobucket.com/albums/v134/NotWalkinBlind/windtunnel.jpg

[wqqdster]

Better'n I could do! That looks excellent. Personally, I'm going to move the fan further away from the tunnel and put one or two more along side that one, widening the duct to accomodate them. I think that would increase the wind speed through the tunnel without changing the speed of the fans by increasing the vaccuum between the duct and the tunnel. That may also collapse the duct!

I believe yours will work just fine.

What sort of 'smoke machine' are you using?

[ajlovering12345]

Foam is porus and permiable. I think your looking for non permiable foam

[fasterthanspeed]

http://www.musiciansfriend.com/product/Lighting-400W-Fog-Machine?sku=801401&src=3WFRWXX

That one. Im going to moify it so itll have a hose so i can have a fine line of fog and itll be more contralable.
That design is about what i designed.
As for the plexi, home depot has some that i was thinking before that id use for the windows.

[B4Maz]

I did a research paper on exactly this. Here is the link to a few pictures and some of the result:

http://fastrc.blogspot.com/2007/02/windtunnel-testing.html

I have a link to my full paper in PDF format on that page. Here are a few pics from my testing:

http://bp0.blogger.com/_3N0hwXEVK9I/RcloQfzAsCI/AAAAAAAAANg/lUTpK9DVy-8/s400/DSC00490.JPG

http://bp3.blogger.com/_3N0hwXEVK9I/Rco-V_zAsRI/AAAAAAAAAQQ/pxVBAHgWBgA/s400/DSC00508.JPG

http://bp3.blogger.com/_3N0hwXEVK9I/RcloQPzAsBI/AAAAAAAAANY/HH2R6EL97oY/s400/DSC00483.JPG

http://www3.aere.iastate.edu/wind/images/BillJamesWindTunnel.JPG

http://bp3.blogger.com/_3N0hwXEVK9I/Rclo5PzAsHI/AAAAAAAAAOI/M-OyJUEa3vU/s400/DSC005191.JPG
Wing at 80mph

[Cossack71]

You might want to glance over this article then

http://www.allcarmodels.com/acm/news?mv_arg=20070405_windtunnel

[NotWalkinBlind]

Well, as we can all plainly see from those photos B4Maz posted... especially the next-to-last one where you can see the way the small part of the tunnel flares out to the fan housing...

...that he copied my design.

Bummer. LOL



fasterthan... all your work's been done for you.

Just use a Monte Carlo body.

Without that wing. :D

[B4Maz]

Another pic of the windtunnel I used for testing:

http://www3.aere.iastate.edu/wind/facilities_pics/billjamescad.jpg

More specs about the windtunnel here:

http://www3.aere.iastate.edu/wind/smallscalefacilities.htm

[wqqdster]

Nick,

What an awesome experiment and paper. I’m afraid I wouldn’t have been able to finish the paper because I would have had too many more experiments to try!!! Your photos and paper bore out what I was saying that what happens at 20 mph isn’t 1/4 of what happens at 80.

A couple questions....if the bolts supporting the air foils would have been in airfoils themselves would that have made a significant difference in the drag?

As the speed was increased, the L/D ratio seemed to drop off. Do you think this was due to the close proximity of the roof structure of the car disrupting the air flow? If there were a cleaner air flow would the L/D ratio have stayed closer to the lower speeds?

The photo of the lexan wing all buckled up is amazing but not surprising. Did you consider installing a more substantial support frame work under the standard wing or was it so inefficient that it wasn’t worth the work?

Where can I get the profile of the NACA foil you used?

Lastly, you have a drawing of your concept car on your site. Wouldn't the tail dropping down actually give the car a wing profile and cause it to fly?

Thanks,
Gary O

[NotWalkinBlind]

You know... it might be smarter to just use a body design that somebody else has already spent a gazillion dollars perfecting in a wind tunnel.

http://img.photobucket.com/albums/v134/NotWalkinBlind/porscheGT1.jpg

Of course, you could make it even better by leaving the mirrors off and enclosing the rear wheels. :D

The Porsche GT1 and Mercedes CLK are available in 1/5th scale here... click on "Bodies" to the left, then "1/5":
http://www.netshop.nl/shop/rc/enter.html?lmd=39386.550394

[ajlovering12345]

They are also availalbe in 1/8th I have 2 waiting to go on my car!

[NotWalkinBlind]

I would recommend just using one... to save weight. :D

[ajlovering12345]

Its too long so I have to cut them and use 2 joined together. lol

[NotWalkinBlind]

:eek: :eek: :eek:

Can you draw us a pik-chah?

[fasterthanspeed]

No body on the market will fit my car, its as wide as a 1:10 200mm body and longer than a 1/8 onroad. It has a wb of 14in.

[NotWalkinBlind]

You're gonna need 4 of those GT1 bodies.

And a roll of duct tape. http://img.photobucket.com/albums/v134/NotWalkinBlind/LOLdude.gif

[B4Maz]

Nick,

What an awesome experiment and paper. I’m afraid I wouldn’t have been able to finish the paper because I would have had too many more experiments to try!!! Your photos and paper bore out what I was saying that what happens at 20 mph isn’t 1/4 of what happens at 80.

A couple questions....if the bolts supporting the air foils would have been in airfoils themselves would that have made a significant difference in the

As the speed was increased, the L/D ratio seemed to drop off. Do you think this was due to the close proximity of the roof structure of the car disrupting the air flow? If there were a cleaner air flow would the L/D ratio have stayed closer to the lower speeds?

The photo of the lexan wing all buckled up is amazing but not surprising. Did you consider installing a more substantial support frame work under the standard wing or was it so inefficient that it wasn’t worth the work?

Where can I get the profile of the NACA foil you used?

Lastly, you have a drawing of your concept car on your site. Wouldn't the tail dropping down actually give the car a wing profile and cause it to fly?

Thanks,
Gary O

Yes, the bolts add a significant amount of drag. I did account for that much drag in my calculations. I subtracted the difference in drag of the wires holding the standard wing and the bolts holding my experimental wings.

It is possible that since the airflow was so turbulent in the back of the car at speed that the L/D ratio could drop off significantly. One of the other experiments I didn't get to do was adjust the height of the wing relative to the rear deck of the car to see if it would make any significant changes in downforce.

Adding more support to the standard wing was another experiment I wanted to do, but didn't have time.

You can get generate profiles at a couple places:

NACA 4 digit airfoils:

http://www.pagendarm.de/trapp/programming/java/profiles/NACA4.html

NACA 5 digit airfoils:

http://www.pagendarm.de/trapp/programming/java/profiles/NACA5.html

The concept model I have on my site is designed to use ground effect (http://www.nas.nasa.gov/About/Education/Racecar/physics_groundeffect.html) to create all of the cars downforce. I want to minimize how many wings I will need on the car to create downforce. The most efficient way to create downforce on a car is to use ground effect. I was hoping through windtunnel testing to be able to have a low tail to keep my pressure drag low while still creating large amounts of downforce from ground effect (http://www.nas.nasa.gov/About/Education/Racecar/physics_groundeffect.html)

[Mod Man]

Good stuff!

I have an anemometer (wind speed meter) and a smoke pen already for this type of testing and haven't gotten around to building the wind tunnel yet. Glad to see someone else pushing to do it.

Wind, what a drag! ;) :D

Matt

[Cossack71]

I thought about building a complete wind tunnel / dyno combo for RC.
Your average owner probably would never buy one, but you could sell it to shops and race teams. I'm sure it wouldn't be cheap.
If it was at a LHS, they could sell test time to people that wanted to do fine tunes.

[B4Maz]

I found this site where someone did flow experiments with cars in a water tunnel.

http://www.instructables.com/forum/EP10RQOAA7ETVPKQ6T/

http://www.instructables.com/files/deriv/FJ6/RDNF/RRUETVPKPOK/FJ6RDNFRRUETVPKPOK.MEDIUM.jpg

http://www.instructables.com/files/deriv/FBC/MCOG/X0QETVPKPO0/FBCMCOGX0QETVPKPO0.MEDIUM.jpg

http://www.instructables.com/files/deriv/F3R/OKCZ/8BVETVPKPNO/F3ROKCZ8BVETVPKPNO.MEDIUM.jpg

[trailranger]

I thought about making a water tunnel. Since low budget is key, I think I would have to make the tunnel for 1/24 scale cars. I don't remember the specs but I think water flow of about 20mph is about the same as air speeds of 250mph?

Supplies needed:
TWO 55 gallon Water barrels (BLUE PLASTIC, surplus for $35 each.
ONE sump pump, NEW $50
Garden Hose, $10
2" and 3/4" PVC PIPE and Fittings $45
Plexi-Glass, 12 X 24' sheet $15
Scap PLywood and Fasteners $10
Bulk Straws $10
WaterTight Paint $15

The idea is that since low power is cheap, we will have to compensate for that by using gravity. By having one barrel higher than the other you can create a gavtiy flow of water. In order to see how the fluid flow interacts with your model, you need a viewing box and I would use Plexi-Glass for that.
Since the barrels are already threaded for 2" Pipe, we will use that size pipe and port that into the viewing box. At the front of the box will need to be a diffuser system to get the water flow to spread. After the diffuser section is the straw box. The straw box gets the water to flow laminar before hitting the test object. I Planned on having a box about the 8" square by 48" long. The first 24" would be for the diffuser and staw box, the next 12" viewing windo, and final 6" just the end cap with exit pipe. On the exit pipe mount a valve and then run the pipe into the catch barrel. There is a need for air vents on both barrels, I was thinking making the system closed loop with a small 3/4" pipe connecting the two barrels. As the water rushes out of the top barrel it creates suction on the air vent and that pulls water into the catch barrel, CLOSED LOOP>> LESS LEAKS. Fill the system to the point where the water tunnel is fully filled and so is the catch barrel. Now using the sump pump and the system valve closed, fill the top barrel. Have FUN

[glypo]

Sorry to post late in the topic.

What kind of useful information are people expecting to find in their own wind tunnels, or water tunnel as the case may be? You can't expect to measure anything accurately, all you can hope for is perhaps a rough idea of stream lines. This won't help in anyway in shell design. So before anyone build a tunnel specifically for this speed challenge I would urge that you don't, money could be better spent elsewhere on the project.

I agree with NotWalkinBlind in that it's certainly worth looking at what has gone before, where testing has been done. By that I don't agree with the Porsche though, they are designed for high cornering speed and downforce, not needed for speed run car. Best looking at what full size land speed record holders have done in the past, particularly with wheel driven vehicles.

[ajlovering12345]

For me there is not an issue with not enough HP for an electric car and even an IC car has all the power at the top end so torque is the main issue. If your spending money on designing a car use it to get the most HP and the most effective drive train.

[wqqdster]

A couple years ago a Mercedes GTP car, with aero designed by real live engineers, caught a minor bit of air down the Mulsannes Straight (sp?) and flipped over several times. A change in the attitude of the car of only a degree or so contributed to this accident.

I have a lot of intuitive reasoning but not enough to send a couple grand worth of engines and electronics up to 100mph+ without some sort of test. At that speed the bodies can deform and that nice down force you had a split second ago is now picking up the front of the car and tumbling it down the strip.

I have an idear for a method to test it but I’ll wait to post it until I’ve tried it.

[glypo]

I couldn't agree more. I'm an aerodynamicist, and as such my speed car is going to be very heavily aerodynamics based. But what I’m saying is that simply a home built wind tunnel really won’t help at all. Better off putting the money towards anything else, better cells, motor.... heck even a book on aerodynamics.

I’m afraid sending your couple of grand down the strip at 100mph+ without prior testing is the only real option. There are two options to test the aero of the car, that’s wind tunnel or CFD. Wind tunnel needs to be at least a 1.5m working section so ensure there would be no wall interference. These cost thousands to hire a day, so not a real option. Plus you need to skill to know where to measure pressure, and how to evaluate it. And CFD, a good package costs truly thousands, and to get any kind of accurate results you need to be trained to use it. And just like the wind tunnel it will give you raw data that needs analysis.

I would also imagine with CFD the results might be varied, being a small model at sea level the Reynolds number will not be conventional for the code. Usually they are suited to analysing bigger objects. Not sure though, I haven’t tried plotting anything small in the panel method code we run. So what my long, and not to the point, post is trying to get at is: aerodynamics for a speed cars has to be educated guess work.

People seem to think you can analyse things by looking at things and going “hmm... looks slippery must be good”. Sadly not so easy, slippery means nothing lol. If you are truly worried about aero design, your best bet I say is to use a book to get an understanding of pressure distribution around the car. I know at university writing my final year project (dissertation if you prefer) my project supervisor was an aerodynamicist. (make sense with me writing an aerodynamics dissertation) Although he spent most his career before on aircraft design at Airbus etc, he did work for McLaren F1 team for a couple of years. And his recommendation for aerodynamic car reading was a book written by Simon McBeath called Competition Car Aerodynamics.

I took the book out on loan from the library and it certainly seemed good enough. Being car aerodynamics it was of little interest to me at the time, and there were some basic errors like it used the equal transit theory to explain lift generation. Which in a sense I suppose is good for the level the book is aimed at, which isn’t an aerodynamics expert, closer aimed towards someone who is technically minded but with little aero knowledge. Therefore if you haven’t already read it I suggest this is worth a read from your local library, and if you’re impressed and it addressed areas you need to look at on your speed car, it’s cheap to buy (here in the UK at least).

Anyway sorry for my long rambling post, I seem very ‘typeative’ tonight :)

[ajlovering12345]

Great post! Aerodynamic does play its part but you need downforce more than making the car slippery

Look at it this way

At the moment you only have 1/4 mile to run up and for a 5lb car to get to 150mph you need to accelerate at 0.6 G for the full 1/4 mile. If you accelerate more than .6g the wheels will spin as you dont have enough traction.

For this you only need 2.5hp

Now if you want to get past 150 you either need more room but then you need to be 100ft in the air to see the car or more weight or downforce.

Power isnt a problem because you can get 5Hp electric motors and 3 IC engine will give you 5-7hp

[NotWalkinBlind]

I don't remember hearing about a Mercedes CLK flipping at Le Mans, but a Porsche GT1 back-flipped at the rise about a third of the way down the back straight during a Petit Le Mans race at Road Atlanta a few years ago.

The reasons it flipped was because of a combination of factors...
1) it was so close behind another car that its downforce was "interrupted"
2) while it was under acceleration
3) while it was cresting a small rise in the track

Here is video of it:

http://www.youtube.com/watch?v=wM4guvo6Ifo

With top speed challenge R/C cars being alone on a flat track, I would think a GT1 or CLK body would work fine... the real cars are pretty slippery because their downforce is mostly underneath for the cornering, and the body itself has to be slippery enough for straightaway speed.

And I think they each have enough downforce to keep the nose and the rest of the car down at high speed, but whether the nose would stay down during acceleration is another matter.

Also, body deformation could be a factor at 80+ MPH if the tip of the nose were not supported.

It would also be smart to cut off that roof scoop... it's not needed and would only add drag... it would be a simple matter to glue a piece of sheet Lexan underneath to fill the hole.

Eddie Weeks was using a GT1 body on his turbine car's initial tests...
http://www.corpcomp.com/weeks1/Car/Car.html
http://forums.radiocontrolzone.com/showthread.php?t=176409

If ya wanna go all Bonneville with it, Sigurd's body design was doing pretty well the last we heard...
http://forums.radiocontrolzone.com/showthread.php?t=180309
http://www.ruschkowski.se/
http://forums.radiocontrolzone.com/showthread.php?t=175811

[glypo]

Really people have to get past the whole "slippery" idea. If you calculate the amount of drag that is caused by screen friction, then compare it to that caused by lift induced drag over wings, and massive pressure voids on the cars you will notice that on certain areas not a lot if important.

I maintain though that those bodies won't give you amazing aero characteristics, but then again if you don't have the ability to produce your own body choices will be limited. You just have to worry with the Le Mans style body about the pressure void behind the car, the wedge shape leaves a steep back edge. You would need very effcient diffuser so help with that.

Wow that second link, now you're talking! Proper aero data. Not to dissimilar from the shell I am making for my car. My underbody is different though.

[wqqdster]

I'm curious about that number.... isn't the tire compound and surface relevant in determining that number? It seems that a sticky tire on a sticky surface would allow more acceleration than that. Top fuel dragsters pull a couple g's at the start...why wouldn't that be possible with a car with less mass to move?

[ricebubbles]

hey guys i think you have look at this before you reinvent accidents please.

www.mulsannescorner.com/techarticle1.htm


www.mulsannescorner.com/data.html


its worth a read boys

[ajlovering12345]

I'm curious about that number.... isn't the tire compound and surface relevant in determining that number? It seems that a sticky tire on a sticky surface would allow more acceleration than that. Top fuel dragsters pull a couple g's at the start...why wouldn't that be possible with a car with less mass to move?

Yeah I looked at the spreadsheet calcs I have and If I put in the numbers for a 1/8th circuit car I know goes from 0-65 in 2.5 seconds it says it cant so I'm sure there is a flaw somewhere. I would have thought you could get 1.2G all the way along the straight in which case you would get 200mph before 1/4 mile is up!!

[fasterthanspeed]

Well since this project is still cheap in price, about 30 dollars total. Im still going to do it. I have also thought of an idea of how the fan can blow the air and not get the swirling affect. I actually plan on making it as soon as i finish this post. So wish me luck and ill try to get some pix up for u guys tonight.

[B4Maz]

These are all the bodies that Nic Case has made for his speed run cars.

http://farm3.static.flickr.com/2043/1858972737_34b3c29374_o.jpg

[B4Maz]

Yeah I looked at the spreadsheet calcs I have and If I put in the numbers for a 1/8th circuit car I know goes from 0-65 in 2.5 seconds it says it cant so I'm sure there is a flaw somewhere. I would have thought you could get 1.2G all the way along the straight in which case you would get 200mph before 1/4 mile is up!!On the spreadsheet, there is a box to input coefficient of friction. This will effect the overall amount of traction force the car will have. How much traction the car has also depends on where the CG is located relative to the front and rear axles and how high the CG is relative to the ground. Once all the other data is inputted and the CG is located and a coefficient of friction is calculated, a good estimate of how much traction available can be calculated. Keep in mind this is also assuming a constant acceleration.

[wqqdster]

"These are all the bodies that Nic Case has made for his speed run cars."


I.......I.......I just don't know what to say.