Wind Tunnel Smoke Testing

[A2WindTunnel]

Smoke testing a Superbird

http://www.youtube.com/watch?v=s-t5WKkgMDk


D

[A2WindTunnel]

Ford GT durring a smoke sequence

http://www.youtube.com/watch?v=3EjFkPL_H_4

D

[A2WindTunnel]

It has been a long time since I have been on the site or done a vidieo but I had some time today.  This is Gene McKeen's 32 Ford in A2 for the second time since his build.

http://www.youtube.com/watch?v=8PoIoQNIoMQ

Dave

[Bville701]

That's awesome! It doesn't look like the air is hitting the rear spoiler though

[wobblywalrus]

A basic question about the superbird.  Would it be better to have the engine air duct at the very front of the car near the radiator instead of on top of the hood?  The duct on the hood seems to cause turbulence when air passes over the car.

[A2WindTunnel]

That's awesome! It doesn't look like the air is hitting the rear spoiler though

That’s the nature of a 32 Ford is that the spoiler is not working as effective in the wake of the cab and the center portion is not working as efficient as the outboard sides.  There were several configurations tried on the rear back in 2008 and even again this year to find an optimum design.  For those that don’t know, I cannot discuss specific details of our customer’s tests.  I was told by Mr. McKeen that I can post the video and vaguely discuss since I will use his car for an example a second article I will write for myrideisme.com 
If anyone hasn't read part one the link is http://www.myrideisme.com/Blog/race-tech-wind-tunnel-testing-1/

A basic question about the superbird.  Would it be better to have the engine air duct at the very front of the car near the radiator instead of on top of the hood?  The duct on the hood seems to cause turbulence when air passes over the car.

A forward facing scoop is a good way to get air but not a very efficient way.  The scoop will add drag but you would not have the pressure loss that you might see from running it at the front (ie more available air). On the Superbird, I think the scoop will only open at low speeds where drag is far less important than the HP needed to move a heavy car around the streets.

[A2WindTunnel]

Video of a landspeed bike in A2:  I think this got posted somewhere, but I could not seem to find on the site.  My apologies if it is indeed in another thread.

http://www.youtube.com/user/A2WindTunnel?feature=mhum#p/a/f/0/zG8Htax6_Og

D

[WhizzbangK.C.]

Video of a landspeed bike in A2:  I think this got posted somewhere, but I could not seem to find on the site.  My apologies if it is indeed in another thread.

http://www.youtube.com/user/A2WindTunnel?feature=mhum#p/a/f/0/zG8Htax6_Og

D

I have a question that I hope you can answer without revealing any proprietary information, and I hope that the answer will be helpful for other folks as well as me.

At about the 3:00 mark the smoke wand is placed into the air intake, and we see a lot of turbulence there. Obviously the engine is not running while testing is taking place. From looking at the smoke trail earlier it seems to me that there is a higher pressure area there, and that the intake is pretty well positioned to take advantage of it. Extrapolating from that, it appears to me that if the engine was running as the test was being done that the results would show quite a bit of difference at the rear of the bike, both from the air being sucked into the engine, and from the exhaust filling in the lower pressure area at the tail.

My question is would it be worth the effort, for a team  preparing to test in a tunnel, to come up with some way to simulate the intake and exhaust air flow? It seems that this could be done, possibly by using a high volume electric fan to draw air into the intake, and mounting something like a CO2 bottle inside the body work with a solenoid to dump it into the exhaust pipe to simulate the added volume of combustion by products to the exhaust flow.

To my eye it looks like this would provide more realistic data, but I'm a complete novice at this kind of thing, so I'm asking. 

[joea]

whizz, what about the confluence's (sp) ie impact of the missing ground effects and
air coming off of spinning wheels and its important impact on the entire system...?..

[WhizzbangK.C.]

whizz, what about the confluence's (sp) ie impact of the missing ground effects and
air coming off of spinning wheels and its important impact on the entire system...?..

Good point, there are "rolling road" wind tunnels to at least partially address this, but time in them if you can find one is a lot more costly than at A2 or similar facilities. Everything is a compromise and I'm pretty sure anyone going into tunnel testing realizes this. I was just thinking that this is something that a person could build on their own and fit into the machine as part of the preparation for tunnel testing, to get a more realistic idea of what's going on with their machine within the limits of the facility.

I would think that some basic calculations would tell you how many CFM you need to be pulling in through the intake and pushing out of the exhaust at a given RPM figure to match the air speed in the tunnel.

I wonder if it would be possible to build an external setup to do this? Ducts laid along the floor straight back from the rear tire connected to blowers calibrated to deliver the air flow needed to simulate intake and exhaust, with the blowers themselves outside the tunnel. The ducts would be in the wake of the tires, which is already acknowledged to be a compromised area of the test since the tires aren't turning.

Maybe that's something that A2 could build into the tunnel to offer more realistic testing, that may give another option to teams. The team would than just need to build ducts into the vehicle that would hook up to their hoses. Blowers could be turned up or down depending on the displacement of the engine.

Something to think about anyway. 

[A2WindTunnel]

My question is would it be worth the effort, for a team  preparing to test in a tunnel, to come up with some way to simulate the intake and exhaust air flow? It seems that this could be done, possibly by using a high volume electric fan to draw air into the intake, and mounting something like a CO2 bottle inside the body work with a solenoid to dump it into the exhaust pipe to simulate the added volume of combustion by products to the exhaust flow.

The Cup teams actually did a lot of development several years back where they had a wind tunnel "engine" that internally had a fan that would pull the correct CFM through the cowl/intake and exit the exhaust and what they found is that it might change the absolute numbers slightly (around 1%) but the incremental information was still the same. i.e. a change on the car resulted in the same incremental change with the engine pulling CFM or not.  The reason.. At 200mph think of how much CFM is going over the car.. The amount going it the engine is a very very small amount compared to that and that so you wont see a big change in data.  The Cup teams stopped running these "engines" because they were expensive and time consuming and they didn't find "better" information with the use of them.  Instead now they just block off the cowl inlet. 

Now, would it make more of a difference on the Kolb bike since it’s in a different location? It might, but they would have to build a fan to match CFM and pull air in and duct out the exhaust to find out.  As I said, I would expect the absolute levels to change a bit, but if you changed something like the windshield shape I think the Delta (data change) would be about the same.  It would just take a lot more time, money and effort for the team to make the simulation better.  We have seen people actually run the engine to see for them self’s, and don’t beat me up, I know the engine NOT under a load at XXXX-RPM is not pulling the same CFM as it would be under load (on track) but they just wanted to pass air into the system and out the exhaust to see if it had any effect, but the teams that have done it found virtually no change in absolute data.  Will that be the case 100% of the time? No, you might find a bike that it makes a significant difference on because of all the variety and inlet/exhaust configurations.  I personally would have a team focus on the testing goals, but if they wanted to fan and duct the engine/exhaust and had the time, $$ and effort to do it I wouldn’t tell them not to, but I think they could get better use out of their time and money with other things.

[A2WindTunnel]

whizz, what about the confluence's (sp) ie impact of the missing ground effects and
air coming off of spinning wheels and its important impact on the entire system...?..

There would be a difference with spinning wheels and wind tunnels are just simulations (some better than others).  We could build A2 with a rolling road, yaw and much more, but the cost would have to go way up to pay for these systems. AeroDyn (our other tunnel) with a boundary layer control system and spinning wheels is $1690/hr and Wind Shear (180 mph rolling road) is $4000/hr.  A2 was designed to be an affordable option for teams just like Bonneville racers that know aero is very important for what they do, but don’t have the multi-million dollar budgets.  And you can learn lots of usefull information in A2 that will translate into speed on the track.  (GM, Chrysler, & Ford wind tunnels don't have spinning wheels or a rolling road)

I do have a design ready for a platform (in A2) similar to our bicycle platform (attached) that would allow for spinning wheels on a motorcycle and Yaw capability.  I had a poll out on this site about a year ago to see what the interest level would be for you motorcycle guys and I got No response.  I can’t go sell the idea to my boss if there is no interest but if I got a group committed to testing I could move forward with the project.  With a system like this you could then test more accurately wheels, fender shapes and distances from wheels along with yaw.

Link to a video that would give you an idea of what this system would be like:
See Here: http://www.youtube.com/watch?v=JJP9A6VAWpY

Bicycle Platform in A2.

[WhizzbangK.C.]

Thanks for the reply.  I can certainly see how it wouldn't make a dramatic difference on a Cup car, since the air intake is not projecting up into the air flow. Makes perfect sense.

In the case of the Kolb bike, and many other vehicles that run on the salt, where the air intake is protruding into the air stream ahead of significant areas of body work, it seems that the difference would be much more pronounced, especially when the air intake is positioned to draw in some of the turbulence behind the rider and hopefully allow better re-attachment of the airflow to the body behind it. (I don't know for sure that's the theory that Kolb is working with, but it is on my bike, which has some similarities, hence my interest in this.)

It also seems that the effect of exhaust flow filling in the low pressure area behind the bike would be much more dramatic than on a car with the exhaust exiting out the right side.

I didn't know that it was an option to run the engine in the tunnel, that would seem to be the simplest way to make a quick comparison to see if more in depth study was warranted.

[A2WindTunnel]

Thanks for the reply.   I can certainly see how it wouldn't make a dramatic difference on a Cup car, since the air intake is not projecting up into the air flow. Makes perfect sense.

In the case of the Kolb bike, and many other vehicles that run on the salt, where the air intake is protruding into the air stream ahead of significant areas of body work, it seems that the difference would be much more pronounced, especially when the air intake is positioned to draw in some of the turbulence behind the rider and hopefully allow better re-attachment of the airflow to the body behind it. (I don't know for sure that's the theory that Kolb is working with, but it is on my bike, which has some similarities, hence my interest in this.)

It also seems that the effect of exhaust flow filling in the low pressure area behind the bike would be much more dramatic than on a car with the exhaust exiting out the right side.

I didn't know that it was an option to run the engine in the tunnel, that would seem to be the simplest way to make a quick comparison to see if more in depth study was warranted.

Cup cars yes, but remember we also test Pro-Stocks (and others) with a hood scoop and there have been studies there as well with inlets.  Most teams will just cover and tape them off.  And when I say teams have run the engines I am talking cars and motorcycle both.  I agree that not all cases are the same and the simplest way would be to run the engine (which you can do in the tunnel most cases) to see if there is a big difference and further study would be necessary. We would have taped off the inlet on the Kolb bike, but the team had the intake instrumented with pressure taps to get readings to see if changes not only helped drag but what it did to the pressure inside the box. And as I always say, I cannot talk about what they learned or tried at their test because all of our customer’s information is proprietary.

I will tell you that exiting exhaust in the wake will have little to no effect unless it is something in the realm of a top fuel (i.e. high HP engine).  Imagine whatever you race in a wind tunnel like AeroDyn (or driving through a tunnel with same cross sectional area for those who don’t like to think that way) at 130MPH there would be 2.5 MILLION (2,500,000) CFM going by the vehicle.  Now tell me how much CFM your engine puts out and what the percentage difference is? (that is only at 130mph)  Do you kind of understand what I’m getting at?  I’m not saying don’t make the simulation as close to the real world as you can.  If you think that for your test it would be important then go right ahead, but I’m just telling you (my opinion) what I wouldn’t waste my time/$$ on.  For a first time customer visiting any wind tunnel I would not worry about CFM in or out of the motor under most circumstances.  Now, if you have tested a dozen times and now starting to look at things through a microscope then things like this might be worth perusing.  The reason, like I said before, lets say it changes the absolute values 1% but at the end of the day you helped the drag 15%. Now let’s tape it off and run the test over and you have improved the drag 16%. Was it worth the effort? 

Wind tunnel time is not cheap, and to get the best bang for your buck you need to define what is important and not important for the simulation.  No wind tunnel is a perfect simulation and there is always going to be a compromise (always), but as long as you understand what that is then you can account for it.  If you had an F1 car dumping exhaust into the diffuser then I would tell you to absolutely have the right CFM in and out, spinning wheels with a moving or simulated ground plane (boundary layer control).  On a low HP motorcycle like Kolb I don’t think there was anything wrong with the inlet left open testing for air pressure and wouldn’t expect much of a change if we taped it off or ran the correct CFM through it.  At the end of the day the team would have seen the same improvements within a few percent. 

Thanks for the good questions. 

[A2WindTunnel]

 Two of the most popular ways on scoop testing:  The cone in the first picture simulates what the air actually "sees” as the inlet builds pressure the air slows down resulting in a high pressure at the inlet.  The same type of thing can be said for the plate which you won't see (because we cant see air) but as the pressure builds on the plate (more in the center or stagnation point) then the air pressure will build up similar to the cone shape.  The cone shape might be a better way to simulate, but unless you know the shape/length through CFD or another study then a flat plate would work just fine.