The DZ dummy could :-)

You can have my zipp 2001. It's a 700c bike and I ride a 54cm road bike so the frame size is representative of an 'average' size tri bike.

PM me and let me know how you want it built up/built down and where to ship it to.

---

----------------------------------------------------------------

My training

I'd want to see a P3 aluminum in there also.

---

Want: 58cm Cervelo Soloist. PM me if you have one to sell

Vintage Cervelo: A Resource

Guys, I appreciate all the input on this.

In no particular order...

I will be borrowing bikes. I will be setting them up as as identically as possible. Bikes of similar size normalized to stack and reach is how I am going about this. Same wheels and tires of course. Zipp disc and deep front and possibly Hed disc and Hed front if there is time and money for that many runs/bike. All will use something like Vision basebar and clip on (the same set and angle and brake levers and shifters and stem length) for all bikes EXCEPT the ones that come with a proprietary and unswapple set from the manufacturer ala the Shiv etc. Cables will be in place and trimmed or made to look as trim as possible to the tunnel on all bikes. Pedals will not be on the bikes but I will be using a standard crank aligned the same for each bike. Spacer stack will be accounted for. Obviously bikes with hidden cables won't be showing any cables they would not normally. Protocols will be professional and transparent. I will tell you all what I intend to do before I even do it.

I have no direct experience in the wind tunnel but I will be working both with the tunnel engineer as well as an outside consultant who has run these types of tests before. Got no horse in this race. I have followed every tunnel and aero thread and bit of info on this site and the web in general for numerous years so I would like to think I know what I am getting into.

I will update the list in my OP as to what bikes will be tested and which ones I have already secured. I will add the Look 596 and I already have offers of a Zipp 2001 and a newer beam bike as well.

As to expenses and pre-orders, let me get a little further along with the planning of this and I will let you know. When it looks sure to happen I will re-address this issue. Expenses for this will mostly be shipping and tunnel time and that will be very close to the time of the test itself. I have two excellent mechanics on the project and I am soliciting a few guys to help design and verify the test protocol.

And Flanagan, I am going to spend some time down there on my new Giant tweaking the position. I'll give you that info for free in exchange for me picking your tires at the state TT next year.

Since your intent is to compare the frames, I think you should probably do something similar to what QRoo did in their Cd0.1 testing with some minor "tweaks".

http://forum.slowtwitch.com/...=cd0%201%20;#2225340

In other words, test the bikes without bars. My suggestions for "tweaks" to that protocol would be:

1. Place a capped round tube (PVC?) over the portion of the steerer sticking out of the top of the headtube. Use and appropriate length tube and/or spacers under it, so that the vertical distance from the BB to the top of the tube is the same on each tested configuration. This would mean figuring out which of the "units under test" has the most stack and then working from there. This would eliminate having to play around with stems and such throughout the testing.
   
2. Use a consistent seat and seat position (again, relative to the BB) across all of the bikes. Alternatively an adjustable "blob" could be made to go onto the top of each seatpost to eliminate the fact that portions of the seatpost heads that aren't normally exposed in use ARE exposed during the testing without a rider. Again, the top of the "blob" would be set relative to the BB in a consistent manner.
   

---

http://bikeblather.blogspot.com/

Tom: how to you test the Trinity Alliance, shiv, speed concept with that setup? Also, you mess up the 596 testing (or is that stem removable?). Additionally, that means cleaner air hitting the top tube - doesn't that exaggerate the effects of a good top tube (to a small degree).

Well...with the Shiv, just leave the bars off ;-)

Seriously, though...you make a good point. But, that difficulty in comparing across those models with proprietary and not easily removable bar setups is going to be there no matter what. I was just trying to make a suggestion that would lessen the test time and variability.

Never mind...

---

http://bikeblather.blogspot.com/

There are a couple ideas I am considering for spacers. One would be to make them all the same as the highest bike but that would give the "no spacer" bikes even more advantage. Another is to see if it is possible to calcualte the drag effect of different stacks of spacers and subtract it out. Option three is to just lie to all my bike donors and cut everything flush. <----big joke there...you're bikes will be returned as you remember them. Actually better, as part of the protocol will be to test clean and waxed bikes.

Dave Luscan

I love the idea of testing clean v. waxed.

I always wax my ride before a race with Speed Shine. Takes <1min to apply so this, or a similiar spray-n-wipe product, may save you some time.

http://www.griotsgarage.com/...aners/speed+shine.do

-tim

X2 on the Planet X and Aluminum P3. A Zipp would be very interesting too, and what about a Lotus and/or Hotta? Those would probably be tough to come by, but it would interesting to see the results! I'd likely be willing to throw in some money to see the results.

Cheers,

Mike

It's a pretty common practice in kayak racing to wax beforehand - same concept, just denser fluid.

Hi Dave,

I'm a big supporter of this concept, but of course have to get the go ahead to share bikes. We don't have many Shiv's right now, so delivery would be based on time of year. I have a couple of questions and notes here:

1. Mike Giraud at the A2 Tunnel knows what he's doing, so I'd trust independent results from him. Who's the other expert consultant you're looking to bring in?

2. Yaw testing at A2 takes a long time right now. To do this test right, you'd need to test at least one bike 3 or more times throughout the test. Current protocol puts yaw sweeps at ~45 minutes with no wheelbase length changes. This would be ~7 hours for 10 bikes tested plus controls. This extensive of a bike test (to garner reliable answers) would take ~12-15 hours. So tunnel time alone, we're talking $6,000. Sounds like you anticipate this cost, but to be clear, I'd only support the test if the protocol is entirely scientific including tare sweeps and the re-running of 1-2 bikes multiple times through the test (whether it be one or more days long).

3. I'd simplify and run with only one wheelset that you test on everything. If you want to check any wheel stuff afterwards, finish your test and pick one or two bikes to go back and study. Any typical aero wheelset would work well -- but don't run with crazy narrow or fat tires. A solid 23c clincher or 21-23c tubular is nice.

4. Spacer stack can not be messed with or calculated out. I sincerly recommend picking a stack and reach you're trying to meet (i.e. a "typical" triathlete position for a guy who's 5'10" etc.) It's hard to be locked into one position, I know, but making changes or getting close does not work. Variations in stem length are less significant than stem angle -- i.e. if you go up higher with stack, even with a round steerer tube, you should lock into 1 angle of adjustment for the stem. Use one stem manufacturer and various lengths. Additionally, stack matters more than reach for aero drag in the cases you're looking to study. If you can only match one perfectly because of a frame's geo, match stack.

5. You need to cut steerers. Long cylinders put into the wind generate a lot of drag and you cannot accurately compare the drag of two bikes with different steerer lengths sticking up above the stem. If Specialized were to support this kind of a test, it wouldn't be a big deal to get you a fork you can cut the steerer on.

6. Cable routing matters -- all bikes should be fully ridable IMO. The mechanic just needs to do the best job he/she can when building the bikes.

7. Cranks -- crank arms should be tangential to the wind, or at least consistent for all tests. If you angle them up/down, the crank length matters a tiny bit.

8. 2x on using the same saddle and height. Should also use the same elbow pad, or no pad at all.

9. Back to 4. I would use a good stack and reach tool to measure and re-measure the bikes. The skeleton should be fixed and the bike be made to fit that position, not the other way around.

10. Good luck! This is great and very ambitious but all the best. Let us know more as you continue in the process.

And the wind tunnel aero helmet/mannequin thing I did with a friend as a senior thesis at MIT. We tested 14 aero helmets at 3 different yaw angles and 3 angles of attack. PM me your email and I'll send it over. Or someone tell me how to post a file up here. I keep forgetting how. We can get a thread going on that study. It was pretty extensive and showed the range and mean improvement with aero helmets in the marketplace.

Cheers,

Mark

---

--
Mark Cote
MITAerobike
Specialized Bicycle Components

I'd add one more bike frame to the mix...the BP Stealth or Martec frame design. This is a competitor to the Titanflex style of bike in the style of the old Lotus superbikes...and afaik there is no actual tunnel data anywhere on it. I ride a size medium and would happily send the frame/fork off for testing if you are interested, as long as it isn't in the middle of race season when you finally determine a protocol and run the tests! I'm actually featured on the splash page for http://www.bpstealth.com at the moment.

I'd be very interested in a third party objective evaluation of the frames. I'd considered going to NC myself (and still am thinking about it) for a tunnel session, but it would be valuable to me to know if my bike is decent/crappy/competitive with the best that's out there before bothering.

---

---

Sent you a PM in regards to loaner bikes. Have a Pinarello Montello FP8 and LOOK 576 available. Also have access to a few more such as Fuji D6 and Jamis should you wish to include them.

I would really like to lend help in this as far as being at the tunnel to provide physical man power for moving bikes, set-ups and documenting if needed. I sent you my contact info in an email. Thanks.
-Mike

Do you mean this study?

http://dspace.mit.edu/handle/1721.1/40486

That was linked to in a thread here last week on aero helmets:

http://forum.slowtwitch.com/...ing=helmets;#2557630

I wouldn't mind it if you emailed or PM'd me the "decoder ring" for the helmet identification though ;-)

---

http://bikeblather.blogspot.com/

Working on it. It would be sooner rather than later. I'll let you know about chipping in. I also don't know how the data would be presented, where, etc.

---

Brandon Marsh - Website | @BrandonMarshTX | RokaSports | 1stEndurance | ATC Bikeshop |

Assuming I have a road bike by next year I'll loan you my 596. The only thing is that it is a XS which probably is similar to a 51 P3C. I don't know what size range you are trying to stay within.

Exactly what I was thinking. Which is which?

---

Want: 58cm Cervelo Soloist. PM me if you have one to sell

Vintage Cervelo: A Resource

Dave,

My P4 (54) is gonna be hanging on the garage wall from now until early March, so if you'd like to have use of it during that time for this, let me know.
-M

---

float , hammer , and jog

Thanks for the reply. I hoped you would. I have been kicking this around for some time and running projected costs and just decided to go for it today. Mike Giraud has an email, and if he is interested, I would use him exclusively.

As to the tunnel time and $$, I am probably going to address this by culling the list a bit and making it most relevant to buyers of a new bikes who place primary emphasis on speed over style. I would rather test the fastest bike from each manufacturer, rather than all of them. We already have a good idea as to how the P2 compares to the 3 and 4 so why spend the time and money. Probably won't test bikes like the Ceepo Venom as buyers of those are not primarily interested in aerodynamics, IMO.

One wheelset makes the most sense. I was waiting for more specific info as to how long this will all take per run and you provided it first. Thanks. Question I would ask is, why so long for full yaw sweeps?

Spacer stacks is the biggest issue I have run into so far. I would love for donor bikes to be all the same but yeah right... I think I am going to go direct to the manufacturers and see what they have to say. Along these same lines are the bikes with proprietary seatmasts that need to be cut to fit. Daunting logistics to be sure but I intend to do this right or not at all. Leaning heavily towards doing it right however.

Advice taken on cable routing. I just learned that there will be plenty of time between runs and having two identical set up for this is probably the way to go.

3 hours passed since I started this post, gave me daughter anap, cooked dinner etc. Got an email from Mike at A2 and I should be speaking with him on the phone tomorrow.

Honestly my biggest concern with this is not going broke. If I knew right now I could break even and never make a dime I would still do it cause it ain't about the money. I think this needs to be done.

Dave Luscan

I definitely agree that you can not "post-process" the results in terms subtracting/adding drag for a spacer stack. The flow off the tip of the cylinder stack is complex and different frames will interact with the shed wake differently. If you are limited to borrowing personal bikes, I believe the most uniform way to test would be to set the test stack to the max stack out of your sample of bikes (as you have previously suggested). It should be relatively easy to raise the stack on bikes with steerer tubes cut too short (using some sort of temporary strut to clamp the stem to; the strut can be bolted into the star nut and would then be sheathed in spacers to maintain the same O.D. as the rest of the steerer). Hopefully the range between max stack and min stack is minimized by carefully choosing frame samples. This method, of course, will benefit new bikes such as the Shiv (since the desired stack will need to be reached using pad spacers instead of steerer tube spacers).

Chris

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Dave,

Sent you a PM about a loaner bike. I wonder if it won't be better to test bikes with either a flat or bludge disc depending on which is optimal for that bike, so we would determine the fastest overall system. A TTX for example is supposed to be faster with a bludge disc than a flat, and a P4 will only take a flat disc. Just a thought. That might make things too costly though. Sounds really cool.

Dave, I would add the Blue Triad to the list, I know they offer free wind tunnel testing at A2 with purchase of bike and that they have bikes for testing at A2 so it should be fairly easy to talk to Chance at Blue and see if they would be in.

Kevin

I've got a L transition you can use depending upon when you do it.
We start racing in Feb out here.

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

I've also been meaning to ask (either you or anyone else who knows) whether facilities (A2, LSWT, etc..) have the capability to back sweep at shifted yaw angles, so as to not repeat the same angles as the forward sweep. When I used to test in our tunnels, we would offset the sweep coming back (only repeating the zero angle point) to get a better idea of what the force curve looks like and more importantly better define any flow hysteresis that may be masked by sampling the same points.

It would admittedly be hard to tease out with limited sweep angle resolution, but signs of significant flow hysteresis could point to "reluctance" of certain stalled areas to reattach in dynamically changing yaw situations; usually a bad thing in the real world. I'm just thinking aloud here about other ways a bike can be better aerodynamically than another...ie; even though drag is lower in steady state for bike A, bike B may be dynamically better if it manages separation and reattachment better than bike A. This is probably a higher order effect though, as yaw angle would have to change relatively rapidly throughout a ride for this to matter.

Chris

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc