Think of it as more of a "gentle nudge/wink" ;-)

The fact that you put the crates in place to mimic the yet to be constructed twin control room points to the fact that your return leg isn't exactly "poorly designed" :-)

Lots of REALLY good stuff comes out of open-circuit designed tunnels.

---

http://bikeblather.blogspot.com/

Sorry I assumed you meant that the balance by itself was accurate to 2 gF. i.e. with a calibration rig or some known masses, not with the tunnel on and measuring the drag on something.

I have to admit, whatever you have done to get that level of consistency is pretty impressive.

In reference to the wind tunnel comparison I was curious if you had done back to back testing with the same bike in multiple tunnels and compared the results. In the auto racing world when ever a new tunnel came online the manufacturers would do a comparison test to the tunnel that was considered the "gold std." (for the mfg. I was privy to it was Lockheed in Atlanta) and compared the outputs for a given car. For instance the most dramatic differences I ever saw was front downforce for the Chrysler tunnel was reported to be a couple counts different than Lockheed due to the vertical return and how they did the laminar tripping in the Chrysler tunnel vs. the horizontal return at Lockheed. I was just curious with the emphasis on the test area size whether the data in your tunnel trended in one way at high yaws vs. data taken at the little A2, or whether the data mirrored each other pretty well across all the yaw angles.

That's a cool tidbit about the Chrysler/Lockheed tunnel comparison! Yea, I would imagine for performance cars especially, how ground effect is modeled is pretty important. We don't have enough data yet that I would be comfortable claiming seeing different yaw trends (and furthermore directly attributing that to blockage differences). Also, as far as cycling tunnels go A2 is actually pretty good in that regard which is why we tested there. It's the smaller aeronautical tunnels that I would be more wary about.

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

How much of the wind tunnel project was developed or learned from the relationship with McLaren? Will you guys share data with each other and review together? How does this affect the ongoing relationship with them in future? Besides the aero testing and technical know how, I know they also helped with the carbon lay-up on the McLaren Venge. Will this turn into more of a technical materials relationship since you are taking the aero testing in house?

We currently have a bunch of projects going on with McLaren, from composites to dynamics & vibrations. We'll continue to work on aero etc with them. The only thing that changes now is that we have a tunnel here and some more CFD and on-road tools. With McLaren, it's as much or more about their engineers and technologists as it is their tools. The relationship with McLaren is as strong as ever and we're all pumped on the next big things coming!

Mark

---

--
Mark Cote
MITAerobike
Specialized Bicycle Components

Hey Chris

I have always been curious about calibration of strain guages. Is that something you have to do before each new measurement or set up? Is there some kind of NIST traceability for that?

Thanks,
Mike

I just wonder how much of the inconsistencies between data sets of different manufacturers is based on different tunnels used. For instance the Mavic data showing different high yaw data for Zipp wheels than Zipp shows. It is possible that it is due to testing protocol differences, different equipment choices, but some of it may be a coloring of the data by the tunnel that is being used. I was hoping you guys had done some of the leg work to figure out the differences to try and help us all piece different information together into a coherent picture.


I have no experience with the small A2 tunnel, but from my limited experiences with the tunnel next door they are a class act and are constantly trying to improve the usefulness of the tunnel as a tool.

Yea, it's actually kind of like using a power meter. We zero the readings periodically during a test. For calibration (e.g. finding the slope relationship between applied force and voltage output of the sensors), we will do that when anything on the balance apparatus itself changes. It's pretty stable otherwise. We will also verify the cal regularly. We perform the cal with NIST weights but everything on the balance end is custom in house.

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Yea, the guys at A2 and Aerodyne are definitely class acts. I think you're right about the different tunnel facilities playing some role in seeing slightly different data from different sources. It'll be interesting to explore further but that may require some collaborative effort at other facilities (after all, we built the tunnel so we wouldn't have to travel anymore!).

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Hello Nate and All,

"Why do I have to put shoe covers on to make my shoe aero? I can't do that in a tri (but shoe covers save time).

We need some type of shoe with the ratchet in the heel (maybe entry point is in the heel too). The rest if the shoe should be really aero/smooth."

Suggest that you must consider shoe/cleat/pedal combination as an aero unit ..... not just the shoe.

http://forum.slowtwitch.com/...post=4558598#4558598



Showing smooth pedal retraction (like aircraft wheels) into shoe body

Covers not required for View-Speed Aero modified shoes (although a smooth top shoe like the Bont works best) and you can run on the shoes.

Saves 40 seconds per hour San Diego LWST = saves 4 min in 6 hour race.

Available .................. now.

Cheers,

Neal

+1 mph Faster

Nothing builds trust like the truth. Hopefully, you are the fastest most of the time. When you aren't, say so. That way when you are, people listen. The truth in a marketing driven, spin infested space would be refreshing.

More or less. If you stand in the warehouse room that holds the A2 tunnel you can actually feel a fairly strong breeze blowing opposite the direction of the tunnel flow. Enclosing an "open return" tunnel has advantages, particularly for testing humans (better control of environmental conditions), but I have to believe there's a fairly substantial efficiency gain too. A2's power requirements at bike testing speeds are LOW--under 50hp//37kw as I recall. Granted, they have a smaller TS, but I think the enclosure of the tunnel plays a small part.

---

__________________________

I tweet!

Hey everyone, sorry for the last second notice but we will be hosting a live Q&A session with video this afternoon from 4-5pm PST. Chuck Teixeira and I will be answering questions this time around. We'll do the same on Thurs but at 9am PST (I believe) with Mark Cote and Chris D'Aluisio online.

Link: http://www.specialized.com/...ws/latest-news/16025

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

I'm reminded me of a quote that said something like: all wind tunnels are closed loop. The difference is only the degree of efficiency of the return path. :-)

---

AndyF
bike geek

You're hoping for 2 gF for the bike alone? And you've been seeing 5-10 gF for the bike alone, or for the bike+rider, at these other facilities?

The best I've seen from a field test (with bike+rider, of course) would have been in the ballpark of 5-10 gF at 14 m/s.

Hi Robert,

That's correct, the target is 2 gF for no (live) rider and we have typically seen 5-10 gF at other facilities, again with no (live) rider. A really good/steady rider will get us numbers in the same range but a poor one can be a lot worse so the window is pretty big once you factor that in. As you know, the same is true for field testing - which as you point out requires a rider! So for R&D stuff, it may take a bit longer to hone in on the effect of a design change due to the additional variable. Not saying it can't be done, just using the different tools for different pieces of the puzzle.

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Are you talking about me again? :-)

---

http://bikeblather.blogspot.com/

Well, Chris said:

so, yeah, I'm talking about you, damn you.

:-) It's getting warmer here in the mornings and is getting into prime field testing season...I think I'll be stopping by the local craft store to procure some "standard shapes" to help calibrate my new half-pipe venue (as AC calls it, the "Tom Compton Challenge")...

BTW, and I don't recall if this was brought up earlier, but as you well know, one of the "advantages" of field testing is that the "power to rotate" of the wheels is included in the overall demand.

Chris or Mark, can you go into detail about how you guys plan to measure that value consistently and include it in your measurements?

---

http://bikeblather.blogspot.com/

An alternative is to re-visit the "delta mass" experiment. Since a Newton is a Newton, hold your CdA constant and run with empty vs. half-full water bottle (~200 or 250 ml or so, depending on what the Crr is). That should also be a difference in drag of ~0.1 N. If you can detect that amount of difference in rolling drag you ought to be able to detect that amount of difference in aero drag.

I don't know whether that's an advantage or a disadvantage -- I just think of these sorts of differences as characteristics of each particular method. But it is something I'd like to know, too.

Good idea. Even better would be if I'd have someone there who would change the water level unbeknownst to me so that when I analyzed the runs I wouldn't know ahead of time which was which ;-)

---

http://bikeblather.blogspot.com/

Ah, yes. That was a funny critique.

You needn't actually only put 200 ml of water in the bottle. You can put any mass in there (as long as you weigh it; how about something dense like krugerrands?) and see how closely the measured change in drag force matches the known change in mass. That's your standard and you won't have to worry about standard "shapes." Much cleaner experiment. You'd maybe want to change the test protocol into a hill climb and then compare the VE profiles.

Initially, we'll be measuring the electrical power that the drum motors are drawing. That, along with the motor efficiency tables, should get us pretty close. However, if that doesn't look like it's hacking it, we've designed the apparatus so that we can insert some rotary torque cells in the drive system. We'll also have load cells to measure the normal force applied to the tire. As I had mentioned earlier, this measurement will include rolling resistance on our smooth (and small diameter) drum so to tease out rotational aerodynamic drag, we'll need to either use the same tire and pressure and do comparative tests or measure Crr using a lab fixture with a similar drum (and subtract that value out from the total power-to-spin number measured in the tunnel).

Edit: I just realized you may have also been referring to how we plan on reporting it....well, good question. How would you like to see it? A total drag curve including everything? 2 separate curves, 1 including all and another including just the translational or rotational drag components?

And as Robert mentions, the different tools each have different strengths - the tunnel allows us to isolate the effects of different parts of the bike/rider system a bit easier (and going to the extreme, in CFD we can isolate drag contribution from different sub-components). Imagine having different parts (e.g. wheels or mannequin) on an isolated stinger so we still get the aerodynamic flow contribution but we can include or exclude its force contribution. So say we make a change to a wheel and the overall drag reading decreases by X amount. Is that X all from the wheel change or is it now interacting differently with the frame or rider and we're getting some benefit there? (or in an extreme case, maybe the wheel is actually higher in drag but the frame/rider has benefited enough to compensate).

Sorry I'm rambling now and shouldn't spill all the beans ;-)...

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

"Yes, technically speaking it's "closed.""

i don't see how you can describe this tunnel as such. this would mean that every open circuit tunnel indoors is really a closed circuit tunnel, and the only real open circuit tunnels are non-recirculating tunnels that are outdoors. is that what you're saying, or am i misunderstanding? not that i think it matters, because of everything you do to condition and straighten the air.

---

Dan Empfield
aka Slowman