I forgot to mention that I'm using gF units since that's what everyone is familiar with (and honestly, it's more visceral...people can imagine what it feels like). Anyways, the number I quoted was for 50km/h...so scale appropriately for different speeds.

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

*Very* convenient in this case ;-)

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Yup, that's correct. We're running variable speed fans rather than variable pitch to get the response time that we want. At 50km/h we're drawing roughly 100 kW total across the 6 fans.

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Hope the operations guys don't notice the hit to the electric bill and shut you down.... Ha ha.

100kW x 12hr per day x .15 $ per kW x 300 days a year is not a small bill.

I saw you've got plans for a solar panel.

Aero test request!

Are you guys willing to release data on partners, even if they "lose" in the aero test?

If so...

Can you test the Flo disc/60 combo with conti 4000s 23mm vs zipp firecrest disc + 404 with zip tires?

Disc value opening tapped on both. Same skewer.

Do this on a s-works shiv please :).

*actually, any wheel combo testing with the shiv and venge would be appreciated. Change out tires too to hit the most popular ones and the 'aero' ones*

Also, please test bottle between the arm mounts on the shiv with the specialized carbon shiv aero bars.

---

CEO at TrainerRoad
Co-host of the Ask a Cycling Coach Podcast

As most of us STer's products are already tested in a rented wind tunnel, I am most excited by the potential for webinars. My question is will the tunnel be solely used for R&D of products or will it also be used as comparison amongst top brands for marketing purposes? If it will be used for comparison purposes, how do you plan on making it so no one can claim unfair testing protocol? Do you suspect product prices will drop since there is no longer the overhead of renting tunnel time?
Thanks,
Dave

Haha, fortunately our operations guys are awesome and yes, we're looking at ways to partially or fully offset the fan power draw so we can claim to also be the greenest wind tunnel! But in seriousness, it's not a small amount but that number equates to just a handful of test days if we were to travel to another tunnel to do a test (accounting for rental, travel and shipping costs - not to mention engineer time away from the office).

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

You got it...well I can't promise every single combination (because we do have to buy our test samples!) but we will try things that are either in popular use or make sense and show you all what comes out. Yes, we run into the possibility of something testing faster than what we make - in that case I hope to be able to argue why we may still have the most beneficial or advantageous setup overall, but if not, hey at least we learned something!

Edited to add: We want to get all this out to you guys asap but we have a queue of stuff we need to get to (like I said, everyone here at HQ is waiting to test something). I'll keep the ST crowd posted when we start doing the fun stuff...

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

That's a good question. To be honest, I'm not sure how the product marketing will happen. Up until now, for aero products, it's been a graph or two with the product release. We can brainstorm on what would be more effective for you guys. Maybe we do a video documenting a test? Either way, tunnel testing is becoming more and more common - many magazine reviews are even including test data (Velonews, Tour Mag.). So I'm sure there will still be independent data out there.

As for the product prices dropping - I'm definitely the wrong person to ask about that one! But it's likely a question of how much more can we do for the same R&D dollars as before rather than how much we can save.

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

This is great guys...keep 'em coming! I know Mark (MITaerobike) has been running back and forth like crazy the last couple of weeks but he promises to come on to answer questions as well!

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Crazy Idea: aero crank arms

Does the shape of the crank arm have any affect on drag while pedaling at multiple yaw angles?

Crazy Idea: aero shoe

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.

Crazy idea: big fat shiv seat posts

Seems like there would be a lot of crazy air around your seat post while pedaling. Can you dramatically change the shape of the seat post so that it decreases drag while pedaling?

Simple idea: aero center pull brakes

This one should be easy(ier)

---

CEO at TrainerRoad
Co-host of the Ask a Cycling Coach Podcast

I've used shoe covers for an IM. Pre put on ver the cleat then just pulled ver the heal. Probably took 4 seconds if that. Although there are some very aero shoes out there, ie Giro Empire, but I'd tape over the laces.

---

Styrrell

What air speeds are available to you for testing? I ask this because everything that I've read that you and other companies post is ~30mph, this isn't really indicative of your average triathlete's speed.

Do you plan to set up a PIV system (DEHS or similar)?

---

http://cds-0.blogspot.com

Nice...that's a good list of ideas! Can't give away everything that we're looking at, but I will say that looking at crank arm shape is one of the reasons why you need a dedicated balance with high precision - at the tunnels we have gone to, any changes we studied have been within the noise. On the topic of shoe covers - there are actually shoes out there that test fast without them (you would be hard pressed to find a cover that made the shoe faster). The new S-Works road is one of them...

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Yea, 30mph or ~50km/h has been the standard. And really, scientifically speaking, it's perfectly valid. In the speed range of any cyclist, the flow doesn't qualitatively change - the forces just scale with velocity (actually velocity squared). So by testing at a higher speed you get a cleaner signal since you're higher off the noise floor. Also, in most tunnels 50km/h is about the min. that you can go. Anyways, we can run our tunnel reliably from 10km/h up to 100 km/h (at the low end, the balance needs to have super high resolution and be ultra precise to tease out any changes).

Edit to add: To directly answer your question, we will run tests at lower and more typically attainable speeds to demonstrate what I pointed out above (heck, maybe I'm completely wrong and there's some crazy flow phenomenon that happens...wouldn't be the first time).

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Yeah, good visualization techniques will definitely be a part of the process. We have a high freq., multi-component pitot that we can use to do velocity and pressure mapping and will be looking at PIV systems as well as other high speed flow field viz systems. It's all a question of whether the juice is worth the squeeze - for example, simple micro tufting is still a really powerful and cheap tool.

---

Chris Yu
Applied Technology
Specialized Bicycle Components
@chrisyuinc

Hey all,

I've been dormant on this site for a while....well now you know why. Chris, myself, and the team here has been jamming on this windy hallway for a while now. I'll do my best to be present for questions but between Chris and I, we'll make sure we're available.

On this question:

What air speeds are available to you for testing? I ask this because everything that I've read that you and other companies post is ~30mph, this isn't really indicative of your average triathlete's speed.

We can test at all speeds from ~5-60 mph with fairly good data, but within the cyclist airflow regime, generally testing at 30 mph doesn't mean that we're only considering this speed. We capture wind speed (truly, dynamic pressure) and use drag measurements to calculate the riders coefficient of drag (or more detailed, their Cd * Frontal Area).

We run at 30 mph, as many people do for these tests, because the forces on the balance are a bit higher at the higher speeds (versus testing at 20 mph) and we can get better measurements. Then we use the coefficients and calculate what the drag or power req'd to overcome that drag would be at say, 18 mph or 24 mph, etc.

We'll of course be looking at running true speeds in the tunnel for an athlete as the wind will feel more realistic. Many are surprised what 30 mph feels like until you're sitting in the tube and catching all of that air!

Goal is to make this real -- so it feels and relates to cycling.

Mark

---

--
Mark Cote
MITAerobike
Specialized Bicycle Components

Last question, I promise

- I know that the industry standard is testing with a wheel rotation speed that matches the WT speed (bike-air relative speed) but have you analysed the effect of wheel rotational speed in both wheel and induced drag? Once you set your yaw and WT speed, there is an "infinity" of real world wind directions and bike speeds (and obviously wheel rotational speed) that can produce that yaw so even if you are in the same Re regime, the different rotational speed can modify the system drag.

---

http://cds-0.blogspot.com

Thanks! I ride a sportbike, so very aware.

Awesome! I have an sworks road on order already :)

---

CEO at TrainerRoad
Co-host of the Ask a Cycling Coach Podcast

You guys need to start aero.specialized.com. Make it a blog dedicated to the wind tunnel and aero testing. Like you said before, show the results even if you aren't the fastest. Educate me as a rider why this stuff matters and what to look for. You'd create a bond between the company and the consumer. Your marketing department should be all over this :).

For the CFD stuff you mentioned that specialized uses its own servers for this. Why did you guys go that way instead of aws it azure. Looks like you can spin up some huge aws servers working in parallel on cfd.

That might reduce the time it takes to do your cfd modeling and save some money. If you already have the servers though its a hard sell to management.

---

CEO at TrainerRoad
Co-host of the Ask a Cycling Coach Podcast

THIS

 
You said that building a test section with a large cross section was easy(er) because of the low speeds. My question is why soooo big? It seems that it wouldn't be so hard to test everything in a tunnel half that size. The only thing I can think of is analyzing longer test subjects (Like a TTT mentioned above).

---

___________________
"TRIATHLON ISN'T ACTUALLY THAT HARD OF A SPORT" -ALISTAIR

The largest issue we've had with other tunnels testing bikes has been blockage at higher yaws. A tunnel should only have 5-6% blockage in frontal area to represent good freestream velocity without accelerating air flowing around the test subject.

So at 16' x 10', we have a tunnel x-sec of 160 sq ft. Our splitter plate takes up about 16 sq ft of that, so we're at 144 sq ft of testable area. 5% is about 7 sq ft so you actually do need a pretty big tunnel to do good athlete testing at yaw.

We also want the ability to test a bunch of other stuff too, so a big tunnel is great for future crazy ideas. Honestly, I've been pretty obsessed with the idea of getting some wing suits this week...just ask Chris!

---

--
Mark Cote
MITAerobike
Specialized Bicycle Components