I'd like to get some thoughts on a protocol to try tease out CRR from CDA and whether anybody has done something like this and what their results with the protocol was.

Basically it would be to do a bunch of runs in an aero position - call this position A, and then runs in a non-aero (sitting up position), call this position B.

So the set of Runs in A will give you various combinations of CdA, and CRR that satisfy virtual elevation for those runs
The Runs in B will give you combinations of CdA, and CRR that satisfy virtual elevation for those runs. However we know that runs in A and B should theoretically have the same CRR to "link" the sets of equations such that only one set of CDAs and CRRs satisfy all the equations.

If you did the runs at different speeds of A and B as a traditional mechanism for teasing out CRR from CDA, that would then probably further assist - as you would have various further constraints that would assist in converging on the "true" CDA(A), CDA(B) and CRR i.e.
CDA (A)slow = CDA(A)fast
CDA (B)slow = CDA(B)fast
CRR = CRR (A)slow = CRR(A)fast = CRR(B)slow = CRR(B)fast

Haha, I don't do the "vary speed to tease CRR" thing.

I use representative tarmac at my testing area where there is some kind of "useful" amount of vertical grade. Then I go up the hill or small rise in aero at a very very very slow speed. Like 4 to 6 mph.

Doing so, I pretty much know my CdA from other testing and also know that having that "pretty close" would mean it's really good enough at such a slow speed. From this, CRR becomes dominant and any error in my actual CdA largely irrelevant.

Days I do this, I weight myself and kit and bike on a scale also altogether.

Right, wrong, indifferent...........that's how I do it.

From your description it sounds like speed variation is exactly at the heart of your protocol, but yes I get you on the nuances of your method vs what is traditionally described

Your method surely requires very accurate elevation though?

My problem is I am testing on an outdoor velodrome, and CRR seems consistently really high but I'd like to think of ways to validate it.

I have had mixed results with "shen" method.

Outdoors? On the road outdoors I have seen my sensor that cannot read wind at yaw cause what appears to be a CRR high error because it cannot decipher the airspeeds when wind is "at yaw" like that. The same could occur on a velodrome.

With the wind your body may shield the sensor some versus against the wind, so despite having an oval shaped lap seeing all the wind every lap..........it may skew the CdA.

Pure head/tail wind I've seen it do better.

Spent a whole $25 on like 5 sets of aero socks and shoe covers from Aliexpress. Excepting maybe a few sellers, everybody else is using the decade plus old designs and pricing them like they just spent a year in the F1 wind tunnel. When they are just standard copies. Honestly and sadly, they all fit better than a few name brand ones I have tried.

Like the shoe cover doesn’t have the seam sewn for an aero trip like a Nopinz or Aerocoach flow cover but the material of the shoe part itself conforms better to the shoe. The Nopinz ones have a huge seam right up the middle of the toe that lays funny.

Maybe do a long zone 2 day come spring to test them out.

I gave up on shoe covers for a few reasons

1) I could never prove they were faster or at least not within margin of error
2) I have big feet so typically need to order xlarge size. But I have calves disproportional to my feet. I guess people with my shoe size probably have much bigger calves. So I rip them putting them on, if not they float around my calves.

Yeah, my Aerocoach ones and these Aliexpress ones both fit my size 10.5 Giro shoes fine.

Either way, now I have a bunch that I can wear them for club runs always and have no worries about wearing out my "race day" kit.

I doubt I'll find a major difference. Just gain that whole couple seconds for each club run come spring time.

Sometimes I wonder if this thread isn't me, you and Ryan now and then.

Sometimes I wonder how much people really care about aero.

What do you think ?

IMO?

I feel like aero is largely a consumerist venture in cycling like everything else is. A small subset of folks have tried and tested things out over time and generated a pretty good go-to shopping list for folks. Then it's a matter of "set wallet on fire, go faster" for the majority of folks.

Folks want the easiest pathway to success. And a lot of endurance sport jocks don't want to be bothered to think or problem solve. They want to spend money or just "grunt sweat mash harder" to win.

Ignoring equipment, but going to bike fit, I see the same phenomenon. The Facebook TT bike fit group is a prime example. A ton of those folks on there don't rub me as being bothered to spend the time to go generate data. They want someone to look at their photo, tell them the fit coordinates to change and the helmet to go purchase, so they can go faster.

I make good money, but not "wind tunnel splurge" money. I don't have infinite training time, but I can log data and make training time on the TT bike double as data gathering ventures that "might" make me go faster.

There isn't a legal equivalent to "set wallet on fire, go faster" with draft legal events or things like cyclocross and gravel. Some, yes. But largely the drafting or skills nature means it's a brute force "mash pedals harder Thor!!!" or an acquired skill thing. So, your options are optimizing your time spend on training to win. Since you're only fighting largely the competitors, not aero drag. So you see folks blow money on pain caves, coaches, etc.... And you have the folks who outright cheat and use drugs or hidden motors also.

The issue I'd have with making time trial only Merckx is that IMO it takes away a prime problem solving thinking aspect that I enjoy. I would take up gravel only and maybe racing SCCA cars or something like that.

I check in on this thread daily.

Aero is a hard sell bc you can't see it, touch it, show it off to your friends. people would rather spend $1200 on a set of wheels that might make them :20 faster or :20 slower over 180km than spend $800 on aero testing that might make the 5 minutes faster over 180k.

It's only when you ride 5 min faster on 15w less that you can show it off and how often does the average person race? The people who test tend to race more often.

The market for aero testing, the amount of people that should test is HUGE. The subset of people that will aero test is tiny.

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

The thread title is misleading, until the cost of an aero sensor is comparable to the cost of a powermeter, your average cyclist isn't able to contribute much to a thread like this. Just about every bike and wheel maker can say their product is the most aero because there are so many variables to testing, yaw angles, wind speed, tested with a dummy or a half dummy or with or without a water bottle/s. I don't race much just local TTs and a tandem. But I do wear a aero helmet, calf guards/aero socks, I use 38cm aero bars and all the holes on my bike are taped over. I can't really contribute because I don't have a aero sensor. Maybe it's time for a thread name change?

I debate the cost thing, as folks blow $2500 on some ENVE for a difference not detectible versus $1500 in some HEDs.

But you're 100% right on the topic title not bringing in enough folks, maybe we should update to a topic title to something that's both sensors AND aero in general. But NOT a bunch of random useless "give me free speed looking at my bike fit for free" nonsense.

For me too the thread name is somehow problematic.

On one side I think the current aero sensors are a dead end when it comes to aero testing a cyclist, on the other side I don't want to be always negative.

Therefore I mostly only read the thread, even though I am very interested in cyclist aerodynamics.

I was thinking exactly the same. Maybe the thread should be aero testing which would include aero sensors.

It could cover aero testing and protocols, results that people perceive or have measured as well as experiences in tunnels, velodromes and road testing........

The OP probably reads this thread so maybe he could change it.

As for the comment on the cost of sensors and testing, I'm all with you.
I am hoping 2023 will bring some change :-)

Why do you believe this ?

Oh, that‘s complicated. For me it is not a financial question, I already bought more expensive gadgets.

First, who wants to ride around with such a bulky sensor mounted in front of the handle bar? I not. When I am cycling outside I want to train or just have a nice ride and I don‘t want to ride so that the sensor works.

The friends with whom I aero tested (generally velodrome testing) liked to have a fixed date for a test session with as many runs as reasonable in order to be dialed in for the upcoming season. The likelihood for good stable conditions should be high for this date in order to not waste time.

What more information give‘s such a sensor one can not also get from a speed sensor? Ok, a good and calibrated sensor may give air speed and yaw with high precision, but how to compare two runs when the distributions of speed and yaw were different? I don‘t want to compare apples with pears. By averaging and / or extended ABABAB testing errors do not necessarily become small because it takes time and conditions may change.

Among other things but by following this thread I got the impression that it is at least difficult to get precise aero results, i.e. the famous cda, with it. However, it has to be more precise than my simple feelings.

I care about aero very much and always make sure to read your posts on the rest of the forum as well Burn the sheep

Well you know from our couple of PMs I do like to read about aero

Not sure I am smart enough to do the Ching method

Add to the fact I am currently obsessed with underwater filming my swimming so as to improve and I spent a bunch of money on new wheels, another bike trainer, Omni one etc etc lol

Thanks

Sadly I think your list is right on target

It also describes me 😞

But as I say above I am obsessed with underwater videoing my swimming

Maybe in a few months when I finish that project I will start on bike aero

I KNOW that will yield better gains for me

But seriously the math, program etc just seem so complicated every time I read it

 
I agree that hanging expensive devices off the front of the bike and expecting Joe triathlete to figure out a consistent CDA is going to have limited success in the near future. I would not want to be a sensor vendor betting on this model.

I do believe aero professionals have a big role to play in getting actionable data to more people. But I don’t believe the one time trip to the aero tester is the holy grail. The whole aero testing model needs some innovation.

I believe the tech has evolved. I also believe that not all vendors are at the same stage.

But I believe how the tech is delivered is more important than the tech itself. And this I don’t think is being addressed. Yet. I have some ideas and am looking forward to 2023.

Back to the substance.........

Any opinions on me doing a little experiment buying a apparent wind direction indicator off Amazon and attaching the Notio to it on the mount so the Notio can rotate freely to coincide with the resultant wind passing over the sensor instead of being fixed?

Something like this:

That will not solve your challenges.

You need to know air/wind speed in the direction of travel.

If you are getting an incorrect wind speed in the direction of travel that is a different problem.

If you want to know yaw, that would be a different problem.

I also follow this thread, and since it has turned somewhat into a philosophical discussion, will throw in my 2 cents. This is from 76 year old who started racing in 1971, but now only does TT's to prevent broken bones.

1st the bad - I abhor the cost. For sure, there is no such thing as "cheap' racing of any kind, but there comes a point where it becomes detrimental, especially for attracting new people into our sport - we are reaching that point (to be competitive).

I like everything else. I find the speed addictive. I like that it keeps me in shape - I may not live longer, but I certainly live better. I like the competition. Lastly, I like toying with aero - it's one of the main reasons I follow Slowtwitch, but this seems to be the most controversial/complicated - witness the sheet show that occurred here when they tried to windtunnel test some bikes and report the results.

I guess I've adopted what would be considered a more ZEN approach, similar to what s13TX describes in the "Fast Tires - 40K" thread. I've been riding a long time, so I just listen to my body - No Aero Sensor, No Power Meter, No GPS, No Speedometer, No Heart Rate Monitor - just a stop watch and a 4 mile long, well paved bike path that I'm fortunate to have near my home. I ride 5 days a week, and do a "timed" 8 mile out/back run on a regular basis, and log this along with position, equipment, and weather changes. Even with varying wind speed and direction, over a long period time you'll see patterns develop - I obviously opt for what is fastest.
That's my story, and I'm sticking to it.

Surely, the forward groundspeed will just mostly keep this sort of straight anyway unless a really strong and conssitent crosswind. And I suspect will just introduce a ton of noise with the thing spinning right round.

Marc asked the question a while back - is airspeed measurement really the limiting factor on the notio for you currently?

It was just a silly idea after it was posted the thing becomes a random number generator if testing at yaw and you need a road mostly straight into and with the wind.

Also, the averaging time mostly fixed my personal issues. It just would be nice to be able to test on days with some yaw on favorite roads.

Also wish in general the sensors and stuff didnt depend on such long time averages as that makes testing a bit shit for the non velodrome folks as you need a lot of miles for each test.

I mean, your cda is your cda in reality the whole time you are in position so someone should figure out the sensor timing and calcs so as soon as you are at speed your cda is good to go. Not a minute long average after you have been at test speed forever.

I feel elevation slope accuracy to the centimeter per meter bike travel would be the holy grail. They tell you to pick perfect roads for a reason.

I imagine lasers, one pointed straight down to road surface at front tire and another shot about a meter ahead of bike. Geometry tells you the coming change in road slope. I already posted that way back im this topic as a hairbrained idea.

If I had the top 10 things I'd want these devices to do, 1 being the thing I want the most and 10 the less important, yaw would probably be somewhere like 6 on my list. But when you don't have it, you always wonder "is yaw the issue ?". I'd say 9 out of 10 times I look and "nope that's not a factor", but yes it's great to have.

I would put accurate elevation #1 on my list and this is why I have "obsessed" on it as we have been doing some development. We are an order of magnitude better than we were but still room to do things. I don't think your idea is hairbrained. Maybe hard to implement but probably fun to try. There would be a few challenges, some obvious, like the wheel getting in the way, others like the extensions are not installed at a "fixed" angle during a ride. We do have a board with a 2m +/-1 mm proximity sensor. We use it on the drone to test barometers. We point it down to the ground and measure the barometer tracks to true altitude. We also used it in TTT testing.

As far as the window in which a CDA can be calculated, one can do way better than that 1minute rolling window. That is probably 2 or 3 on my list, but not for the need/want of "real time" CDA but more for the calculation of a standard deviation to quantify the confidence in a CDA measure.

I may make a top 10 list just for hoots. I'll get my development buddy to make his and we'll compare. I suspect they will be the same list, sorted in opposite order :-) He'll probably put something silly like "not have to press 50 keys to get a measure"....like really who cares about that :-)