I did not try S mode as this means no obstacle avoidance. It has slower maximums when avoiding obstacles. It has it's quirks, especially on accelerations.

I do need to try it more. To be honest I put self filming aside pretty quickly. But if Tufts/Drone testing is the next frontier I am willing to do it for science.

Youtube even has a video https://www.youtube.com/watch?v=DJ4oQ03HqyE

Next week I am up at the country place. I just need to fly it 1m above the occasional oncoming logging truck and I should be good in S mode.

Yeah, if you live in a spot where nothing will get in the way of the drone, then you won't have any problems at bicycle speeds with the Mavic Air 2 and sport mode enbabled (which disables all obstacle avoidance). Though, it'll by default slide slightly back then directly off to your side.

Skydio can stay directly off to your side, at high speeds (for a bike), and with obstacle avoidance.

---

-
My tiny little slice of the internets: dcrainmaker.com

When I bought it the Skydio was still in pre-production sale so it would have taken a few months. I also considered how much weight it can carry

When aero sensor testing on the drone, it flies with an aero sensor, a Garmin, one of two distance measurement devices, ant+ simulator, leg extensions to get it will above the ground. A real Turkish bazar flying around. Not sure if the Skydio could carry all that baggage. But Father's day is soon so maybe I need to speak to the kids.

How close are the values of the important parameters measured near a drone to these values at the drag center of the bicycle and the rider?

The idea isn't to use the drone to test aerodynamics, but you can use the drone to test sensors that will in turn be used to measure aerodynamics on a cyclist. So you can do some pretty cool component unit testing.

For example, altitude. Some vendors use barometers as the primary sensor for altitude measurement. Barometers are very subject to influences such as wind, wind at yaw. Accurate elevation data is KEY to many of the things we want aerometers to do....A drone is a perfect device to observe how these sensors behave and more importantly how to complement/correct them. Hover at 1m above ground, rotate 360 degrees around the Z axis and get hit by wind at various angles and you will see how the barometer behaves. This is one very simple example. Well, at least it sounds simple :-) You end up needing sensors to test the sensors

There are many things you can test better on a drone than the road. Altitude and air speed sensors, which are the two most sensitive components for aero testing work quite well on the drone.

These legs are too old for 2x per day testing and my wife said no to the wind tunnel so we settled on a drone.

Ok, now I got it.

Measuring is somehow my job, but even measuring the simple parameter temperature can become complicated. In German we say „Wer viel mist mist Mist“ (who measures a lot measures a lot of rubbish).

Hilarious :)

It’s cool that you’re intrigued. I imagine tuft selection and locations will be a something of a study on its own.

Without having tried it, I would guess that a somewhat rigid and thin wire, like high-vis fishing line (maybe bent through a tape “platform” such that they tend to protrude normal to the body) may work best, since flow reversal or local shear may be too weak to activate tufts with any tendency to adhere to jersey material. Then, the direction these things
take on will be less likely to be corrupted by gravity or fabric properties, more correlated with local flow, and hopefully thin enough to not affect aerodynamics. When I think about this, I wouldn’t be surprised if such flow tripping late down in the lower back would be beneficial (good flow-tripping devices add turbulence just before it is needed, to avoid excess losses to shear, and are sized as small as possible to do the job, unlike say, jersey rear pockets filled with stuff).

Lets ruin any patent prior art here by suggesting tension, pressure or vibration measured mechanically, visually, or acoustically of any distribution of wires or transducers can be used to predict aerodynamic performance, and their integration into an article of clothing is not novel to those in this art :)

I haven't decided if you're genius or not.

What you actually want: A drone
What you tell your wife you want: A full-on freakin' wind tunnel
What you end up getting through negotiation: The drone

Clearly, I have much to learn.

---

-
My tiny little slice of the internets: dcrainmaker.com

I'm gonna tell my wife I'm thinking of taking flying lessons. Then I'll settle for a drone.

Oh god I'm fucked. She thinks flying lessons are a great idea.

Don't panic

This is where you say "Hmmmm, maybe financially it would make more sense to get a drone"

If she agrees, get a drone
If she says, "you work hard, we can afford it ",
You say "well, maybe you're right and i should get a drone and a new bike instead". Then you look at the kids and say "it' better for the environment, I'm doing it for you".

Win win

You’ve heard of a tricked out cycling pain cave? Do the same with theater quality projector, sound system, controls, vibrating seat and all that for your airplane sim.

E

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

Back on topic…

I’ve been doing some A/B testing and A/B/C testing of long sleeve cycling jerseys and I’ve found one interesting result.

Caveats, I test with Notio. The raw CdA is slightly inconsistent, but the delta is always the same. I’ve averaged, thrown out highs and lows, thrown out only the outliers, etc. and they all produce the same delta. I’ve got dozens of days of testing over a couple three months.

I usually do six 5min runs of each, so 12 total runs. Within each run I snip off the few seconds where I’m getting up to speed and the few seconds at the end when I’m stopping.

I have found a random “aero” jersey that is consistently .050 CdA faster than a totally “smooth” jersey, the kind clubs would use because the fabric is good for graphics and sublimating.

Anybody ever hear of these types of fabric numbers or see holes in my method?

E

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

I'm guessing that's .005 ? ie .210 vs .215

No, 50w, .050. The phenomena I’ve quantified, if right, is the difference between aero fabric and “smooth” or “flat” fabric. Basically absolute best case vs. absolute worst case.

Makes me want to know where aero fabric comes from. Latest TT suits have different fabric types fit every panel it seems, even sleeves are different upper and lower arms and front and back.

E

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

That is a LOT. I'm a little sceptical but who knows.
There are ways to double and triple check underlying measures but they aren't straightforward to explain.
Are you seeing B .05 better than A then retest baseline to find C = A ?

Edit :
Let me rephrase that. If the state of the art of suits has leaped forward that much in 2 years, OK. Maybe the regular tunnel testers can comment.
2 years ago, I was seeing 12ish watts between a regular (but tight jersey) and a Castelli skin suit.
There are several things that should be investigated if you suspect a large error and there are ways to investigate them. 50 watts is a lot.

Yep, agreed it sounds like a huge difference or it would mean comparing wind jacket half open VS fast TT suit ???

In my experience, between the worse tri-suit on me and the best tri-suit I've found for me the difference is about 0.015. That's already a lot IMHO !

.05 m^2 is like the size of a sheet of A5 paper.

Baseline always retests and A = C. Baseline is within .01 between days for the most part.

I should mention that both jerseys are somewhat but equally “loose”, so perhaps aero fabric is even more aero is that condition.

E

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

I guess it’s even bigger because I’m finding .050 CdA difference, so some number multiplied by the A5 paper.

E

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

Are you using the same calibration factor in the tests ?

Are you checking the calibration factor post ride ?

Most importantly, do you believe the number ?

I don’t really believe the number but yes to your other questions

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

If you share the GC Json files (Garmin+Notio merged) I can look for errors/anomalies and feedback if I see anything.

Sure, PM me your email?

E

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

tufts with a follow drone is such a cool idea! I wish I was 15 years younger, or I'd totally be doing this! haha.

Here's some really old tunnel video using a tuft jersey - sorry about lo-res, but that was all we had at the time:

https://drive.google.com/...rU8/view?usp=sharing


https://drive.google.com/...gVS/view?usp=sharing


https://drive.google.com/...5Q1/view?usp=sharing




...and regarding "feeling" what is and is not fast, I can totally remember (to this day!) this DOE set of runs and how much pressure built up in front of my shoulders during the beta=15 run with the narrow elbow width. wider elbows at that same beta=15 did not have this same sensation and the cxa showed it.


https://drive.google.com/...kjp/view?usp=sharing


that was a really cool set of runs to explore any interactions that might have been going on!