what is it about Canadian companies and aero sensors ?

Aerolab, Notio, Alphamantis and now this.....

Thanks for the link, but that website doesn't tell us jack !

I like this ... ' Gibli is working toward developing a real–time aerodynamic sensor '.

So, should we just give them our CC number now ?

db

Up front, big and bold... "Real time aerodynamic sensor".

Nope, won't be useful for field testing... probably.

Do any of them actually make a product I can buy and use?
Seems like a lot of vaporware.

Notio definitely makes and sells a product.
You need to determine if it meets your needs.

I said this to you just a couple days ago!

Best of luck to anyone who enters this field.

Yep and as soon as I saw where they were from I said to myself....wow....the man is a clairvoyant.


And contrary to what rruff believes, I believe real time is a possibility

Let's
Notio is one of the more reputable (Wattshop and MTM are using them and getting results) and it is on sale for 40% off. I got one and have been trying to figure it out. Let's just say that it is probably like the early days of power meters. You can look at the numbers, but the utility comes from really understanding how to use it properly.

From the CAD model on their website, (and I believe notio and some of the others are the same way) the only air measuring device is a pitot tube. This is only useful at getting bike-axis relative wind velocity. Since we know CdA changes with yaw I would think some sort of directional measurement would be important to get the CdA calculation right. Am I missing something there? Maybe put the pitot tube on a bearing that allows it to spin, with a little tail (like a weather vane) to make sure it's always lined up with the relative wind direction and you could put an encoder into that pivot joint to measure yaw angle.

---

-------------
Ed O'Malley
www.VeloVetta.com
Founder of VeloVetta Cycling Shoes
Instagram • Facebook

That is the design of the Skippy Kitten/Red is Faster sensor. The other alternative is to go really fast so the yaw angle is low.

In my initial fiddling with the Notio I am guessing that getting it further out front would be a good idea,

Good software and a test venue and protocol that leverages VE.

As a delta to compare moment to moment “better, worse, better” .....maybe, as an absolute..”. Your CdA is actually “ .210, .215, .210”.....hell no.

Yup. I have tried doing the outdoor Chung/VE method a fair amount and the combination of little to no traffic and calm conditions makes it really difficult where I live. While the Notio does a lot of stuff, if the wind sensor makes outdoor Chung testing possible it will be worth the cost. Just the fact that it has a built in air pressure sensor for automatically measuring the altitude corrected tho value goes a long way to making it worthwhile. A Weather station to do that costs half as much as the Notio.

Hey Mike- How easy would it be to send me a test file that contains second by second data with (i) speed or distance (from a speed sensor not GPS) (ii) airspeed (from Notio) and (iii) power (from a power meter), and (iv) ideally two laps of data from the same course run once in each direction but almost as good would be a lap on a loop course that starts and ends in the exact same spot (no need to split out the laps as long as the file contains the lap info). If it's easy to create files like this then I think Notio could be useful, though that would depend on the quality of the airspeed data. I'd love to see a file just to run some numbers on it and see if the data looks to be good quality. Also happy to help you analyze your data in return.

Yep. Good software, good protocol, good and well calibrated auxiliary power meters, good internal sensors + algorithms to properly and precisely track the balance of energy expenditure and input energy. And yeah, that whole protocol that enables pulling apart CdA and Crr reliably (even in variable wind conditions).
From there, if the device does what it is designed to do, then the question is how to efficiently adjust and improve (aerodynamics and rolling resistance) without sacrificing your power production (among other things like comfort and stability).

---

Chris Morton, PhD
Associate Professor, Mechanical Engineering
co-Founder and inventor of AeroLab Tech
For updates see Instagram

Altitude corrected tho value? Air density? The pitot measures dynamic pressure directly, so air density is not needed.

If you have passing cars then that's going to mess up your data. Based on what I've seen with similar sensors they won't be able to accurately measure what your bike/body sees when that happens. So I think we are stuck with that restriction.

Dealing with wind is what we hope to get from an aero sensor. Plus software that uses good test protocols and lets you analyze data in the field. Real time data will inherently add extra error vs a good protocol. At the current state of the art (AFAIK), it's not worth trying to use real time data for field testing.

The best protocol IMO is a out-back where you ride close to the centerline and can mirror the runs in both directions. This lets you check for wind bias; a small error in that causes a big error in your CdA result. Also, after a few runs you can nail down the actual elevation instead of treating it as an unknown, which is another nice feature for error checking.

Last year I used the CdACrr app and anemometer, and I think it works better than anything else I've seen. Cheap too. I was hoping to motivate Pierre to code an outback mode... but with 2020 a bust and this being my last year racing, I lack incentive...

The testing I did last year with anemometer and CdACrr app didn't show any effect of yaw, within the measurement uncertainty at least. I always did out-back and normalized the data to eliminate wind bias... zero net tailwind or headwind.

Fairly certain you cannot eliminate wind bias with a static measurement from an anemometer unless you assume steady flow (not the case) and assume the airflow does not change with terrain and surroundings (e.g. hills, buildings, trees, etc).
Perhaps your equipment and body position are not significantly affected by wind yaw (holistically speaking, not component level), at least to the level of the measurement uncertainty as you have indicated.
We have certainly not observed what you are indicating. We have found substantial effects of wind yaw on CdA of the rider, particularly for yaw greater than 5 degrees. That said, we test a lot in Calgary where typical wind speeds at the rider elevation are >10kph.

---

Chris Morton, PhD
Associate Professor, Mechanical Engineering
co-Founder and inventor of AeroLab Tech
For updates see Instagram

Well good question. I have looked at the flat file with the data, but haven’t tried exporting it yet. I will say that one of the things I learned with my fiddling today was that I need to be more careful with the sensor install. I failed on getting it level Enough and on the central axis Of the bike. As a result I think my wind speed data is off a bit.

As of now, I don’t have a good out and back that doesn’t require touching the brakes. I won’t ride tomorrow, but on Friday I may head over to an oval at the local college campus. I have successfully Chung tested there, once (all other times I just had too much breeze to get a really nice VE profile). I will send some data once I am sort of confident that I didn’t screw anything up.

There are things you can do to put bounds on the error but, in a strict sense, yeah, you need a bit more info. That said, as one of my mentors once said to me years ago, yes, you want a good estimate; but if you can't get a good estimate the next best thing is to know that the estimate isn't good -- and by roughly how much. The worst thing is if you have a bad estimate but think it's good.

I think it's a possible now in certain situations, and it's a future possibility in others. I think there are other situations where it will be very, very difficult.

I disagree.

I am not a big fan of looking down at my head unit and trying to read the immediate CDA if that's what "real time CDA" means to you.

However the ability plot my CDA over time and pick any point and it tells me my CDA for the that 30 second window, and it's within a reasonable tolerance, let's say .005 and i can see influences on that number, say Yaw.....that is achievable IMO

There are easier ways to get yaw. A few holes and pressure sensors can get various wind direction info.
I was trying to see in the Gibli photo if we can see how many holes on the pitot. Not sure if they do yaw.

"if you can't get a good estimate the next best thing is to know that the estimate isn't good". This is gold.

One of the first things I do is go into Aerolab and see where my VE and actual elevation diverge. I immediately see the question marks. My modified version of aerolab allows me to overlay all other info such as power and speed ( to see drops), wind, yaw, and a whole whack of measures I don't want to pollute this thread with. I can pretty quickly tell what went wrong.

Example....when I'm testing and a pick truck drives by, I stop pedalling for 5 seconds, just to get an indicator in my file. When I am looking at my VE chart I see a "blip". I then click the cadence checkbox, overlay that data and sure enough I see my cadence went to 0, my indication a truck went by. I then toggle on wind and I see a big disturbance in wind/air pressure. I snap to the point to re-attach my VE/Altitude lines and move on.