Aero Sensors Stalling?

Late last year during Eurobike and afterwards, there was a lot of buzz from various companies on aero sensors. Notio Konect, Velocomp, Aerolab, and Swissside have since been relatively dormant with no major product announcements, and the conversation has really become muted. In DC Rainmaker’s keynote at the Connect IQ summit, he highlighted that too many companies “came out” at too early a point in their respective developments. Visibility was low and the execution of the technology was fumbling. He even went out to suggest that some of the companies mentioned would be going out of business, in his opinion.
Do you agree?

Why do you think this technology has not created more buzz and interest? What do you think is the missing link? Is it poor user interface, too high price, unattractive product design (Pitot tube)?

No company has implemented it well at a good price point.
See also Boeing 737 MAX

As one of only a handful of users of the Aerolab sensor, I can provide a little insight.

I believe the largest issue is coming up with a combination of both hardware and software. You might see something by 4th quarter 2019, but more likely 1st or 2nd quarter 2020.

In the end, I think only a few viable sensors will make it to market. Most will run out of money before ever getting there, which is common for any new product. Viable to me is proven, consistent data with yaw (why would you develop, or spend money on, a sensor that doesn’t provide yaw data?). Aerolab appears to be ahead of everyone, but you have to think Garmin will have something soon, too. Notio has stumbled, likely rushing a little too quickly to be first to market. Software is certainly a problem they need to solve. Still, I’m guessing they pick themselves up and move forward, though without yaw I see them falling away very quickly once Aerolab and Garmin have product available to the public. Just my opinion.

Then there’s education on how to properly use these sensors. That’s a whole can of worms, but nothing the power meter didn’t need to overcome.

So, they’re coming, just not as quickly as the developers thought. What person/company doesn’t underestimate how long it will take to develop a new product? Overcoming unforeseen challenges is what separates the contenders from the pretenders. I was the first to deploy the Alphamantis system back in 2012, so have been through this before.

I have been involved with the Notio product but have also seen and tried some of the other products out there. I have seen dozens and dozens of tests and customer reports so have an opinion of where we were, were we are at and where we are going.

Notio has shipped north of 100 units. As a new product in a new market segment they had to overcome challenges that are part of the game. Did they underestimate some of the challenges? I think so.

I would say the challenges fall into 4 categories. a) Accuracy and precision of the technology, b) manufacturability and reliability of the hardware, c) ease of use and customer experience, d) sales and support. I could probably add more but those are the main ones.

There are probably parallels to be drawn to the Powermeter market. I think it’s a tougher nut to crack

On the topic of manufacturability and reliability of the hardware: i’s not until you get many units out there and in a huge variety of operating conditions that you start seeing some of the challenges. This is probably where companies like Garmin will have an advantage. They have been through this dozens of times. I know Notio posted a FB pic of some testing they did for heat/cold/vibration, shock…Drawing the parallel to PMs, it took several iterations for the Quarq, Garmins…of this world to solve all waterproofing batteries doors, seals, cracking…and these are companies with a ton more manufacturing experience. Ship 1000 units overnight and have fun dealing with these issues. Especially a small company.

On the topic of accuracy and precision. First of all these devices are not only at the mercy of their own precision, but at the mercy of all the other devices they use. How any times did we see inconsistent CDA numbers due to PMs reading high (or low). Even worse are the PMs that drift and vary during rides. But the aero sensor hardware itself has tough problems to solve. I actually don’t think wind speed is the tough nut to crack. And having tested with and without yaw measurement, I don’t believe this is where the biggest challenges are. Time will tell. If I can control test conditions, I can get really really good accuracy with the Chung method. With these sensors I can get equally good numbers in less constricted test conditions. Are we at point we get very accurate and precise numbers in all conditions ? This is where the manufacturers need to get. I think with time this will be a major differentiator for the various products.

Ease of use is big. I have seen so many people struggle to get things configured, paired, calibrated, upload, downloaded and analyzed. While I haven’t used some of the newer incarnations of these tools, I know some are investing heavily here. Customer success will depend as much on this as accuracy and precision.

Sales and support. I have heard many opinions of how to sell and support these things. Do you go through aero coaches, do you sell direct to consumers? I see the pros and cons of each. But I would say support is a big challenge and the organizations need to ramp up, which I know some are doing.

So there are some thoughts on why bringing these things to market is not as quick as some would have hoped. I think the fact this has gone gradually may end up being a good thing after all. Not for one product, for the market as a whole.

There is one thing I am convinced of, accurate and precise devices are very possible. Will they be useable by the masses ? Like the PM, it will take a few incarnations. I dont even think the PMs are there yet.

I have no connection to any product. My only connection is that I have an interest in how well these products will do.

I suspect that one of the manufacturing problems is consistent production and calibration for each device.

I suspect that to be used in the way that consumers will want, both the hardware and software will have to be refined.

I know that if you want to get good results, field testing is complicated. These devices will make field testing simpler and possibly faster but it’ll still be moderately complicated, and the rider or the rider’s coach or consultant will have to learn what to do to get good results. For the foreseeable future (and perhaps forever), you can’t just attach a sensor to the handlebars, ride around, and get high accuracy and precision.

Best of luck to anyone and everyone who’s entered this market. It’s a tough nut to crack. When I first came up with this approach I thought there’d be maybe a couple dozen people in the world who’d be curious enough or obsessive enough to do it, so that’s an indication of how well I understood this corner of the market.

Aerolab seems to be going in a lot of different directions, and perhaps putting more emphasis on military / surveillance drone applications, which may rub some people the wrong way philosophically. I would have expected them to master the cycling science piece first, get a steady stream of revenues, solidify a proven track record, maintain a recognized customer base, and then re-invest in other applications. My gut tells me that the bike industry hasn’t bought into the Aerolab concept (or Velocomp, Notio, SwissSide…for that matter) and is waiting for someone to come up with the right formula of value proposition, price, user interface, technical robustness, and product design. Spitting out a “CdA” number on a head unit and then having to do some complicated “post ride” analysis is clearly not the right approach. The fact that it has been over 2 years since Garmin bought Alphamantis, it tells me that they are also facing some issues or challenges in coming up with a compelling product / solution that will generate reasonable revenues.

It’s cost. You’ve got folks as a target audience who are interested in the aero, but can’t afford tunnel time.

Then you turn around and want to charge $300+ for a device requiring a $400+ power meter AND a new head unit because they only work with a Garmin. So another $200+.

For me, needing another head unit is a deal breaker. I am NOT about to fiddle with multiple interfaces because they only want them to work on the Garmin.

I thought the Velocomp acted as both meter and aero but not direct measure meter. Calculated. So there you’re using a calculated power to generate a calculated aero figure. At least with the real meter you’re eliminating one of those. Plus, the combo unit is more expensive.

IMHO, rules would probably forbid it in racing, but an ideal scenario would have this built into a Garmin or Wahoo head unit. Just have a removeable pitot tube slot on the front of it. Then add $150 to the price.

That would be slick and cost effective both.

I h
Ease of use is big…

:slight_smile:

I´m happy with my results in the last 18 months using Notio (as customer, and “auto beta tester”), reliable wehn comparing with Bike Valley and Boardman Performance Centre.

Next generation; Reliable and instant data without uploading and playing with Golden Cheetah.

Meanwhile, I continue playing with the system and offering only to clients which I´m 100% confident.

So far I’ve yet to see what exactly I’m gaining (vs. a very small amount of time by eliminating a step) vs. using the Chung Method / Aerolab. I see most of the same limitations, and most people agree the “instant” cda on your head unit is pretty worthless…

I tried the (early) Notio on an outdoor velodrome (helped a lot to have Marc right there doing the hard work) and then went to A2 the next day, got pretty good results (imo) definitely close enough to be helpful testing equipment. However again… I’m not sure how much better it was than just recording it on and running it through aerolab.

So far I’ve yet to see what exactly I’m gaining (vs. a very small amount of time by eliminating a step) vs. using the Chung Method / Aerolab. I see most of the same limitations, and most people agree the “instant” cda on your head unit is pretty worthless…

I tried the (early) Notio on an outdoor velodrome (helped a lot to have Marc right there doing the hard work) and then went to A2 the next day, got pretty good results (imo) definitely close enough to be helpful testing equipment. However again… I’m not sure how much better it was than just recording it on and running it through aerolab.

I am not just talking about one product here. I have a Notio, but I have used and currently use other technologies.

To me the biggest advantage is being able to test on any ride, any weather conditions, not being constrained to an out and back, not being constrained to not hitting brakes, not having to worry on a windy day with gusts and especially being able to test during a race. I do a 15km TT every 2nd week and it turns out to be the perfect scenario to test. But it;s a course with some hard braking in some hilly conditions and pretty varying wind. It gives me great results with an aero sensor. But I also know most of the do and don’t dos.

I agree “instant” cda is not there yet, but lap cda is. I program Strava segments and get CDAs for those “reference” laps. I just get home, upload and I have a CDA for those segments.

The technology has some room for improvement and when a couple of issues are improved it opens the door to other metrics that will further improve the whole experience. But there is value today for a certain segment of users.

Fair points, and admittedly most everything I test now is already on a 250m track, so the lap cda function isn’t really helping me particularly, which I understand is a corner case and not really what Notio/others are targeting audience wise. Still, even on the road, I’m not sure that the inconvenience of having to do loops would justify the cost of the current gen products.

There’s also a lot of talk about how close the sensors are mounted to the handlebars having too much noise to give reliable data vs. changes to hand position/cockpit nuances. That is worrisome as that’s the primary thing most people would try to test after helmets / skinsuits I would think. *In all fairness that’s just hearsay, I don’t know that for fact.

I can see that instant CdA readings wouldn’t be very useful. There is probably too much noise and you need a lot more data.

I do think one of the biggest advantages would be getting something like a calculated CdA for a “lap”. I did an entire day of testing and found out that I had a sensor problem only after I got home and loaded things into Aerolab. About 5 hours in the car, track rental and a day of vacation time down the tubes. Smart people (i.e., not me) would drag a laptop along, download the data and run a couple of Aerolab calculations at the track, but that would probably cut in half the number of runs you could finish in a session.

,… justify the cost of the current gen products. …

Given what you and i have spent on tunnels and time testing, I think we are both disqualified from judging reasonable expenses for AG athletes

I am doubly disqualified since i have a ceramic speed pulley.

,… justify the cost of the current gen products. …

Given what you and i have spent on tunnels and time testing, I think we are both disqualified from judging reasonable expenses for AG athletes

I am doubly disqualified since i have a ceramic speed pulley.

Hahahaha! So true!

I do think one of the biggest advantages would be getting something like a calculated CdA for a “lap”.

Been using the CdACrr app and the Weatherflow anemometer with good results; 12 test sessions and counting. It has a few different test modes. I’ve been using “defined lap length” on an out back course that has a steep hill for the turn. CdA SDs ~.003 in a variety of windy conditions.

The Weatherflow meter costs ~$70. You need the Bluetooth/smartphone model, plus a mount. The app is a couple bucks and runs on Android phones only. I’d be happy to help anyone get started if they are interested in trying it.
http://cdacrr.blogspot.com/

https://image.winudf.com/v2/image1/Y29tLmNkYWNycl9zY3JlZW5fMF8xNTQ5NzI1MDE1XzAxMA/screen-0.jpg?h=355&fakeurl=1&type=.jpg

https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTAObs0EWVjgOnOWLc9MVFusO2apU4l4jP95ZrJn-RA9huKCXXd

I remember seeing that at least a year or two ago and thinking “Geez I wish I had an Android phone”
.

I’m using a Samsung S4 and I think that’s what Pierre used when he developed the app. They are around $50 on ebay. Even the S6 is ~$75. I don’t think you need to set it up with a phone plan, but I’m not positive.

What do you think is the missing link? Is it poor user interface, too high price, unattractive product design (Pitot tube)?

Reality.

Some companies apparently thought “real time CdA” and a small unobtrusive “tucked in” sensor was the way to go. While those features are nice, the “price” you pay for them is useless (imprecise, inaccurate) data.

For it to be useful for field testing it needs to be combined with a test protocol and software. And it needs to be located far from the airstream influence of the rider and bike. So mounted either way out in front or off to the side. This is not something you leave on your bike.

User friendly and universal testing software is the key. You are trying to conduct a controlled experiment, with a lot of variables and ways to screw it up. If you want to make it appeal to the masses the software needs to be easy and near idiot-proof, and able to handle a variety of testing scenarios, and hopefully provide AI-like error checking.

Pierre’s (bugno) CdACrr app is a good start, and works well with a $70 airspeed meter, so money isn’t an issue. He has been updating it lately, so it will keep improving. I know of a couple people developing more sophisticated sensors with yaw (and hopefully software too), but they will be in the $1000 range. I’ve gotten good day to day repeatability even with strong winds from a variety of directions, so measuring yaw may not be that important.

Would love to try and see the kind of results I can achieve with cda crr app even though it means buying a weather meter and an android phone.

I’ve been testing with Aerotune for a little while (https://www.aerotune.com/) and it’s super easy to use with a Garmin Computer and phone in your pocket and the feedback given runs after runs on the Garmin is super great to have ! Would love to use both apps on the same day and see how they compare !

The Notio Konect is available for $950 right now. I’d love to get one but it’s currently a want and not a need.