Even the downhill bit has enough flat and a dip that it's not a zero power segment. however, there is a good bit of elevation loss from the crosswalk to the back of the roundabout that would require hard braking from speeds 24-27mph.

---

My YouTubes

rruff wrote:

If it's really a 3% grade, I think that is too steep. Mine is ~1.2% and I wish I could find a flatter one. Particularly if you have a tailwind on the uphill, then you will need to go really hard to outrun the wind, plus all of your aero-drag-energy for the lap will occur on the downhill part, which isn't ideal. If you could find a flattish road with a hill at one end for turning, then that would work well.

Due to the topography in my area a tailwind on the uphill segment is as likely as me putting out 300 watts over 40K. can I take it that this is a good thing?

---

My YouTubes

You can do just the uphill leg, but make sure you do them at a wide range of power/speed (like: I do one run basically as fast as I can, then one run at about half that power, then one in-between). I discuss this on (I think) slide 98 of my white paper.

In early tests I used to forget to hit the lap button at the end of each lap. I hit on duct taping an old car mat to the road at the end of my run; when I went over the mat at the end I could see a "bump" or discontinuity in the VE.

I'm guessing the brick bits should/might simulate the floor mat?

Also, thought some more about mounting the anemometer....thoughts on this position (red K-edge mount)? I can go either up (pictured) or down. Thinking either are going to be pretty close to center mass. Just worried a bit about being too close to my hands and that it affects my grip ever so slightly.

---

My YouTubes

It's a very good idea to calibrate windspeed when you ride, which means going out and back both. Another issue with only going uphill is that your aero-drag-energy is going to be quite small. Since that is the thing you are wanting to measure, any errors in other metrics are going to have a larger effect on your computed CdA. Re# (airspeed) will also have an affect on CdA. A flattish road with most of your time ~race speed is best IMO.

There various ways you can mark laps using the data... but do you really want to graph the results, then go into the file and hand-manipulate the data for each lap? I guess some people's tedium threshold is substantially higher than mine...;)

I think the anemometer that close to your head and hands is definitely not optimal. Try the basebar handle. You'll need a different clamp anyway... standard camera mount threading I think.

When you are talking about "custom segments" is that CdACrr or MyCda?

Hey trailerhouse- I just tried to give your app a test run and it wouldn't load any of my fit files. They are from a Wahoo Bolt. Guessing you probably already know about this problem but I couldn't test it out as a result.

Yeah, that's what I was thinking is that it was too close. Really I went out to the garage to see if the GoPro adapter would fit where the Garmin insert goes and then thought I would see how useable that k-edge might be once I had my hands on it.

I can do tedious, but if I don't have to that would be best. 😉

I really would like this route to be my test course but you're probably right about race speed, which I definitely wouldn't be doing on just the up bit.

Probably too many things going on in my head and I conflated the two, but the premium version does have automatic segment in cdacrr. Could this be set to do the same or similar?

---

My YouTubes

Can you send me a file so I can take a look at it?

Email support@mycda.app
Thanks!!

---

What's your CdA?

Auto segment mode in CdACrr is for on the fly CdA (or JRA as I like to call it). You don't want to use that to test because it uses the barometric sensor for elevation and isn't very accurate.

You might be able to use that course. I'm going to renege on my statement about the grade being a problem; it might be ok if you are measuring airspeed. Before I got the airspeed sensor I would manually determine a wind vector for each lap, and for that you don't want all the aero-drag-energy to be on one leg. But you don't need to worry about that. So I think you should give it a shot. I'd advise using the fixed lap length like I do, so your laps will have one turn that you are able to negotiate without braking. If you are able to have your start/finish on a flattish section that helps reduce the error from having that point be off a little.

I've said this before here and on other threads so I'll repeat this one last time and then drop it: if all your tests are done at one identical speed then all your data will be identical.

I guess me and Lanier had the same idea...;)

Is the "wind model" functional? When I click on that I get a spinning icon that never stops. That's uploading a CdACrr .fit file with airspeed data.

I wish that was true... ;)

I'll take a look. The wind model is supposed to approximate air speed based on weather station data and your direction of travel / heading. It depends on a lot of things, accurate weather reports, GPS data, mapping data etc. Sort of what best bike split does.
If you have airspeed in the file then you don't need this.

Anyway, I'll check it out in case something broke. It uses a few 3rd party web services so maybe something changed since I developed it.
---
Fixed it, should work. You'll need to do a browser refresh and log back in.

---

What's your CdA?

Thanks, I was mistaken, I thought it was airspeed. Cool app! Is there a way to say pick a point and generate a "lap" that is equal distances before and after that point?

I also have a suggestion. Since CdA is determined visually, it would be good if peaks and valleys in the elevation plot were detected, and an elevation line (horizontal line) drawn at these locations.

How do you do this ?

Distance averaged headwind/tailwind for all runs =0.

You calculate some kind of factor and enter it into the software ?

ah..okay. well, then let's hope snipping is added to the functionality of the app as that would make life so much easier and less tedious. ;)

I have a 1b course that is marginally closer, is lollipop in nature, and is pretty darn flat for the important bit. However, weekend AM traffic might be an issue at that site as it is the local highschool and the ball fields are in use on the weekends.

Anemometer, mounts, speed sensor, and other bits should be arriving tomorrow at some point. Therefore, I should be able to get out and play on Sunday morning at a bare minimum. With an adapter I was able to secure the K-edge to the bullhorn well enough that it should provide for a good mount. However, without an extension I will only have the height of the device plus the threaded adaptor. I think this may be enough since I am not very tall on the bike. I still have to sort out where I will place the phone so that it allows me to easily start and stop the runs, but I am not sure how I can do that without being in an upright position when hitting the run button. Will it matter if I am sitting up to start and stop or is this something that will be apparent and I can merely ignore? Again, would be nice if we could snip those bits. ;)

---

My YouTubes

dumb question, most likely, but i'll ask anyways. Is there any advantage to running a vane type setup for the anemometer? perhaps adding on something like this?

---

My YouTubes

I was doing it in a spreadsheet, but Pierre is now doing it in his app. Actually he computes a scaling factor for each lap, while I was computing one using all the laps... but the average of the scaling factors for each lap is close enough. So I adjust the values I get from the app after the fact.

I was hoping at first that the scaling factor would not need adjustment, but it appears that isn't the case. I'm not sure what causes it; maybe alignment since positioning is not precise. But it seems that adjusting it each day gives better day-to-day consistency than not. It averages ~1.082 but deviates from that up to 1%, which is a 2% effect on CdA.

So you take the windspeed measured and multiple by 1.082, ie wind speed used in the calculations is more than the actual wind speed measured ?

Hopefully one day aerosticks that play well with other apps will become available.

As I mentioned earlier, if you use fixed lap length mode you won't need to snip... provided that you have at least one hill where you don't need to brake.

I thought about the weathervane deal, but don't think it's good place to put attention at this point... mainly because no one knows what to do with yaw anyway. It's going to affect Crr as well as CdA and IMO wind tunnel data is poor representation of what happens on the road. If yaw is significant it's probably gusting and pushing you around requiring steering corrections, position adjustments, etc.

Ideally you want an airspeed sensor that has a response to yaw that is about the same as you and the bike; so that the computed CdA is about the same regardless of wind conditions. IME the cheap Weatherflow meter is pretty darn close to that.

https://patents.google.com/...t/US20120221257A1/en

And that is my thought on the lap i.e I will hit the "uphill" bit complete the roundabout and then head back to the start on the "downhill" section. Now, can I sit up through the roundabout as long as I don't grab the brakes? My thought is to ease up on the effort as I approach the roundabout so that I can navigate it without brakes, but I'll be damned if I am confident enough to take it in the aerobars.

---

My YouTubes

Yes, I'm pretty sure that is the case. The scaling factor is one of the settings. According to Pierre's comments that is normal, but I'd like to try even more remote positions for the device to see if it approaches 1.000. I'm testing at 7,000 ft, so maybe that has something to do with it... like maybe the factory calibration is only good at sea level.

Notio uses a scaling factor as well doesn't it? As I recall from Martin, his was in the 1.25-1.3 range and that was mounted on the basebar handle.