CliffordK

Whether or not the Arofly uses tire pressure for power calcs, it would make a nice sensor... However, I can usually feel the wheel bonking when it gets a bit low. Too much practice with sewups

A pressure sensor + slime/sealant sounds like a gooey mess

Homebrew01

April fools day ?

MinnMan

This is a topic for a different thread, but I'm not so sure Strava is spot on. I have seen some funny things on Strava segments, such as indicating the my average speed on a segment was greater than my maximum speed on the same segment. I had a KOM a while back (no longer mine) in which that was the case, and I'm fairly sure Strava just didn't have my time right. Also, I upload my rides to both Strava and RideWithGPS, and they clearly sample the raw data differently and give somewhat different results over short intervals. I'm not saying that RideWithGPS is completely accurate either, but just yesterday Strava thought I hit 30 MPH on a flat where I know for certain I wasn't above 22 or 23 MPH, whereas RideWithGPS has it about right.

PepeM

I managed to hit 30,000+ watts during a race according to my Powertap.

WhyFi

I love threads that bring out both the 41 engineers and the 41 "engineers."

rpenmanparker 

Doesn't cost anything to have an open mind. I didn't say I was going to buy one.

StanSeven

Right. But perhaps they submitted a PCT application which does cover many nations. Also pendacy is high iand approaches three years in the US.

CliffordK

What device are you using? Garmin - something?

As I mentioned earlier, some of the Garmins seem to sample every 5 seconds or so. Since the segment is calculated based on the time on the segment, if Strava cuts it short at both ends, then one can be off by quite a bit, leading to a seemingly fast time (or on occasion slow time).

Look at the granularity of the power estimate to get an estimate of the frequency of sample intervals.

MinnMan

Yes, a Garmin Edge. I just scanned a few raw .gpx files and in most cases there are data at once per second, but indeed there are other parts of the files with only one datum every 5 seconds or so. I would hope that the low sampling rate corresponds to slow changes in position (low speed), but perhaps it's not that systematic. Are there devices that sample reliably with greater frequency?

CliffordK

I suppose a bit off topic... but I'm still waiting for my new really expensive valve cap to come... so no news for a few weeks.

In the mean time, here is a comparison of various GPS devices.

GPS Distance Accuracy Test: Smartphone Apps vs. Dedicated GPS | Singletracks Mountain Bike News
GPS Elevation Accuracy Test: Smartphone Apps vs. Dedicated GPS | Singletracks Mountain Bike News

It was a tough challenge. The Garmin GPSMap60CSx had a lot of problems with the corners due to a low sampling rate.
The Garmin Fenix2 looks interesting as it apparently had more sampling on the corners than the straight stretches.

Perhaps that is why the Garmin devices have troubles with Strava segments. Some may use a uniformly low sampling rate. Others may vary the sampling rate which may not correspond with Strava segments.

The Android and IPhone devices seem to do reasonably well with Strava, although perhaps not perfect tracking.

Seattle Forrest

Most (maybe all?) Garmins have two choices for sample rate:

• Every second
• Smart

Sounds like you have smart recording set. You can change it if you like. You should change it if you use a power meter, but it doesn't matter otherwise.

Your device spends the same amount of time talking to the satellites no matter which option you use.

Smart recording means the device records a point to the file when it needs to, to show a change in speed or direction. Every second means the device writes to the file every second regardless of what the data says.

Seattle Forrest

Wow, talk about ancient! Garmin is about to release the Fenix 5 this spring.

Nope. That's a fundamental misunderstanding of how Garmins work. See the post above this one.

Seattle Forrest

Because somebody started a thread now, and talking about stuff is what we do here.

merlinextraligh

Correct, but a little bit pedantic. Strain guage power meters are measuring the force being applied to the drive train (at the pedals, the crank or the hub.) Admittedly, there are still calculations involved to derive power.


But the point is that strain guage based power meters are much closer to making a direct measure of power than a device like the ibike, which is extrapolating power from a bunch of measurements, and assumptions about aerodynamics, without actually measuring the force being applied by the rider.


Given that the Ibike, after years of work on their algorithm is still rather questionable in both ease of set up, and accuracy, I can't imagine this new device will begin to approach the accuracy or useability of a decent strain gauged based power meter.

FlashBazbo

Yes. Your Garmin will. It has settings for "smart" sampling (which I suspect yours is set to) and for "once per second." The default is "smart" sampling. You have to change it to get once-per-second sampling.

If people are really trying to do live Strava segments on a Garmin set to "smart" sampling, I'm not surprised that they are getting it wrong. That's like timing a 100-meter dash using a calendar.

MinnMan

This leads to the dumb question about how these newer devices perform when one's position on the bike is not constant? If the devices require some assumption about aerodynamics, how do they adapt when one changes position from sitting up down to the drops? It seems that they don't?

rpenmanparker 

Why do folks keep thinking the device approximates power from road and wind speed, rider weight, aerodynamics, etc. I don't understand the basis for the measurement, but a careful reading suggests it has something to do with pressure effects inside the tire due to power applied to the wheels. Maybe road speed is also used, but all this talk about trying to figure out slope of the road, direction of the wind, speed of the wind, etc., is just nonsense.

Seattle Forrest

Does the pressure in your tire change when you ride over a bump or a crack in the pavement?

CliffordK

Hmmm...

When I placed the order, it said approximately 2 weeks.

This morning I got this message.


I suppose it isn't surprising. But, nothing on the web page indicated it was a "preorder".

I'll have to see if they actually charged the credit card, but 2 or 3 months, and it makes it much more difficult to recover payments.

wphamilton

Some of us already agree with you.

wphamilton

Uh-oh. If they charged already I'd start a charge-back.

FlashBazbo

Sadly, I think this fits with everything else we really know about this "product." I still suspect the product doesn't exist in a working format.

They wouldn't be the first where the marketing department is well out ahead of the engineering department. (That's a nice way of saying that they are selling something they do not have.) Multi-million dollar alternative energy projects have been sold on this basis.

MinnMan

As already noted earlier in the thread, high frequency noise such as potholes can be removed by Fourier transforms.

Which is not to say that I'm sure that the gizmo actually works as advertised.

FlashBazbo

What? Are you trying to look at this SCIENTIFICALLY? You're trying to introduce facts and/or the laws of physics! (And also pointing out, rather obliquely, that this thread should have been about two posts long.) No fair!

This is the old "Air a car tire up to 36 psi while it's off the car. Now mount it to the car and let the car down. How much does the pressure inside the tire change?" deal.

If the "product" isn't just a practical joke, it's a promising scam.

Doge

Off topic a bit.

Some case materials expand more than others with pressure. Nylon vs cotton for example. Which means that in a flexible case tire the tire is expanding a really tiny tiny bit as the tire is deformed and pressure goes up. In a cotton case the tire expands less and the pressure should go up more. In every case I know of tires are less elastic than any gas. Tire expanding and contracting wastes energy.

As we generally all use air for inflating tires the part about which gas may be better is not a factor. But I have spent some time on that and turns out pretty hard to get information about that. Might be a nice diversion sometime. Nitrogen, CO2, Air and Helium are/have been used for inflating tires. But when pumped up to 100psi I'd think they'd tend to behave more the same than different. Or...100PSI of Helium takes more moles than 100PSI of Nitrogen.

OK - I have no idea how that PM is really working.