bettrave

I use an Edge 530 and a Forerunner 920xt.

I have Continental GP4000 s2 700x25c

My Edge 530 is set to auto calibration (value 2134mm).

My 920xt is manually set to 2105mm

Interestly, I always get higher distance with my 920xt, even if the circumference is set to a lower value...

Anyone could explained that?

CyclingFool95 

Is this indoor or outdoor riding? If outdoor, your Edge 530 should be using GPS data for distance.

DrIsotope

If outdoor, you should be using the calibrated speed sensor as source #1 , and the GPS as data source #2 . Because consumer GPS is decent at locating a stationary target, and pretty spotty when moving at +20mph-- as it polls once a second, and you're moving ~30 feet per second, so the GPS accuracy is +/-30 feet @ 20mph. It gets worse under trees, in canyons, next to tall buildings, etc, etc.

Get a string, wrap it around the tire, lay the string out flat and measure the string, input that number. My Edge 520 was set to auto-calibrate and shorted me ~4% mileage for 2 years.

bettrave

From the manual:

About the Speed and Cadence Sensors

Cadence data from the cadence sensor is always recorded. If no speed and cadence sensors are paired with the device, GPS data is used to calculate the speed and distance.

biker128pedal

I like to get on the bike and roll at the desired tire pressure to measure circumference. Makes a slight difference. Use a metric tape measure. I tried putting line chalk on the floor and roll through it but could not get it to mark the tire. Helps in a garage so I see the string would work in doors.

example
Road Bike Aeolus R5 wheels
Rear Conti 4000 112 psi 2110mm

UniChris

In terms of rotation counter computers, something that's useful is to reference known distances.

I've now got mine dialed in to the point where I can do a 48.8 mile ride and be off the official trail distance by maybe two tenths, which is really in the range of choices I made at water stops and taking a road around a downed tree.

A *good* GPS track would have 1% error, and often they're worse.

Miele Man

When I was using a regular bicycle computer and wanted the most accurate measurement of distance I could get, I'd get on the bike with the load I intended to carry and then roll forward whilst a friend measured the distances of three completer wheel rotations. I'd divide that by three and input that distance into the computer as the circumference. I don't know it that'd work for you.

Cheers

Sy Reene

This sounds contradictory? If outdoors, the GPS I would think is more accurate than a speed sensor. Over an eg. 30-50 mile ride, if the start and end points are both off by by the referenced 30 feet, that still sounds pretty accurate.

DrIsotope

No, your ride could be +/-30ft if your activity was 3 seconds in duration. Any or all of the three polling points could be off by +/-30ft. That's why we see people get PRs/KOMs on streets they didn't even ride on-- there's just a line moving straight through buildings, because the head unit lost the satellite for a few seconds. On a mountain ride a few years ago, my Edge 520 lost the satellites, got confused, and added 22 hours to my moving time. I basically time travelled to the next day. Had to fix the time code error in FitFileTools. Ever since, it's been a speed sensor on every bike.

noimagination

There was a lady (this forum? Bicycling forum? other?) a few years ago who would argue that bike computers using wheel sensors were always inaccurate because the track of the front wheel is not the same as the rear wheel, and the front wheel is not always tracking in a straight line. So, in her view, the distance measured by bike computers overstated the distance actually traveled. (I wondered what she looked like on a bike, we've all seen those riders who weave back and forth with every pedal stroke - I figured this was probably her.)

I'm not all that concerned with accuracy, myself. I get that some riders really get into the data, and that's great, but for me whether a ride is 40 miles or 40.359 miles is immaterial. I generally measure rides by time anyway, since that is what matters for me (I have 2 hours to ride before I have to be somewhere, or I have the energy to ride 3.5 hours but not 4 hours today, etc.).

woodcraft

I know a guy who habitually rides in a serpentine line. Riding behind him on a fog line I was able to estimate that it was about 6" per pedal stroke. He rides a lot of DCs, & by measuring a piece of string laid out in that pattern I estimated that he added 2 miles to a DC (IIRC). Since I was observing the back wheel, perhaps his front wheel traveled even further.

Iride01 

This is the number on the devices and not when you view the ride activity that has been uploaded to a site or software?

While we would expect that the auto calibration number is used to correct miles that you've already ridden, I personally doubt that it does. And some devices might auto calibrate several times on a ride.

This is with the devices on the same bike and not on a trainer..... right? If on a stationary trainer, auto calibrate is not going to work and the decision process it's program uses to provide you with a number is even more a guess to us.

It'd be interesting to know what happens if you put the 530 to manual with the same setting as the 920xt. Likewise if the 920xt will auto calibrate, what's the result with both in auto calibrate?

And even so, the differences in the wheel circumference are only 1 percent.

Sy Reene

Ok got it.. but don't the various sites recalculate your distance, and therefore your speed, by actually comparing your ridden route to the mapped distance it should be? Ie. isn't this why Strava relies on GPS for all of its segment times, since it feels that a speed sensor unit can be more likely incorrect (intentionally or accidentally)?

bettrave

Hi Iride01,
I have putted 2105mm in my Edge 530.
I tried pedaling inside and I got exactly the same distance on both devices.
I did not tried the auto calibration on my 920xt.

spelger

Sounds more like a programming bug induced by gps loss rather than the gps loss itself.

spelger

She is/was right...to a certain point. the rear wheel does not follow the path of the front exactly. the front travels more. but not so much to lose sleep over.

spelger

this article may hlep: https://support.strava.com/hc/en-us/...-is-Calculated

unrelated but equally interesting is this: https://support.strava.com/hc/en-us/...-Your-Activity

Andy Somnifac 

The GPS tells you where you rode, the speed sensor tells you how fast you rode it.

Iride01 

What is unknown in relation to the OP is whether this was on a trainer or out moving in the real world.

I wouldn't expect a mileage value for riding nowhere on a trainer.

bettrave

Real world