Fitted a Wahoo speed sensor to my commuter at the weekend. Set the wheel circumference to 700 x 25, and off I went.

Now my commuting distance has changed from 14.4 miles (GPS) to 15.0 miles (speed sensor).

Bit of a leap – anyone know which is more accurate? And why?

I’d say there’s more room for error on the speed sensor, even a few mm out multiplied by the number of revs will be a bit of a leap.

What does Google maps say? I’d bet on the gps

the speed sensor if yo measure the circumference of the tyre properly and it allows you to put the real figure in. Just putting in 700×25 is only an approximation

Speed sensor will be more accurate with no tyre deflation and an accurate measurement if it is spinning in contact with the ground all the time and never slips.
GPS will be more accurate in good open ground if you keep a good straight line and don’t wiggle under the GPS sensitivity.

precision vs accuracy.

wheel sensor will be precise but if it isn’t calibrated correctly will have a gross error. if it is calibrated correctly then it will be both precise and accurate.

GPS will be less precise but doesn’t require a user calibration.

pretty sure when i had a garmin 520, it auto-calibrated the wheel-size against GPS.

Remember to subtract a bit if you actually measure the circumference of the tyre – quite a bit if you run low pressures – as the tyre flattens out on contact with the road surface.

Garmin calibrates the speed sensor using gps. Does Wahoo not have that option too? Then they’ll read the same

TT courses are calibrated very carefully using a mileometer. A 0.1% error in circumference over 100 miles makes for a difficult situation come records time! That’s 2mm by the way.

I’ll switch the calibration to automatic for the journey home and report back.

Thanks.

For most accurate wheel measurement, just mark the tyre tyre and the floor, sit on the bike kitted up, roll forward one wheel revolution, mark the floor and measure the distance between the 2 points. That’s about as good as it gets for a speed sensor i reckon.

?? a 0.1% Error is 1/1000 of the distance is 160m
(assuming 100mile = 160km)

?? a 0.1% Error is 1/1000 of the distance is 160m
(assuming 100mile = 160km)

What would 0.1% error in wheel circumference be? Around 2mm by any chance?

So, switched to automatic.
Rode home.

Have now lost 0.3 miles from my commuting distance.

The GPS distance always matched the route when I first started working there and planned my route.

Might bin the speed sensor. 😩

If you brake the 100 mile TT record by five seconds and the course is short, all hell breaks loose.

Richard Bideau – Another record-breaking 100 Mile Time Trial, but again no certificate

Can you measure the circumference to 2mm accurately?

I would go with a properly calibrated speed sensor all day long. I don’t trust any kind of auto calibration. Pop a bit of paint on the tyre, roll the bike (straight) across a hard surface then measure the distance between the 2 paint marks. (and wipe the paint up before it dries 🙂 )

Might bin the speed sensor

Don’t do that.  The problem with GPS ‘speed’ is that it isn’t very smooth from second to second as the GPS position jumps around as GPS isn’t that accurate, especially if you are moving.  Over longer distances that error percentage goes down as the error amount is pretty fixed – 2m in 2km is much better than 2m in 20m.  The speed sensor fills in for that by being quite reliable from second to second although over long distances it would be wrong if you don’t allow it to be calibrated by te gps

as said chalk (check last century) and a measuring tape will help enter accurate circumference

…if you look at the earlier split speeds some of us need all the help gps errors give……

in fairness to ridewithgps was running phone off faulty external powerpack

But there is no way you can know that you have ridden the exact same route every time. Unless you can ride on a painted line all the way, there are bound to be variations in your distance. Passing a parked car, taking a different line round a corner all add up.

Right, got the chalk and measuring tape out today. The circumference of the wheel came to 2100mm exactly.

This is ‘only’ 5mm off the setting that the wahoo had for the 700 x 25 circumference (2105) – which would suggest that the 15.0 miles it recorded with that setting is closer to the actual distance.

Went to ridewithgps site and plotted the exact route that I do in the morning – every little cut through I do or back street I take to avoid traffic or lights. It came to 14.9 miles.

Which suggests that the speed sensor matches the plotted route and the gps route is actually quite inaccurate (which if you zoom in on my strava route after a ride, is not that surprising).

Might bin the speed sensor. 😩

Might have a re-think about that…

GPS will be affected by tree cover, buildings, inaccuracy going up/down hills etc

Speed sensor over GPS, so long as its correctly calibrated.

the gps route is actually quite inaccurate

What’s the sampling interval, and how spikey is the track?

Wahoo interval is 1 second. By spikey, you mean hilly? Not very – around 600 feet of climbing over 15 miles – most of it is heading across london (which may account for GPS problems).

GPS accuracy is also affected by resolution, as is any analogue to digital conversion. Resolution here comes in two main flavours, firstly the number of satellites that have sight of you (pretty sure full resolution here is still restricted to US Military and everyone else gets at best 50%, often less) and secondly sampling frequency. Sampling frequency is how often the GPS checks for satellites and (obviously?) the longer the periods between ‘knowing’ ‘precise location’, the more time it’s just guessing where you probably are based on the last thing it ‘knew’ and your direction and speed as calculated from the pings to date.

Increasing sample frequency is a major battery hog, and most devices are trying to balance acceptable accuracy vs desirable run time. Some devices give you control here and allow you so specify sampling frequency and therefore increase accuracy at cost of run time (Suunto watches I think still do this) but most leave you in the dark about it.

When correctly calibrated and operated, the direct reading analogue approach is always more accurate, but as demonstrated further up the thread, GPS is likely more accurate than a rotation and frequency counter that hasn’t been correctly operated or configured. In a situation such as a car, where it’s far less likely you can control the speedo configuration, I’d probably trust GPS over dash but wouldn’t expect either to be ‘correct’.