Which is more accurate? GPS or Wheel magnet

[gaz]

A couple of my colleagues at work have seen my latest videos with a HUD on displaying my speed. They were questioning it's accuracy, as soon i set off my speed gets to 15mph pretty fast.
It moved on to how i record the data, and the fact that i use the garmin speed and cadence sensor, which works off a magnet on the rear wheel.

Sooo which is more accurate, a correctly set up wheel magnet with wheel size manually meassured or GPS from the garmin?

I believe it's the wheel magnet, as it should be more accurate as you know how far the wheel travels in one rotation. where as the GPS isn't going to plot your exact course.

One of their arguments was when they had a wheel based computer they said their rides varried in distance, even though it was the same route, i think this is because you take different lines in the road everyday, and this affects your distance over the whole journey.

Basically my colleagues where doubting that i could go 35mph without being in a group.

 

[thomas]

Well, I'd of thought the magnet would be better...GPS relies on a lot more to get your position.

 

[Alan Whicker]

My GPS recorded me as doing 43mph at one point on my route yesterday. I can't even go that fast on my Vespa. A wheel magnet is effectively a 'trundle wheel' as used by surveyors. If it's correctly set up it should be very accurate.

 

[Neddy]

On a similar vein, I've noticed that my regular commute has dropped from 7.55 to 7.4 miles, as measured using a wheel magnet. I have recently invested in a decent track pump and my tyres are now pumped up harder than they've ever been. Could this really account for a 2% difference?
Perhaps I should measure the wheel circumference and re-calibrate the computer, rather than accepting the pre-programmed wheel size.

 

[Rob3rt]

I vote for wheel based, not used gps, but wheel based is a constant measure of what your bike is doing, not a point to point measure like GPS.

From what I gather based on the technology there will also be some form of lag on gps updating of speed based on position, this could explain your jump to 15mph very fast?



Ive only used a front wheel magnet based computer for 1 commute (today) and it tracked my distance as 1.77 mile using the generic wheel size. Google maps said 1.7 mile. Im pretty happy with this accuracy. Error of ~4% vs Google maps. The 0.07 mile could quite easily be due to weaving about on the road etc combined with a slight innacuracy of the wheel diameter entered, or because the tyre pressure is not high enough. Or maybe even google maps is out. I will be fine tuning the wheel size entered and running a few different routes to see how they match up over the next few days.

 

[addictfreak]

Gaz

Its obviously the power generated by your awesome thighs that gets you going so quickly!

At least thats what i would be telling my colleagues

 

[3tyretrackterry]

i have a computer set up and i recently invested in a garmin streetpilot. when i ran them both together they both said more or less the same thing. gps lags a bit on speed but distance wise they were both within 0.2 of a mile of each other. that i can live with

 

[hillrep]

Neddy wrote

On a similar vein, I've noticed that my regular commute has dropped from 7.55 to 7.4 miles, as measured using a wheel magnet. I have recently invested in a decent track pump and my tyres are now pumped up harder than they've ever been. Could this really account for a 2% difference?

Yes, I've seen similar differences when pumping up my tyres. Tyre pressure makes a surprisingly large difference.

 

[fossyant]

GPS plus wheel magnet - eg Edge 705 will be pretty accurate as it will rely upon the wheel sensor as well. The unit takes a mile or so to 'calibrate' when you first get the sensor.

Properly callibrated wheel magnet computer will be accurate too. You'll be say 0.1 mile out over 10.

GPS only..... acurate enough, although prone to 'skips' - GPS plus wheel sensor - I don't get any silly 'speeds' recorded as the wheel sensor is used as well.

 

[BentMikey]

I know you can get to 35 Gaz, having ridden with you. The GPS might be less accurate momentarily than a wheel based computer (think jumps in location and thus peaks in speed), but I don't think that's the case in your video. Mostly because the speed is consistent, changes steadily with time, and also because it's consistent with the perceived speed of the video.

 

[HLaB]

I found my gps would record a slightly lower average if anything (but not much in it, say 17.1 as oppose to 17.2). I put it down to the fact that a gps records your speed between two points (does this hundreds of times to make a track, so each individual segment is a straight line, where in real life you're path may be different. Thats my theory anyway.

If you are a fit cyclist, which you seem to be gaz, you'll have no problems hitting 35mph drafting traffic. I usually chicken out before then

 

[GrasB]

Assuming you're not getting severe drift a GPS is going to under-read the mileage but a wheel magnet is going to over-read. When properly locked/setup you're looking at ±10m or so meters in 1km.

 

[biking_fox]

Depends. A GPS will be accurate(ish) all the time. These days not effected by cloud cover, and can cope with surrounding high buildings. Not effected by wheel size or tyre pressues. Just clip it on, press go and you've got a record.

A Wheel magnet will be better, and/or worse. If you've got the outside tyre diameter correct, then it will be pretty much as accurate as you can get. However any error in this will quickly lead to both speed and distance accuracy loss. Even tiny errors will magnify quite fast.

And without an absolute standard you won't know which is the more accurate. Ideally you'd want a really precisely measured perfect 1k bit of flat road to set the tyre size on the wheel magnet through iterative runs. Once you've got that perfect, you can compare the GOS reading.

 

[GrasB]

A Wheel magnet will be better, and/or worse. If you've got the outside tyre diameter correct, then it will be pretty much as accurate as you can get.

As long as you ride bolt upright. As soon as you lean the bike into a corner the rolling radius of the tyre drops. It's more noticeable on motorbikes when you're really leaning into the corners & find that the rpm drops as you accelerate out of the corner

 

[HJ]

Well given that Garmin® GPS receivers are accurate to within 15 meters on average, I think you should be able to set up to wheel magnet to match that...