Hi,
Installed a new computer the other day and after the first ride on a usual course, I noticed my total distance was less than what it should have been. I checked the chart in Sigma's manual and had that correctly set. I rode again with a friend and he came up with approx 1.5 miles more post ride. He suggested that my sensor on the spoke/fork was not at the right distance from the rim. I re-read the manual and found nothing indicating this type of measurement.

Is there a set distance where the sensors must be on the spoke/fork?

Thanks,
Mike

No. You should have to input the wheel+tire diameters to the computer(s), which works off of revolutions/unit time to calculate speed and distance.

Thank you. It's never been an issue with any past computers. Wondering if I should do a roll out to be more accurate.

Yeah, there's quite a bit of variance in manufacturer's suggested circumferences. Measuring the roll out is the best way to get more accuracy.

I've used Sigma computers before, and they were always preset to 622-23 (another way of saying 700c-23). But you should check what yours is set to. To measure rollout, I put a small dab of white-out on the tire, then rolled the bike along the sidewalk and then just measured the distance between the white-out marks on the sidewalk. FWIW most computers will be using the metric system on this measurement.

Since the sensor is only counting revolutions, it shouldn't matter where on the wheel it's mounted. I've seen recommendations to mount it as close to the hub as possible, because that's where the velocity of the magnet will be lowest for a given speed, to it should be able to pick up each revolution better, but it's never made a difference for me.

JB

edit: OP, did you mean that the sensor is too far from the spokes (and thus magnet), so it's not picking up each revolution? That is an easy fix - the sensor should be mounted on o-rings or cable ties - just move it in a little. Or move the sensor and magnet down closer to the hub - that's where the spokes will be closest to the fork.

Maybe your friend's computer is the inaccurate source.

I use the rollout method, then compare my readings to about 4 or 5 mileage markers (painted lines/stripes on the road) of straight sections at the trail. Usually off on a line or two so I'll set the computer to match as best as I can.

"Censor" ?? I think you mean "censer":



Hang it from the nut on the front wheel skewer, or handlebars.

Maybe he did mean 'censor'

This. Always do a roll out for the best accuracy.

Signum Crucis

My bud's computer is dead on. As mentioned, I ran a familiar course and knew what the mileage should have been. I'll do a rollout.

On some bike computers accuracy was improved mounting the sensor and wheel magnet closest to the rim part of the wheel. This worked for me on my Trek computer.

Another method of doing the roll-out is to start with the valve stem at the exact bottom, perpendicular to the road/sidewalk/driveway, and mark the spot on the road with chalk or tape. Roll the bike until the valve is once again at the bottom, mark the road, and measure the distance between them. This way you don't end up marking anything on the tire. Just sayin...

and this

only makes sense if this somehow affects the contact/sensing of the relationship between the wheel magnet and fork sensor. As others said, it's just counting revolutions, and whether it's right at the hub or rim shouldn't matter...as long as the sensor and magnet are properly spaced.

Charles

A magnet near the rim passes the sensor faster than one placed near the hub. Nearer the hub appears to be more reliable (according to Sheldon Brown).

The gap between the sensor and the magnet is more crucial.

This doesn't make a lot of sense unless this corrected (reduced) the gap between the magnet and the sensor.

Keep in mind if doing the roll out method and you want to be accurate you need to be fully seated on the bike in riding position with tires inflated to your normal pressure.

If it's working at all, it shouldn't much matter where the magnet goes. If you get it set wrong where picks up erratically, it might throw the distance off, but it's also going to be showing zero mph part of the time, too.

If it skips a few counts here and there it wont show 0. Most of these devices average a few counts. I recall the cateye I had when it was missing counts would do something like reading ~22mph, the drop to ~17mph, then pop back up.

I just compare it to the GPS data from my phone...

If my phone tells me I went 14 miles and my bike computer tells me 13.97miles, it's good enough.

My bberry GPS using Endomondo tells me I rode 34 miles and my wireless bike computer tells me I rode 40. Of course, my bike computer tells me my max speed was 74 mph, so I suspect there's a little issue with wireless interference . . . .

First thing I thought is AYHSMB

Why? Does the rubber in the center of the tread stretch or shrink when the sidewall deforms?

Although the tire won't be in a perfect circle anymore, screwing up attempts to use radius or diameter to calculate anything, the distance around the center of the tread (circumference) hasn't changed when you are on the bike - it's just in a different shape. Of course, this is very easy to verify by doing a rollout measurement.

JB

https://www.mathworks.com/access/help...rive/tire.html

or if you trust the so called experts...
https://en.allexperts.com/q/Tires-235...ence-tyres.htm

The one measurement that is required is the shortest distance from the center of the hub to the ground (the perpendicular line from the center of the hub to the ground). This number represents the radius of a virtual circle that is somewhere in between the rim and the edge of the tire.

With more weight or less tire pressure, the rim is closer to the ground (and the distance is shorter). That is, the gap between the rim and the ground gets smaller (the tire gets squished between them like a marshmallow).

Since the relationship between the circumference and radius is a formula, it doesn't matter which one you measure. That means one should pick the one (either radius or circumference) that is the easiest to get.

The "rollout" procedure is a convenient way to measure this number in a way that generates an average of many (an infinite) number of samples. This average automatically takes into account the variation you'd get from measuring the radius multiple times (it increases the accuracy of the radius measurement).

The other advantage of the rollout procedure is that it also automatically deals with variations in tire depth between different tires (different brands of 25 mm tires don't have the same depth).

The squished rubber actually moves at a different speed than the rest of the circumferential rubber!

Anyway, theory and calculations aside, roll out and distance measurement tests I've done confirm a difference for both load and psi on circumference and distance accuracy. Even a 10psi delta on tire can change measured distance by about 0.25% - that can make or break actually completing a century.

Nope. If you took a tape measure all the way around the outside of the tire, it will give the same distance whether the tire is loaded down or not (within reason - the material obviously does stretch in the real world, but not enough to matter). This is the circumference. The rollout measures circumference directly, as if you had cut the tire and laid it out flat (hence my question of whether you think the tire stretches or shrinks when it deflects). if you have a perfect circle, you can calculate radius/diameter from the circumference or vice versa, but if it's not a perfect circle the mathematical relationship doesn't hold anymore (I guess you could always figure out a different one, but why bother - the circumference is the number we're interested in and can measure it directly).

Remember, the calculation we are interested in is circumference * revolutions. Non round shapes also have a circumference and can revolve. The infinite sample stuff is just a red herring. Sorry, try again.

JB

edit: what happens when you load a tire is that the contact patch changes - it gets both longer and wider. The increased length of the contact patch basically accounts for the difference in the apparently reduced circle size. However, the circumference hasn't changed - if you are careful and measure using exactly the same point each time, there should be no change.