Brocephus

A few days ago I installed a new Cateye Velo7, and I was pretty sure it was giving me significantly inaccurate (shorter) distances, based on my memories of some known distances (I also felt my averages and max downhill speeds were low, but that's more subjective).
When first setting it up, I went with 700x30 tires, though the tires I'm running are labelled 32's. (many reviewers said these Continentals were actually thinner than labelled, and my measurements concurred.)
Today I hopped on my motorcycle, which is pretty accurate and calculates 1/10 miles, and rode what the bike computer said was a 1.4 mile stretch, and the motorcycle said it was 1.7 miles (so I've got quite a disparity here.).
So, I figured the larger the diameter the computer thinks are on the bike, the greater distance, per rotation, will be assumed by the computer (I get that right?) So I just zeroed the computer, and set it up assuming a 38mm tire. (The next lowest options were 35 and 32mm, and I didn't feel either of those were going to make enough of a difference.)
Anyone else ever have this issue? And, am I on the right track for getting it dialed in?
thanks......

UniChris

Your apparent under-reporting (1.7 miles as 1.4) is on the order of 18% (or to put it another way you need to increase the reading by 21%) much greater than the variation in a tire's contribution to wheel diameter - your error is even greater than the difference between a 26 inch and larger 700c tires.

The manual for your computer seems to indicate that you enter the circumference in centimeters from a table in the manual. What number did you use?

Did you try a rollout test as described in the manual, doing one revolution from valve stem to valve stem?

Do you have any safe/legal bike routes with mile markers? Or a street you could get a distance figure for from Google Maps or a Strava segment of published length with endpoints readily identified by eye, rather than GPS?

Brocephus

Hey Chris, thanks for the speedy reply.
I was just following the set-up flow-chart in the manual, which seemed pretty straight forward, and appeared to go well.. I didn't notice any roll-out test recommended (but I'll take another look.). The manual gives a large list of tire sizes, all with a corresponding number to enter into the computer. The number for a 30mm tire was 215, and for a 38mm tire, it was 218, which is what I entered.
Once it cools off a bit more, I may hop on the bike and do that 1.7 mile ride and see what the computer tells me.

Lazyass

Do a roll out measurement for accuracy. Not all tires of the same size are actually the exact same size. Some 25c tires measure 28 (for example).

Brocephus

Okee-doke, I just laid out a tape measure across the garage floor, and rolling perfectly straight ( on two tries), I got exactly 64.5 inches (so what do i do with that now?)
Also, my brother just left here, and i asked him to zero his BMW's odometer and measure that same stretch for me, and he just called and said he got just over .8 miles, one way ( so basically the same as the 1.7 round-trip I got on my Honda XR ).
I'm about to ride it on the bike in just a minute, and see what I'm getting with this latest set-up.

UniChris

That would be about right for 20 inch wheels, but you said 700c. But if it were correct, you'd multiply it by 2.54 and enter into the computer.

This is the number of centimeters of wheel circumference or rollout. In inches that would be 218 / 2.54 cm/inch = 85 inches

Dividing that by pi gives a 27 1/3 inch rolling diameter which seems reasonable or only off by a few percent, not the huge difference you found.

Perhaps the magnet isn't close enough to the sensor and it's only counting some of the time?

Brocephus

Well that was weird. I just scooted up the road and back, with (presumably) a larger tire diameter entered, and actually got less distance recorded.(.68 miles, one way). So i pulled out my multi-tool and used the tiny Allen wrench to clear the computer, and reconfigure it, but this time I went with the basic "700" setting, and didn't try the fine tuning option, and once back home it read .82 miles ! So that's basically what my motorcycle, and my brother's car were getting, so it looks like I'm banging on all cylinders here.
I plan to get in about 45+ miles tomorrow, so I'll see how it averages out on a longer ride.
Thanks again to all for chiming in......

Brocephus

Hey Chris, this all threw me for a loop for a minute, so I knew I must have some wires crossed, so i went outside and measured again, (with reading glasses on this time !!) and the tape measure on the floor showed 84.5 inches, not 64.5 ! My bad !
And, I'm sure the magnet isn't too far away, if anything, it's on the verge of being too close.
Anyway, the computer's basic "700" setting appears close enough, I seem to have found my missing 10th+ of a mile, so i reckon I'm all good. Thanks again !

UniChris

Glad it worked!

The cable on mine broke a couple weeks ago when I decided to take the trite trailside repair stand photo. Repaired it, but didn't get it back together right, as setting out on my first century Monday morning the mount disintegrated. Fortunately I'd already figured the intended route's distance and had Strava running on a phone buried in my pack, so didn't really need it, though it's loss did mean I had only a watch and dead reckoning to augment memory of trail features to anticipate planned rest stops by. Plus mine was a cheap no-name anyway.