ERTRO, tyre size and Cateye computers

[Wowbagger]

I'm using a Panaracer Pasela front tyre, 26*1.5, ERTRO 37-559.

I've measured the circumference of the inflated tyre which seems to be 1998cm.

The Cateye instructions give 40-559 as being 26*1.5 but reckon I should set the computer to a circumference of 2010cm.

They don't give an equivalent for 37-559.

Any views?

[hellymedic]

Try (559 + 2*37) * 3.1415926535897923284
1998.63 at first estimation

See how that seems on the road (maybe between mileposts) and adjust accordingly.

[rr]

This was done in great detail in the old place.

My 26 * 1.5 spec Nimbuses on Mavic XC 717 rim is 2010, wider rim or slightly ddifferent size tyre could make 3.5mm difference to the diameter. Try to measure the rolling diameter with you sitting on it.

[andygates]

Do a rollout.  If you've gotten this theoretical then you'll never be happy.  Find a hard floor and a minion, and do a rollout with you on the bike.

[Wobbly John]

A dab of paint on the tyre - ride the bike on a flat surface - measure distance between dabs of paint on the floor.

It is best to do this outdoors.

[LEE]

Get a big Tape measure (I use a 10 metre one), extend it fully on a path and lock it, so tape doens't fly back in.

put valve at bottom of wheel and at start of tape.

Push bike along side of tape (the more revolutions you get, the better (a 10m tape should give you 4 revolutions).

Divide the rolled distance by number of revolutions.

Even better, sit on bike while someone pushes you (squashed tyres make a difference)

I got 856cm / 4 revs = 214cm (incidentally this is exactly what cateye suggested for my 700x28c tyres)

[Ian H]

Or just throw the computer away.

I did.

[Greenbank]

I've measured the circumference of the inflated tyre which seems to be 1998cm.

That's a very big wheel. Approx 6.3m diameter! (I know you meant mm not cm).

I don't bother trying to get it perfectly accurate. The value you get for 100psi will be different than that for 50psi. Given that I pump my tyres up every 2 weeks or so I'll either be under/over reading pretty much all of the time.

[hellymedic]

I've measured the circumference of the inflated tyre which seems to be 1998cm.

That's a very big wheel. Approx 6.3m diameter! (I know you meant mm not cm).

I don't bother trying to get it perfectly accurate. The value you get for 100psi will be different than that for 50psi. Given that I pump my tyres up every 2 weeks or so I'll either be under/over reading pretty much all of the time.

This error is not quite as bad as the report I read in the local paper of a chap who was found to have 222 grams of alcohol per 100 ml blood...

[Sergeant Pluck]

Can you put in something approximate and then use your GPS to tweak it after longer rides, Wowbagger? I did this with mine - it only took a couple of tweaks to get it as right as it will ever be.

[Wowbagger]

I gave it a try on a ride this morning. The Cateye measured 14.85 miles. The Garmin measured 14.7. Oddly, within about 200 metres of the start of the ride, the Garmin's reading indicated a distance about .05 of a mile greater than that of the Cateye. I think I started riding a few seconds before the Garmin had properly found itself. For the rest of the ride, the Cateye gradually caught up. I think the Cateye might be reading about 1% high, but that will do for me.

[hellymedic]

Errr... What setting did you use? 1998 or something else?

[Greenbank]

GPSes don't calculate distance traveled very accurately because of the lack of accuracy of each location datum. The GPS device can perform some smoothing, but this is never going to magically claw back accuracy, the result of summing error prone data is a final value with a huge error range. They certainly shouldn't be used for calibrating anything.

Even if you used the GPS to find two points on a flat straight road which are 1km apart (with good GPS accuracy of +/-5m) this still translates to a possible 10m variance over 1000m, or 1%.

Compared to wheel circumference of 2000mm this could mean the real reading is between 1990mm and 2010mm depending on the GPS errors).

10km of perfectly flat straight road would be better (as you could get the error down to ~0.1%) but it is pretty tricky to find such a thing in the UK.

[numbnuts]

I use 214cm for my 700x28

[Greenbank]

Another problem with estimation and mile/kilometer posts is that the computer itself doesn't display the appropriate precision.

Most Cateye computers only display distance in hundredths of a unit. For km this would be 10m.

Therefore, if you ride a measured kilometer and the computer reads 1.00km then you could still be over 1% out as you may have traveled 1009.99m but the computer still displays this as 1.00km. Also, don't forget that the computer works by counting wheel revolutions which are only every 2m or so (depending on what setting you put).

This means that at the 1km marker you could have traveled 1014m (1.009999 + almost one full wheel revolution at the start and almost one final wheel revolution at the end).

That's an error of 1.4% when trying to compare it against a measured kilometer and the computer saying it was bang on 1.00km. If you did this with a setting of 2000m all it tells you is that the proper setting could be anywhere between 1972mm and 2028mm.