Sunday, April 6, 2014

Technical Aside: Calibrating The Bike Computer Mathematically

Following a cue sheet depends somewhat on a reasonably accurate bike computer, especially in urban areas, where all one seems to do is turn every block.  A reasonably accurate bike computer ALSO means you don't have to wonder where the next town is, when you've already ridden further than the leg distance indicates!

I mean, yeah, you can figure out the difference between your computer and the cue sheet at the next turn and spend the rest of the ride doing math, but...

There are the tried and true methods of 1) using the tire measurement that came in the instructions with your bike computer (usually in 2 point font), and 2) the rollout method.

As I use a less common tire size, my purported tire dimension is never in the instructions.  And the rollout method wants one to ride in a straight line, involves chalk, and a measuring tape to measure in mm.  Can you imagine the inaccuracy that comes from that?

I am lazy.  Why do that when the awesome power of mathematics and spreadsheets are at my command?  Plus, so many cue sheets are generated from online mapping programs (Ride With GPS, Bike Route Toaster, etc), that I would really prefer that my computer match their distances.

Process is as follows:
Go for a bike ride.  The longer, the better.
Map that exact same route out on your preferred mapping program.  I recommend Ride With GPS.

You need to start with a couple of numbers:
1) your bike computer readout from your ride
2) the mapped distance for that ride
3) your current bike computer wheel circumference setting.  Mine is in mm.

First, solve for the number of rotations your wheel made on the ride.  Everything has to be in the same units, in this case, mm.  The formula below converts the bike computer distance in miles, to mm.  If your bike computer reads out in km, well, you'd just add 6 zeros :-)

rotations = (bike computer readout*1609344)/current wheel circumference

The 1609344 is mm in a mile.  If you were solving for cm, you would use 160934.4

You really want to do it with a spreadsheet, to preserve the precision through the calculations.  Of course, I remember doing similar calculations in HS Chemistry, with nothing but a Post Versalog slide rule.  Spreadsheets are better.

Now, the number of rotations will not change for the same distance, so, with the value of rotations firmly in hand, we can now solve for the desired wheel circumference:

desired circumference = (mapped distance*1609344)/rotations

Now the hard part for some, plugging the new circumference value back into your bike computer.

On your next ride, again check the bike computer readout against the online mapped distance.  You might need to adjust, but you will be much closer.

Quicker, alternative method, suggested by my husband:
desired circumference = current circumference*(mapped distance/bike computer distance)

I ran it both ways; the results were within a millimeter.


Kevin said...

Or.... do your ride. Compare your measured distance to the actual. Determine the percentage difference, adjust bike computer wheel circumference setting by that percentage. My newest bike now reads exactly 100 miles for a mapped 100 mile ride now, thanks to that method.

lynnef said...

that would be method #2, discussed at the end of the post.