Many posts have discussed the potential pitfalls of temperature change-induced tire expansion/contraction and its effects on changing the measuring constant. I'm sure most of us have had a measurement thrown off due to a rapid change in temperature, sometimes compounded by a sudden rain, and tried to find a way to combat this scourge of course measurement.
Dave Katz & I often measure at night in NYC for this very reason; the temperature is generally more consistent then, especially when the mercury is down in the single digits, which Dave seems to prefer for rides. (He also claims that riding in daylight would cause him to crumble into dust or burst into flames, I don't recall which, but that's a topic for another post).
However, I serendipitously came up with what I think may be the ultimate solution to this age-old problem.
It happened yesterday when I was preparing to measure a local 5 km course as a cold front was predicted to move through the area. As I dusted off my measuring bike after a winter of disuse, I discovered to my chagrin that not one, but both tires had gone flat since the last ride. With no spare tubes nor a patch kit, I made the drastic decision to remove both front and rear tires and ride on the rims.
I can't tell you what a difference this made! After an initial adjustment period to get used to the teeth-rattling ride, I found my calibration rides to be within 0.5 counts of each other, even though my cal course is significantly sloped, which usually produces a 3-4 count difference between the uphill and downhill runs.
But the real bonus came during the actual measurement. Between the 2 rides, the front moved through, dumping nearly half an inch of rain on the roads as well as lowering the temperature 15 degrees.
While normally this would produce a drastic change in the counts for each ride, the minimal contraction of the bare metal wheels virtually negated this effect, resulting in a difference of less than 10 counts between the 2 rides. Post-calibration, performed in even cooler temperatures, proved almost identical to the pre-cal figures.
An additional benefit of this technique is that the sound of bare metal rims on asphalt, similar to an inbound subway train, alerts motorists and pedestrians to the measurer's approach.
The only drawback I discovered was that to maintain traction on metal rims one must ride at a greatly reduced speed; I found anything over 3 MPH produced slippage, although I am confident with practice and improved technique I can get this into the 6-7 MPH range soon enough. And the time savings of not having to pump up tires before a ride should almost offset this, in my opinion.
No more worries about wet roads, changing temperatures, or flat tires - I'm sold on this new method. I hope to introduce it at any measurement clinics I give this year (Duane, I will send you materials for your presentation at RRCA Convention next month) and will present a year-long study at the USATF/RRTC meeting in December. In line with this philosophy, I intend to participate in the Measurement by Pacing Contest barefoot.
I feel certain this is the biggest breakthrough in measurement technique since the rear-mounted reverse-reading Jones Counter, which sadly never gained widespread acceptance among the measurement cognoscenti. Hopefully, this "bare wheels" technique will be received and accepted with more open minds.
Original Post
