D
Dirtroadie
Guest
I ran into an interesting dilemma yesterday.
My hometown hosts has annual winter festivities involving many types
of events, social, athletic, indoor, outdoor, frivolous, serious etc.
see:
http://snowdown.org/
The bicycle related part of this is the roller races, an event with a long standing history
http://snowdown.org/wednesday.htm#roller
Prior to this year two riders would go head-to-head, the official measuring device being a dedicated
front wheel with a cyclometer mounted to an arm clamped in place with the front wheel quick release,
the same wheels being used for all riders. While this was accurate, it didn't allow a rider to use
his/her own front wheel which generally also meant giving up one's own computer readout unless the
bike had a rear wheel speed sensor. It also didn't provide any direct way for the spectators to
directly compare the progress of the competing riders.
So this year a new system was developed to solve the aforementioned problems but also attempting to
keep the roller system essentially intact (no extra variables , i.e. drag) so that distances and/or
records could be compared from one year to the next.
The new system placed magnets directly on the rollers themselves matched to sensors for Sigma
cyclometers (properly calibrated for the roller size). Preliminary tests showed the system to be
perfectly calibrated to the origanal system. Then the cyclometers were positioned together so that a
video camera could capture an image of all three simultaneously (3 riders at once this time around)
and project it onto a wall. Pretty slick.
Or so it seemed. Come race day it turned out that the calibration seemed way off. And, in fact,
EVERBODY was riding almost exactly the same distance. Every race seems to have the riders within .01
miles of each other from start to finish. (call it a "no rider left behind" policy) And apparently
what was happening was that the rollers (4.5 inch) at race pace were turning about the same RPM as a
normal front wheel would be doing at 240 mph! (Top end race pace is 40-ish mph, our new unofficial
record is 10.3 miles in 15 minutes) In fact anything above about 25 mph was wasted effort as far as
the "official" devices were concerned.
So either the reed switch in the pickup is beyond its capabilities or the electronics trying to
"read" and calculate from the pulses of the switch are not fast enough to keep up.
So I'm seeking proposals of a good way of setting up an accurate system taking into account (1)
spectator friendliness (2) consistency with prior roller setups and (3) no need to modify the bikes
at all. And keep in mind this is a low-key not-for profit, no prize money, fun type event, so
expensive and exotic computerized mechanisms are not likely to to be used. And we are talking
rollers, not trainers.
While we're at it, since this is a little different from actually riding along on the road, how
would you set up a bike (let's ignore the training for now) to optimize the results (go fastest) on
rollers? The first things that come to mind for me are (1) Disk wheels would be nice if possible
(aerodynamics) although nobody used them, (2) the heavier the wheel the better for the flywheel
effect, and (3) tires with low rolling resistance and/or using the highest tire pressure that the
tires can tolerate, (4) squeaky clean and well-lubed drive train.
DR
My hometown hosts has annual winter festivities involving many types
of events, social, athletic, indoor, outdoor, frivolous, serious etc.
see:
http://snowdown.org/
The bicycle related part of this is the roller races, an event with a long standing history
http://snowdown.org/wednesday.htm#roller
Prior to this year two riders would go head-to-head, the official measuring device being a dedicated
front wheel with a cyclometer mounted to an arm clamped in place with the front wheel quick release,
the same wheels being used for all riders. While this was accurate, it didn't allow a rider to use
his/her own front wheel which generally also meant giving up one's own computer readout unless the
bike had a rear wheel speed sensor. It also didn't provide any direct way for the spectators to
directly compare the progress of the competing riders.
So this year a new system was developed to solve the aforementioned problems but also attempting to
keep the roller system essentially intact (no extra variables , i.e. drag) so that distances and/or
records could be compared from one year to the next.
The new system placed magnets directly on the rollers themselves matched to sensors for Sigma
cyclometers (properly calibrated for the roller size). Preliminary tests showed the system to be
perfectly calibrated to the origanal system. Then the cyclometers were positioned together so that a
video camera could capture an image of all three simultaneously (3 riders at once this time around)
and project it onto a wall. Pretty slick.
Or so it seemed. Come race day it turned out that the calibration seemed way off. And, in fact,
EVERBODY was riding almost exactly the same distance. Every race seems to have the riders within .01
miles of each other from start to finish. (call it a "no rider left behind" policy) And apparently
what was happening was that the rollers (4.5 inch) at race pace were turning about the same RPM as a
normal front wheel would be doing at 240 mph! (Top end race pace is 40-ish mph, our new unofficial
record is 10.3 miles in 15 minutes) In fact anything above about 25 mph was wasted effort as far as
the "official" devices were concerned.
So either the reed switch in the pickup is beyond its capabilities or the electronics trying to
"read" and calculate from the pulses of the switch are not fast enough to keep up.
So I'm seeking proposals of a good way of setting up an accurate system taking into account (1)
spectator friendliness (2) consistency with prior roller setups and (3) no need to modify the bikes
at all. And keep in mind this is a low-key not-for profit, no prize money, fun type event, so
expensive and exotic computerized mechanisms are not likely to to be used. And we are talking
rollers, not trainers.
While we're at it, since this is a little different from actually riding along on the road, how
would you set up a bike (let's ignore the training for now) to optimize the results (go fastest) on
rollers? The first things that come to mind for me are (1) Disk wheels would be nice if possible
(aerodynamics) although nobody used them, (2) the heavier the wheel the better for the flywheel
effect, and (3) tires with low rolling resistance and/or using the highest tire pressure that the
tires can tolerate, (4) squeaky clean and well-lubed drive train.
DR