D
Doug
Guest
Hey all,
I ran more tests on these two beasts to determine their treatment of start and stop dead zones.
Refer back to the old thread of the (almost) same title for all the details.
Known previously:
The Astrale started its timer with the first pulse received from the speed sensor. The Specialized
started after roughly two seconds (between 2 and 4 pulses, depending on wheel speed). The two
seconds is about what the time would be, given the average starting speed of a takeoff.
The Astrale overshoots the stop by a full 3 seconds. The Specialized overshoots by about 2 seconds,
the same as the start delay. Hence I theorized that the Specialized would give a more accurate
average speed calculation.
But what about the dead zones, what do the computers do with pulses or phantom pulses?
I set the wheel size to 2000mm and units to km. The 0.01 distance resolution equates to 10m, or
exactly 5 wheel spins. Testing was straightforward from then on.
The new results:
Both the Cateye and Specialized are dead on in terms of distance. Each properly counts the number of
pulses and keeps the count in memory between starts and stops without fail or confusion. No phantom
pulses are counted while the timers run but the wheel is stopped, and no pulses are missed while the
wheel turns but the timer is stopped (Specialized only). Everything is A-OK with both.
The timer issue is murkier.
The Specialized was a bit screwy, in that if the wheel was turned to give pulses with say, 5 secs
between them, the pulses were recorded properly, the distance incremented properly, but the timer
never started. So I could rack up any distance but have no elapsed time. Admittedly a minor issue
given how these things are used, but nonetheless.
The Cateye, since it starts the timer with the first pulse, could rack up elapsed time readings
with very little distance covered. After my testing at one point, the Cateye showed over a minute
of elapsed time, but I had covered only 0.06km or so. That from, say, 10 secs of actual wheel
spin time.
So in the end, how does this mess up average speed? Given my riding style, about 60 miles in 6 hours
with lots of starts and stops, the old 1993 vintage Specialized would indeed give slightly more
accurate readings. Its counterbalanced delay and overshoot give more accurate timer readings. The
Cateye's consistent 3 second overshoot adds a few minutes to my elapsed time, and about 0.1 to 0.2
mph off my average speed.
Doug
I ran more tests on these two beasts to determine their treatment of start and stop dead zones.
Refer back to the old thread of the (almost) same title for all the details.
Known previously:
The Astrale started its timer with the first pulse received from the speed sensor. The Specialized
started after roughly two seconds (between 2 and 4 pulses, depending on wheel speed). The two
seconds is about what the time would be, given the average starting speed of a takeoff.
The Astrale overshoots the stop by a full 3 seconds. The Specialized overshoots by about 2 seconds,
the same as the start delay. Hence I theorized that the Specialized would give a more accurate
average speed calculation.
But what about the dead zones, what do the computers do with pulses or phantom pulses?
I set the wheel size to 2000mm and units to km. The 0.01 distance resolution equates to 10m, or
exactly 5 wheel spins. Testing was straightforward from then on.
The new results:
Both the Cateye and Specialized are dead on in terms of distance. Each properly counts the number of
pulses and keeps the count in memory between starts and stops without fail or confusion. No phantom
pulses are counted while the timers run but the wheel is stopped, and no pulses are missed while the
wheel turns but the timer is stopped (Specialized only). Everything is A-OK with both.
The timer issue is murkier.
The Specialized was a bit screwy, in that if the wheel was turned to give pulses with say, 5 secs
between them, the pulses were recorded properly, the distance incremented properly, but the timer
never started. So I could rack up any distance but have no elapsed time. Admittedly a minor issue
given how these things are used, but nonetheless.
The Cateye, since it starts the timer with the first pulse, could rack up elapsed time readings
with very little distance covered. After my testing at one point, the Cateye showed over a minute
of elapsed time, but I had covered only 0.06km or so. That from, say, 10 secs of actual wheel
spin time.
So in the end, how does this mess up average speed? Given my riding style, about 60 miles in 6 hours
with lots of starts and stops, the old 1993 vintage Specialized would indeed give slightly more
accurate readings. Its counterbalanced delay and overshoot give more accurate timer readings. The
Cateye's consistent 3 second overshoot adds a few minutes to my elapsed time, and about 0.1 to 0.2
mph off my average speed.
Doug
