I’m a small time speed freak, far from world class. Even so, I’ve been wanting to get some accurate measurements of top speed unicyclists. All this talk about how fast we’ve gone is just talk. Cycle computers, be they wired, wireless or GPS, are way to inaccurate to get precise and accurate times that are comparable to other riders.
Here’s what I propose: Let’s get a bunch of fast unicyclists together next summer in a central location (SF Bay Area?), and have every one ride through a 100 meter speed trap with electronic light-triggered timing. We’d need a flat and smooth street that we could close off. Preferably in a place without too much wind. This would take some organizational skills and significant entry fees, but a splendid time would be guaranteed for all… and we’d be able to move forward with something more than just talk.
So you bay area movers and shakers (John, Tom, Nathan, Corbin…) whadya think?
Sorry - I wasn’t meaning to have a go at you - hope that’s not how it came across! It would be a very impressive speed, but not impossible as John Foss has done the same speed on a standard 24" wheel (which is slightly smaller than a muni tyre).
That would be very interesting if it could be arranged. The same riders on multiple wheel sizes would be cool to see as well. I can spin faster on a smaller wheel than I will on a coker on similar sized cranks. The fastest I have been on a unicycle is on my coker, but not by that much (certainly nothing like the difference in wheel size) - my top speed on the coker is definitely governed by bravery (or lack of it) rather than spinning ability.
Not really the same thing as this thread though, which is about absolute maximum speeds rather than speeds sustained over some distance. By the very nature of it, absolute spontaneous top speeds are never going to be properly measurable and are always going to be a bit anecdotal. I reckon cycle computer readings are more reliable than Tholub suggests though, at least under normal riding conditions (and assuming they’re calibrated properly in the first place). Obviously they can be fooled by idling or rollback mounting, but in normal forward riding I reckon they’re pretty accurate except at very low speed.
Not at all! I’m glad you pointed it out. The question of speed comes up a lot with my non-unicycling friends, so it’s good to finally work out what is and what isn’t a reasonable speed on a unicycle.
I might have to invest in a cycle computer now rather than relying on the GPS so much!
Reed-switch cycle computers are known to register inaccurate top speeds due to spurious counts when the switch gets activated by vibration or bumps. This phenomenon is probably even more prevalent on unicycles than bikes because of our wobbling.
Maybe not. A regular (non-GPS) cycle computer doesn’t store individual measurement points, while many GPS units do. I reckon that a GSP is superior if you plot the speed points and then judiciously filter out any outlyers.
Awesome idea…this would give Madison the chance to prove his superiority under conditions that lesser riders will dare not dispute. Unless something comes up, and the time and date is inconvenient for him.
This is a common misconception. Usually good enough as an approximation but not for the purpose of the poster that you responded to. The high gear in a Schlumpf uni hub is about 1 to 1.545.
Yes to a GPS. At least for logging all your movements. EXCEPT that for accurate speed readings, the readings must be over a respectable distance and enough of them. Do not accept peak point readings as correct. If you pay attention to you car GPS it will occassionally throw an absurd but but brief reading.
A GPS is not like a set of rubbers across the path. It is derived statistically and so requires a multitude or readings. And if you lose signal for a moment, (pass under a biridge for example) it will fake the displacement it uses to make its calculations.
Some people may have noticed that the number of satellites that the GPS sees at any given time can vary from many to less than enough. Though I do not know, maybe 100m would suffice.
(As an aside, the US used to introduce a pseudorandom error in all the satellite transmissions so that the accuracy you could get without the decrypting keys would be only 25% of the capability. (Don’t want the enemy to use your satellites to target you as accurately as you target them.) Statisticians very easily defeated that by applying some clever algoritms, resulting in even greater accuracy the 2m that the hardware was capable of.
Yah and in 2000 the military stopped scrambling the signal.
GPSs are great for finding trends but not so hot for top speeds.
Most erroneous top speeds measured by GPSs are from temporary losses of signal while you are on some kind of turn. While the GPS has no signal it just extrapolates your position by last known coordinates, direction, and speed of travel. When it picks up the signal again boom you just went from a few meters off the trail to your current position in the blink of an eye giving a little blip of high speed.
If you look at the data points and filter out those spikes in speed you should have a fairly accurate top speed.
About as concise and clear as can be put. I love entering the Polly Farmer tunnel in Perth at, say, 78kph, add a couple of kph for the duration and then see what it makes of it when I come out the other end. (Small things amuse small minds.)
No. The shortest distance between two points is not a curved line…
For short-distance speed readings, car speedometers are probably the least accurate, followed by GPS and then rotation-counting computers. Nothing beats a timed speed trap.
Wait, but if you’re going on the curved line, and the GPS tracks the straight line, over the same time period, you’ll have gone farther than the GPS measured in the same amount of time so wouldn’t the GPS say you were going slower? Because it thinks you went less distance?
A tour I am frequently cycling is 15.6 km measured by my calibrated (weight-loaded) cycle computer, while GPS measuring typically results in 14.9 km. The reason is right visible on this image:
So, yes concerning the overall distance I agree that GPS gives the shorter distance. However concerning top speed I would believe Saskatchewanian is right:
You could easily set up your own distance area that you want to ride between, and stopwatch yourself and then calc the speed.
EDIT: Like what I would do is set up some markers on say… a bike path. Set something that easily divides from 5280, like 52.8ft, or 105.6ft, then stopwatch yourself between the 2 points as best as you can, and then multiply your time by either 100(for 52.8ft) or 50(for 105.6ft) and then divide that number by 60 to get minutes, and if you get a number larger than 60 then divide by 60 again to get your mph, presto!
Obviously the longer distance you use, the closer it would be to being accurate.
Rotation counters are fine, indeed pretty well as good as timing between two strip sensors. But only as long as the reader takes the average. Unfortunately our egos dictate that we take the flattering max which, I would agree with you, is totally useless.
GPS is ok except in certain situations where the signal is lost. But they do need a decent distance. I suppose hundreds of metres would suffice but even then it depends on how many satellites are tracking. At least from my rudimentary understanding.
Have a cop record your speed with his radar speed gun. If they are properly calibrated, it would be a good way to determine your speed. Finding a cop willing to do it is another thing. :o
Well, if you go for it in a pedestrian zone (or similar) you get at times a good service: print out of the measured speed and photo included. The strange thing is however the better you do, the higher the price.
