's Forum Archives - Components

Archive Home >> Components(1 2 3 4 5 6 7 8 9 10 )

Where do you put your speed sensor?(11 posts)

Where do you put your speed sensor?ParticleMan
Jan 31, 2003 8:11 AM
Hi All
I have a question to ask - where do you install your cyclo meter speed sensor? Facing the front or rear of the fork? At the top, middle or bottom?

I know, this may seem inane, but I'm quite fussy abt this...I try to put it as low as possible, facing the back of the fork.

Also, does it really make any difference in accuracy if the magnet is at the top or bottom of the spoke?

Appreciate your feedback. Have a good ride.
Makes no difference in accuracy; it's a rev countercory
Jan 31, 2003 8:40 AM
You can put it anywhere you want without affecting the accuracy--the sensor is really a revolution counter, and the top of the spoke goes around the same number of times as the bottom for a given distance.
I mount mine low, as close to the dropout as I can get it, because the spoke-to-pickup distance is small and consistent there. I tried mounting one just below the fork crown once, to reduce the length of wire, but it was hard to get it to read.
re: Where do you put your speed sensor?brider
Jan 31, 2003 9:18 AM
Realize the the closer to the rim, the faster the magnet passes the sensor. The faster the speed, the more likely it is that the sensor will miss a revolution (in addition, more wheel flex may increase the distance, increasing the likelyhood of missing a revolution). I had this problem with the first cyclocomputer I had (a Cateye Solar). Never had a problem with Avocets.
low and behindDougSloan
Jan 31, 2003 9:41 AM
I like them low and behind the fork. Low, so the magnet is closer to the hub and less "moment," and also it tends to put the magnet closer to the pickup, as the spokes are closer to the fork there. I then run the wire up the back of the fork, under the fork crown, up the brake caliper and brake cable to the handlebar. This route hides the whole thing as much as possible.

For wireless, though, you may need to mount the pickup much higher to ensure a good signal.

For carbon weave forks...brider
Jan 31, 2003 11:57 AM
I use sections of old inner tube hold the wire onto the fork. Blends in pretty well, can be easily adjusted, won't leave residue behind, etc.
Wherever it works.Spoke Wrench
Feb 1, 2003 6:56 AM
Considering the size of the frame, whether or not it has aero bars, the amount of wire available or the range of a wireless computer, where I can mount the magnet due to the crosses of the spoke, the distance between the sensor and magnet and any other stuff like canty mounts on the fork, usually the computer tells me where it has to be mounted.
Closer to the rimspeedisgood
Feb 2, 2003 8:13 AM
My theory is that you want to make the circle that the magnet traces as big as possible. The reason is to increase accuracy by having the sensor "sense" in the smallest %age of the total circumference. So if you put the mag closer to the axle, the it traces a small circle. Closer to the rim, it traces a big circle. The point is to make the ratio of sensitivity window:circumference as small as possible. I figure the sensor has maybe a 1 cm window of sensitivity wherever it is so: 1 cm/circumference of big circle << 1 cm/circumference of small circle.

In other words, putting the sensor closer to the axle makes it sense the mag for, let's say, 25% of the revolution. Putting it closer to the rim makes it sense the mag for, let's say, only 10%. If this happens, a sensor closer to the axle would sense a rev (one on/off cycle) for 2.5x longer than the one closer to the rim. Thus moving the sensor to the rim would improve the fineness of measuring your speed, assuming flex doesn't fudge things up (I don't think it has for me.) The only negative I've noticed is the sensor sometimes gets knocked by the tire when you remove/install the wheel. If you look at pics of pros' bikes, a lot have sensors closer to the rim.

Now that I think about it, this would only help if the computer counts each on/off, on/off, on/off vs. just on, on, on for each rev.

Hopefully, I was able to make sense to someone out there!
I hope you didn't put a lot of time into that theory.the other Tim
Feb 2, 2003 10:08 AM
As has already been pointed out, the computer counts revolutions. You'll get the same accuracy at any radius. The magnet is in close proximity to the sensor for a longer period of time if it is close to the hub, so that arrangement is sometimes more reliable, but it seldom matters.
Well, I didn't . . .speedisgood
Feb 2, 2003 3:25 PM
but it took me more than a few words to try to express what I was thinking. It actaully seems pretty intuitive to me. And for the record, I did notice that the resolution of my speedo was better after the change. The reading actually changed more smoothly and in tenths instead of 3 or 4 tenths increments.
Well, I didn't . . .Fez
Feb 2, 2003 6:41 PM
I think the best place for the magnet is where it can never miss a reading.

some have said it doesnt miss when it is closer to the hub because it is slower. Others just look to the area where they can put the sensor closest.

Experiment around.
I don't knowDougSloan
Feb 2, 2003 8:09 PM
I don't understand your theory. All the computer does is count clicks -- revolutions. It then counts the clicks over time and calculates speed. A click is a click. You can listen to the pickup audibly click as the wheel turns, if you have the bike on a stand. A click is a click. There aren't long clicks and short clicks.

Resolution is how finely the measurement is displayed. It could be 1 mph, .1, or .01, or some might go .5, etc. The computer does not know where the magnet is located, so how could the resolution vary?

I would think, if we are being intuitive here, that having the magnet outbound, making it cross the sensor faster, would make it more likely a click here and there at high speed would be missed, resulting in a too low speed indication.