Spoke sensor for bicycle

Two advantages: it knows forward from reverse, and it has angular sensitivity to sub-one-revolution movement. It also doesn't need an adjustment-type mount for the sensor, just slap it around the axle and mount the wheel. There's nothing to prevent that kind of axle-mount item from having a magnet and quadrature reed switches, of course; the speedometer-cable is a bit of a hassle compared to a signal wire.

A better solution would be to instument a hub and axle, but the axle diameter is fixed by the bicycle frame, and the hub by the wheel-builders. That means redesigns of the mechanical bits unless one lucks out and finds a compatible generator hub off-the-shelf.

Well, there is also Wiegand-wire sensors:

.

.

But as others have noted, the spokes may be non-magnetic. This can be solved by adding a bit of zinc-plated mild steel tubing (automotive brake line) glued to at least one spoke with a silicon rubber adhesive. The spoke would pass through an inch of tube, with silicon rubber filling the gap between tube and spoke.

Joe Gwinn

Dimiter_Popoff snipped-for-privacy@tgi-sci.com wrote in news:t8aapr$ab6$ snipped-for-privacy@dont-email.me:

Most out there do not sense every spoke, but instead sense a metal flag attached to one spoke and most use a hall effect sensor transducer for that.

John Dope snipped-for-privacy@message.header wrote in news:t8afpb$t21$ snipped-for-privacy@dont-email.me:

With a range of error. At least unless you are using militsry grade hardware and getting military grade resolves logged.

Phones are slow at keeping up. Use one riding a bus and you can watch the lags in real time.

Johnny loses the phone on a bus gps test! Ehhhh!

The biggest win is it gets you away from the wheel. E.g., you could affix a bar magnet to the end of the cable-in-cable and detect it's rotation (to whatever extent you desire) as well as using its motion to generate *power* for the circuit (esp with the ultra low power technologies now available; think about the tire-rubbing generators that used to power headlamps...)

Yeah; those delivered 2 or 3 watts; enough for a halogen flashlight lamp, and to charge a 6V lead/acid battery for standstill lighting. If you use one built into a wheel hub, the dynamo pole count determines the AC cycles per mile travelled. The friction type can be precalibrated easily (the drive drum circumference is better known than the tire's) if it engages the tread rather than the sidewall of the tire.

Most of the time. The 14" studded tires I run in the winter have a smaller rolling diameter than the 15" tires the speedometer is calibrated for so it's off by about 5mph. For that reason I'm in the habit of looking at the Nuvi speed rather than the speedometer.

In some weather conditions or when I'm driving at the base of a ridge that blocks the southern sky the GPS is wildly inaccurate so much so it is easy to detect when it's having issues.

I often carry a GPSr when I'm hiking (Garmin Etex 20). The terrain is mountainous here so satellite coverage can be sketchy. Sometimes the distance is radically off as is the maximum speed. Unless I just rolled down a slope I'm probably not hitting 4.7 mph. I don't think the device has the intelligence to discard obvious false data points.

I've got one of the step apps on the phone. It says it does not use the GPS so it's working off the accelerometer data and it's surprisingly accurate for the mileage particularly considering there was no setup for the length of my stride.

And for most Japanese motorcycles it is +10%. The conventional wisdom is it's a liability thing. Since they can't control aftermarket tires they overstate the speed. You get used to subtracting 5 mph from the reading.

Carbon fiber spokes would be a problem too. The LCD odometer on my motorcycle is old and is only legible if the temperature is close to 90 F so I adapted a bike speedometer/odometer. I epoxied the magnet on the rim near the tire bead and built a bracket for the sensor. It works quite well and since I can program in the rolling diameter of the front wheel it is very accurate.

They just worked until the core of the flexible shaft broke. If you were lucky the strands didn't tear up the housing so you could buy a new core, figure out how long it had to be, cut if off, and crimp on the square drive end. If you were really lucky the transmission end of the drive was above the lubricant level so you didn't have Type F fluid running down your arm.

While I can get nostalgic for the good old days I'm not senile enough to forget what a pain in the butt it was when things just didn't work.

What a wondrous invention they were... All that aggro for a one candlepower headlamp. Somehow I associate those with the Sturmey-Archer three speed hubs.

I suppose driving a generator that way is better than driving the bicycle with a 26cc motor in a similar way:

I thought the concept might have been relegated to the trash bin of history but I've seen a similar design for a DIY low rent eBike.

"This measure must be combined with other factors outside the government's control, including satellite geometry, signal blockage, atmospheric conditions, and receiver design features/quality, to calculate a particular receiver's speed accuracy."

I know they use induced currents in aluminum cans to separate them using a magnet. The magnet spins under the conveyor belt making the cans jump off the end, while the rest of the stream simply falls off. A friend who worked in the industry complained that my crushing cans made this not work as well. But my cans are sold as aluminum, not mixed into the all-in-one stream.

I've not seen one in decades. They seem to predate the "10 speed" time in my life.

But, I recall them as putting lots of drag on the wheel. And, as a "rotation sensor", they would be inadequate as I doubt the signal would "integrate" well.

OK. That would be similar to my "spinning magnet" idea.

I don't think the output would translate into "revolutions" well, across the range of possible speeds. Likewise, I imagine it would still produce an output when the wheel moved *backwards* (?)

I've been looking for an "alternate" form of transportation for the little jaunts -- to the library, post office, etc. Just a couple of miles, likely off-road.

Things like bicycle, Segway, one-wheels, etc. come to mind. But, none have really checked all the boxes (e.g., carrying a bag of groceries on a bike or Segway is really not ideal).

But, recently, I'm liking the idea of a small gas powered generator tucked in the battery compartment of my electric wheelchair (!). This would eliminate the maintenance issue of the batteries ($400/set) as well as increase the possible RELIABLE range of the chair. (of course, you couldn't use it indoors but that's fine)

Building on that idea, a gas powered Segway? <grin>

Cargo bike? Or trike? Possibly electric?

You could carry a genset on a cargo bike.

Ch

On a sunny day (Tue, 14 Jun 2022 22:01:01 +0300) it happened Dimiter_Popoff snipped-for-privacy@tgi-sci.com wrote in <t8altd$uqm$ snipped-for-privacy@dont-email.me:

An other way would be a laser diode shining through the spikes and a photocell picking it up on the other side. Be careful not to burn out eyes of some kids. The acceleration sensor would work nice too, some code is on my site that uses a 6 axis accelerometer but not for speed.. math seems easy enough though. Did I not write code for a navigation system using only such accelerometers, yes, needed an oven, some uni redid the experiment .. its on Usenet..

On a sunny day (Tue, 14 Jun 2022 21:00:44 +0300) it happened Dimiter_Popoff snipped-for-privacy@tgi-sci.com wrote in <t8aicd$2uq$ snipped-for-privacy@dont-email.me:

You will have to average here and there I suppose. Seems reasonabale to me, but have not tried on a bike. Latency is very low.

On a sunny day (Tue, 14 Jun 2022 17:56:50 -0400) it happened Joe Gwinn snipped-for-privacy@comcast.net wrote in snipped-for-privacy@4ax.com:

Just thinking, use a camera looking down The speed can be derived from the moving pattern it sees Almost like a computah mouse...