Bycicle speedo rev counter

[gordon01639]

I have mounted the sensor over the rear rotor so that it is triggered 12 times in each rotation. The maths should be 1000/(rotor diameter * Pi)*12, the rotor is 250mm. The number that comes out of this is massively to big. I did a bit of trial and error testing spinning by hand with a clock in front of me and eventualy came up with the number 1445 which gives me a pretty good reading for rpm I assume that this number is the circumference of the wheel when used on a bike. anyone got any better maths for this. it would be great if it was more accurate. its not far off now.

Gordon.

[ghurd]

Try 167cm circumference, set it to read kilometers per hour.

Then 15.2 kmph = 152 rpm.

G-

[TomW]

gordon;

Are you looking for RPM ? It seems what you want to get? Be lots better if you could get less pulses per rev which could lower your "big number" issue.

Maybe rig the sensor and a single button magnet someplace not influenced by the main magnets and count 1 pulse per revolution?

Just one way to shear that number to perhaps a more usable size.

Cheers.

TomW

[gordon01639]

Hi Ghurd,

This only works if you are only picking up 1 pulse per revolution and the number is 1666. I have used the number 1445 and don't read the 10ths of a kilometre only the full kilometre. Just seems strange that the maths I have come up with don't seem to work. Long live trial and error.

Gordon.

[gordon01639]

Hi TomW,

I thought about that but it would mean fixing a bracket to the back of the rotor long enough to get past the influence of the magnets on the rotor, but I thought that might cause additional vibration as it would off balance the rotor. The maths must be quite simple just cant get my head round it. still thinking at the moment. Who knows I might have a eureka moment, even with a head full of half dead brain cells.

Gordon.

[Norm]

Put a white spot on the rotor infared LED shining on the spot, an infared sensor ( a la mouse parts) and it can count each revolution.

       ( :>) Norm.

[gordon01639]

Hi Norm,

maybe a little basic program

 if mousex and mousy > 0 the t=t+1

 cls

  print t

of course this could be expanded quite a bit linking it to a routine to count t in regard to per second.

Just kidding.

Gordon.

[wdyasq]

I've read through the HOLE thread a couple of times.... WTF is gordonXXXXX looking for?

One can guess 'rpm' one can guess 'Hz'. One can read the rest of the posts and think it might be 'rpm, but one can't be sure.

One can be real basic. State what one is looking for. THEN, some of us may have  clue.

Ron

[gordon01639]

Hi Ron,

I cut and pasted it for you so you dont have to scroll back up to the top.

came up with the number 1445 which gives me a pretty good reading for RPM

plus the subject speedo rev counter (rpm).

are you any good at maths? I need all the help I can get. Its quite a few years since I did much maths, guess I should bone up a bit. I just cant understand why the equation (at the top of this thread) doesn't work.

Gordon.

[Countryboy]

You can do it the hard way, or you can do it the easy way.

Hard way:  Leave out a lot of critical information, and wonder why we can't figure out a correct answer...

or

Easy way:  Put the alternator in a lathe.  Run at a few different known RPM's, and compare to RPM readout of bicycle speedometer.

[wdyasq]

I guess I didn't see the impiled period. Words without puncuation are just that.

One would need to know a bit more about the device. It counts revolutions and turns it into a length/time figure. One needs to know what the circumfrence of the 'ideal' wheel  is for a start. Then it is a matter of finding the number that corresponds with what you care to achieve.

Be sure everyting is in the same 'system'.

Ron

[Nando]

Gordon:

Let's clear the mess.

Tell us the type of bike counter and how many detectors does it have per revolution and what is the K factor adjuster it has ( depends on the diameter in CMS or MMs of the wheel).

Then what diameter you need to measure and the expected RPM

Nando

[SparWeb]

Put ONE magnet on the rotor that is dedicated to tripping the sensor.  Then program the computer with 1667 (mm) or 167 (cm).  I did this myself, and I have found the device very versatile.  I use a ring-shaped neodyminum magnet that I just slide over one of the hub's bolts.  The sensor can be anywhere within 6" of the magnet and it picks up the signal.

I also agree with Countryboy's earlier comment.  Until you run your system on a device with a precisely known RPM, like a lathe, you can't be sure about anything.  A multimeter with a Hz frequency display would also help.

[Capt Slog]

Usually, the sensor part of the bike speedo is a magnetic reed switch.  I'd be surprised if that can cope with or detect 12(?) poles passing it for every revolution, they are just not designed for it.  

My sons and I use these speedos on kite buggys, and we have struggled to find ones which will manage to give data from even a 17" wheel, and will continue to do so when the speed gets upto the 30mph mark.

[ghurd]

No idea how fast a kite buggy goes.

"Cateye" Enduro 8 and Mity 8 are rated for tire sizes down to 14" in their charts, and speeds to 65.9mph or 105.9km/h.

G-

[gordon01639]

All,

Thanks very much for all your suggestions and advice, I have managed to get it to work pretty good now, I mounted a magnet on the rear of one of the rotor studs and taped the sensor to the hub support arm behind it but I had to set the value to half of the 1670 at 835mm, it now works really well and timing it against a second hand comes out spot on. Maybe this particular speedo is calibrated differently, I think this is what Nando meant by the K factor. Thanks again for all your help.

Gordon.

sorry punctuation is not my strong point I get to carried away with what i'm typing.