urgent help with simple bike pedometer

[DavidHasselhoff]

I want to get the speed of 2 excercise bicycles into a pc app with reasonable accuracy. I have used an ezio (www.ezio.com) board (a simple pic based analogue 2 digital convertor) to provide the interface. Connected to this is a proximity sensor on the back fork triggered by magnets on the wheel.

This works, but too roughly. I think the problem is that the sensors and the interface are both working out of sync. I think i need a sensor that changes state each time it is triggered instead of sending a quick pulse.

please help and if you live in the uk(manchester) there could be some work to help on this project.

thanks in advance

 

[Pilot]

The 'problem' may just be the sensors. The interface should only react to the input received from the sensors. As most circuits will detect on the rising OR falling edge of the signal a change state detector would double the input.

A proximity sensor?

For low frequency you can get away with the magnets passing a reed relay. These don't switch too fast.

Another method is.... strip an old mouse down and examine the movement sensor within. An opaque disc passes through the sensor ( which is in a dark enclosure). The required number of slots - same size & equidistant - are cut into the disc in order to allow the sensor to detect motion. Input taken to pic then convert to speed in software.

Just an idea
:lol:

 

[mechie]

Wheel speed measurement

An old forum discussion on this subject went along this line...

Bike wheel with magnet on spoke; reed switch operated by magnet; f-to-v converter gives voltage proportional to speed. (See attachments)

Your current A-D converter is looking for a varying voltage to represent speed, your reed switch will only jump from 0v to +5v (or whatever). You need to either count these pulses over a fixed time (or time betweeen pulses) or produce an analogue signal proportional to speed (the attached drawings).

The LM2917 circuit was tested as OK :wink:

 

[DavidHasselhoff]

Mechie
many thanks for the post.
We have been counting the pulses and doing the maths at the pc end. Am i right in thinking that using the second circuit with the LM 2917 (£3 from maplin) I will continually get a variable voltage based on the frequency of pulses from reed switch. As the ezio board gives +5v will your circuit give a variable voltage in this range.

many many thanks again

if this all works out i might even give you Pamela Andersons mobile :wink:

 

[mechie]

Voltage Range

The output voltage range will depend on the R and C values you choose and the number of magnets on a wheel and the wheel's speed.

There is no reason why the circuit couldn't produce 0 to 5v, maybe limited with a zener diode to protect the IO card ?