Homemade Cycling Powermeter

21,120

71

26

Posted

Introduction: Homemade Cycling Powermeter

Hello!

In my latest project I tried to realise a simple and cheap cycling powermeter based on Arduino. After a month I can say "Yes I can".

There are many commercial products (SRM, SRAM ...), but I'm always looking for a homemade solution. On the following pages I'll show you my successful construction.

Step 1: The Principle

To calculate the power, we have to determine the current force, which is spent by the foot on the pedal. For this purpose I clued four strain gauges (http://www.ebay.com/itm/5Pcs-Pressure-Sensor-Precision-BF350-3AA-BF350-350-Strain-Gauge-Resistance-/361278029289?hash=item541dd93de9:g:oOYAAOSwstxVOFv2) on the sides of my old crank. To get a Signal depending on the load I arranged them in a so called Wheatstone-Bridge.

During one full rotation I sum as many forces as possible and calculate the average. Combined with the average velocity (2 * Pi * crank-radius / rotation-time) I get the power P.

To know the proportionality between the force F and the output-voltage U I strained the crank with different masses and myself (a lot of mass ;-)). The rise of the line is in my case 292 Newton/Volt.

For the amplifier I took a LF353 with a gain of 330. With this Setup I get Output-voltages between 1 and 4 V. I decided to start at 1V to avoid a negative drift and late response. Therefore I have to determine the offset at the beginning of the measurement and subtract this offset from the following voltages.

Step 2: The Setup

For the power-supply I use two lipo-battery packs (7.4V, 1200 mAh for the receiver and a smaller 600 mAh one for the transmitter).

To save space I use an arduino nano and the NRF24L01 are responsible for the communication. They work great and you can send as many values as you want. In my case I send the power (P) and the rounds per minute (rpm).

I also had to check, when a full rotation has accured. To avoid a sensor, which has to be mounted on the bike-frame, I decided to use the gyroscope MPU-6050. The accuracy is about +-3 degree per full rotation, which is satisfying.

Step 3: The Completed Crank and Receiver

The challenge was to fix up all the components (power-supply, gyroscope, NRF24L01, amplifier, arduino) on one crankarm but I succeded. I fixed them with double-sided tape and cable ties.

To mount the receiver on the bar I use the clamp from a bicycle-lamb.

Step 4: The Results

After calibration and programming I first tried my powermeter indoor on my training roller. I got reasonable values and therefore I went outside. Heureka, it works :-)

Finally I can say, that it's possible to build a simple and working powermeter for

* 2 x arduino nano ..... 10 USD

* 2 x NRF24L01 ......... 5 USD

* 16x2 LCD ................ 5 USD

* 4 x strain gauges ..... 10 USD

* 2 x Lipo battery pack ..... 15 USD

* MPU-6050 ..................... 3 USD

* electronic parts .............. 7 USD

less than 55 USD.

Here is the video: https://www.youtube.com/watch?v=v1aaupmxOsI

Thank's for visiting my thread. Maybe you're interested in some of my other projects too:

https://www.youtube.com/user/stopperl16/videos

Share

Recommendations

  • Epilog Challenge 9

    Epilog Challenge 9
  • First Time Author Contest 2018

    First Time Author Contest 2018
  • Sew Warm Contest 2018

    Sew Warm Contest 2018
user

We have a be nice policy.
Please be positive and constructive.

Tips

Questions

25 Comments

Nice project :) I did something like this in the last year, but i kept things minimal. Here are few photos of my powermeter

temp_-1136668075.jpgtemp_2049444351.jpgtemp_912262100.jpg

Hi friend, Would you give us details about your project? I mean, arduino model and etc.

Did you use two interrupts in your program? I think that would make a good idea Instructable. You might win some contests!

At the Moment I use a while-loop. while(angle < 360 degrees) add the forces. if angle >= 360°, calculate the average force F and speed v ---> P = F*v, degrees = degrees - 360° ----> next while-loop

Great work ;-)

Another use of the cadence sensor is power saving. When you pedal, the mcu turns back on and when it times out, it goes to sleep.

Where did you get the calibration weights? Shouldn't they be precise?

Hi! The more precise they are, the more precise the calibration will be.... I took weights from dumbbells and my own half body-weight (Standing on both pedals...)

Actually your transmitter's may work with 3.7V. You may have to change the reference voltage. Half the current through the strain gauges nearly doubles the battery life!