Do anybody know how to build a homemade powermeter for a hometrainer or how to calculate power generated on a hometrainer??????
Mtbiker.
Mtbiker.
home made powermeter or calculator......
- Thread starter highcadence
- Start date
If you're trainer has a stable power/speed curve (meaning it doesn't get easier as the thing heats up) you can just use speed as a proxy for power to track fitness through various interval sessions.highcadence said:Do anybody know how to build a homemade powermeter for a hometrainer or how to calculate power generated on a hometrainer??????
Mtbiker.
If your trainer is in this calibration chart well there you go. http://www.kurtkinetic.com/calibration_chart.php
If not you might have the same trainer as some one with a powertap that can probably give you the power curve.
If not you might have the same trainer as some one with a powertap that can probably give you the power curve.
highcadence said:Do anybody know how to build a homemade powermeter for a hometrainer or how to calculate power generated on a hometrainer??????
Mtbiker.
Hey dude. With a little know how, this could be a fun project. I'm about to do it myself.
First off: A little background with circuits will do a lot for you here but I'm sure you could figure it out either way.
*Eventually I may write up some real plans but for now you'll have to take an idea and run with it.*
1. Get a cheap trainer (even the cheapest will do, you really only need the frame and a few parts from the mechanism that provides resistance)
2. Get an electric motor (Electric motors can also act as generators when work is done on the shaft) - here you may need to search a bit, grab an old one off an electric scooter, or the motor that drives a compressor from a fridge being thrown away. Who knows, maybe a small cheap motor will provide enough resistance without tearing apart. You'll just have to experiment.
3. Get creative - find a way to use the resistance mechanism already on your trainer to drive the electric motor. This may be tricky, it may be easy, it all depends on the specific trainer and motor.
**Note: You need to remove or disable any piece of the resistance mechanism that is providing substantial resistance. If this is not done, the power meter will be waaaayyyyy off.
**Note: Your spin trainer may already be equipped with an electric motor. I haven't taken apart a magnetic resistance trainer but there's a chance there may just be an electric motor(generator) in there. If there is a way to drive a current from the resistance mechanism, you can disregard step two!
THE HARD PART IS DONE!!
4. Once you can drive a generator with your trainer, you can take the current produced and run it through a potentiometer (this is a variable resistor). The strength of the variable resistor (potentiometer) will determine the resistance on the wheel. As for specifics on which potentiometer to use, this depends on the motor you use so it may take some toying around. You can experiment with standard resistors first, they are cheaper (this will give you an idea of what range potentiometer to use). *There are many many ways to achieve different resistances in an electrical circuit, this is where circuit knowledge starts to help.
5. Buy two cheap multimeters from Walmart or Radioshack (you may already have one lying around) - measure the voltage across the two terminals of the motor and the total current in the circuit, Power = current * voltage.
your essentially done!
Of course for step five I intend to build a microprocessor and write a computer program that will give me a full report on every ride I do.
If you got the know how, you can make this VERY nice by means of a microprocessor!!
Good luck.
First off: A little background with circuits will do a lot for you here but I'm sure you could figure it out either way.
*Eventually I may write up some real plans but for now you'll have to take an idea and run with it.*
1. Get a cheap trainer (even the cheapest will do, you really only need the frame and a few parts from the mechanism that provides resistance)
2. Get an electric motor (Electric motors can also act as generators when work is done on the shaft) - here you may need to search a bit, grab an old one off an electric scooter, or the motor that drives a compressor from a fridge being thrown away. Who knows, maybe a small cheap motor will provide enough resistance without tearing apart. You'll just have to experiment.
3. Get creative - find a way to use the resistance mechanism already on your trainer to drive the electric motor. This may be tricky, it may be easy, it all depends on the specific trainer and motor.
**Note: You need to remove or disable any piece of the resistance mechanism that is providing substantial resistance. If this is not done, the power meter will be waaaayyyyy off.
**Note: Your spin trainer may already be equipped with an electric motor. I haven't taken apart a magnetic resistance trainer but there's a chance there may just be an electric motor(generator) in there. If there is a way to drive a current from the resistance mechanism, you can disregard step two!
THE HARD PART IS DONE!!
4. Once you can drive a generator with your trainer, you can take the current produced and run it through a potentiometer (this is a variable resistor). The strength of the variable resistor (potentiometer) will determine the resistance on the wheel. As for specifics on which potentiometer to use, this depends on the motor you use so it may take some toying around. You can experiment with standard resistors first, they are cheaper (this will give you an idea of what range potentiometer to use). *There are many many ways to achieve different resistances in an electrical circuit, this is where circuit knowledge starts to help.
5. Buy two cheap multimeters from Walmart or Radioshack (you may already have one lying around) - measure the voltage across the two terminals of the motor and the total current in the circuit, Power = current * voltage.
your essentially done!
Of course for step five I intend to build a microprocessor and write a computer program that will give me a full report on every ride I do.
If you got the know how, you can make this VERY nice by means of a microprocessor!!
Good luck.
By the way, by adjusting the potentiometer, you will be adjusting the resistance on the wheel, so you could have this all the way up sitting on your handle bars!
Or if you're planning on going the microprocessor route, you could write a program that will automatically vary the resistance with a digital potentiometer. That way your spin routine will be preset! digital coach! haha.
Or if you're planning on going the microprocessor route, you could write a program that will automatically vary the resistance with a digital potentiometer. That way your spin routine will be preset! digital coach! haha.
I have looked into building a DIY power trainer. Making a functional one would be fairly easy - making it accurately read out the power is more difficult. There are a few DIY projects that can be found on the net, but none that I have seen would be usable for high volume, structured training.
To get an accurate reading, you will need to know the efficiency of the generator / alternator you are driving. The power coming out is not equal to the power coming. Without accurate specs, you would need to calibrate it in some manner.
My initial plans were to use either an automotive alternator (cheap and done before), treadmill motor or a permanent magnet alternator (expensive). Then I would build a switched power regulator controlled by an Arduino or similar and network that to a graphical program running on a PC that would determine the power profiles and log data.
I could get that all set up for no less than a decent trainer with a documented power profile and/or power computer, then there is always the chance for failure. As a project it still appeals to me, but you will not save money by building one your own.
To get an accurate reading, you will need to know the efficiency of the generator / alternator you are driving. The power coming out is not equal to the power coming. Without accurate specs, you would need to calibrate it in some manner.
My initial plans were to use either an automotive alternator (cheap and done before), treadmill motor or a permanent magnet alternator (expensive). Then I would build a switched power regulator controlled by an Arduino or similar and network that to a graphical program running on a PC that would determine the power profiles and log data.
I could get that all set up for no less than a decent trainer with a documented power profile and/or power computer, then there is always the chance for failure. As a project it still appeals to me, but you will not save money by building one your own.
My current project in planning right now is to get an ANT+ speed sensor and be able to access the ANT+ data from a (Python) program. I can use the trainer calibration data to compute a virtual power, and plot out the speed, power and heart rate data in real time on a computer display. I could use a $35 raspberry pi for capturing and plotting the data.
Yep, definitely would be some power loss but what can you expect from a cheap rig! If you really put in some effort to make it nice and calibrated it, that would be pretty sweet. I like the alternator idea. Just might take that one.
Good luck with your project.
Good luck with your project.
Here's an update on the ANT+ project.
My initial feasability test worked.
I obtained a Raspberry PI from Newark Element 14 and a Rev2 ANT stick from Digikey. I used the official Rasbian (wheezy) image and installed the PyUSB, python-message and python-serial packages.
I found python-ant project on GitHub. This guy has done most of the hard work in getting communication to the USB stick working and implementing the ANT protocols:
https://github.com/mvillalba/python-ant
His code did not work with the revision 2 stick as is. I needed to add a handler for the startup notification message and change the demos to use the usb2 drivers. But after a few hours of tinkering, I was able to see my heart rate on the computer screen. I strapped the HRM to my 3 year old and had him run around for a test. It works pretty well, the CPU utilization is a bit higher than I would like, but I think I can tweak the code and do less screen updates and less event processing.
Things left to do:
1. Clean up the code (easy)
2. Add graphical capability (medium)
3. Add speed and cadence sensor capability (easy)
4. Add trainer power calibration (easy)
5. Add logging, possibly TCX file output and auto upload to your favorite website (???)
The ANT developer website www.thisisant.com has alot of useful reference material. The most useful is the device profiles which describe the channel configurations and the format of the data transferred. To access it you need to register and to sign the "Adopter" agreement - essentially you are agreeing to develop devices which conform to the standard and to keep the network key private.
So far I have spent about $80 on hardware, the only additonal item required is a speed/cadence sensor.
My initial feasability test worked.
I obtained a Raspberry PI from Newark Element 14 and a Rev2 ANT stick from Digikey. I used the official Rasbian (wheezy) image and installed the PyUSB, python-message and python-serial packages.
I found python-ant project on GitHub. This guy has done most of the hard work in getting communication to the USB stick working and implementing the ANT protocols:
https://github.com/mvillalba/python-ant
His code did not work with the revision 2 stick as is. I needed to add a handler for the startup notification message and change the demos to use the usb2 drivers. But after a few hours of tinkering, I was able to see my heart rate on the computer screen. I strapped the HRM to my 3 year old and had him run around for a test. It works pretty well, the CPU utilization is a bit higher than I would like, but I think I can tweak the code and do less screen updates and less event processing.
Things left to do:
1. Clean up the code (easy)
2. Add graphical capability (medium)
3. Add speed and cadence sensor capability (easy)
4. Add trainer power calibration (easy)
5. Add logging, possibly TCX file output and auto upload to your favorite website (???)
The ANT developer website www.thisisant.com has alot of useful reference material. The most useful is the device profiles which describe the channel configurations and the format of the data transferred. To access it you need to register and to sign the "Adopter" agreement - essentially you are agreeing to develop devices which conform to the standard and to keep the network key private.
So far I have spent about $80 on hardware, the only additonal item required is a speed/cadence sensor.
Hi. I have a Raspberry and would love to use it for logging my workouts. Would like to log power as well and include this in a tcx file.
The ant+ usb device i got is the one included when i bought my Garmin 910xt "Daystream" I think the name was of the device.
Can I help you in any way with this project?
The ant+ usb device i got is the one included when i bought my Garmin 910xt "Daystream" I think the name was of the device.
Can I help you in any way with this project?
Olapalm, what is your technical level of expertice? Are you comfortable in Linux and with Python, TkInter programming? If so I can send you a bit of what I have worked on.
Your ant stick should work with the Pi.
Since I have ridden my home indoor trainer only a couple of times this winter, I have not spent much time working on my project. Its pretty fragmented at this point, one version I have will scan for up to 4 HRMs and plot their data simutaneosly. I was planning on another version to pull in data from a cadence sensor as well and calculate a power output; I was also going to do some major cleanup work. I have not ventured into making a TCX output feature, but the format is simple enough that it should be easy.
There are no further technical hurdles to getting everything to work, now its just a matter of taking the time to do it. If you are looking for a polished, ready-to-use product, I cant help you there. There is a pretty active hackerspace for the Pi, you could propose it as a project.
I am using the code from mvillalba.
https://github.com/braiden/python-ant-downloader
https://github.com/mvillalba/python-ant
Your ant stick should work with the Pi.
Since I have ridden my home indoor trainer only a couple of times this winter, I have not spent much time working on my project. Its pretty fragmented at this point, one version I have will scan for up to 4 HRMs and plot their data simutaneosly. I was planning on another version to pull in data from a cadence sensor as well and calculate a power output; I was also going to do some major cleanup work. I have not ventured into making a TCX output feature, but the format is simple enough that it should be easy.
There are no further technical hurdles to getting everything to work, now its just a matter of taking the time to do it. If you are looking for a polished, ready-to-use product, I cant help you there. There is a pretty active hackerspace for the Pi, you could propose it as a project.
I am using the code from mvillalba.
https://github.com/braiden/python-ant-downloader
https://github.com/mvillalba/python-ant
well, I've just moved from Windows and C# to Linux and at the moment shell script, so to answer you question - No and no, but I learn quite fast. I have studied C# at university for 2 semesters, but that was some time ago and I also read Java at the university. I think I'd have to take a look at the code to see what I can make out of it. I do have the GSC10 sensor mounted on my bike and some different HRM belts that I could se if I can make contact with.
So if you point me to start I will see how this turns out.
So if you point me to start I will see how this turns out.
