Trainer that produces power graphs?

[tbaier]

Physics teacher looking to spend a little bit of the possible leftover end of the year science budget.

I'm trying to develop a lab whereby students pedal on a bike at different cadences and we measure power, heart rate, speed, etc. - whatever we can, but definitely power. I'd like to be able to save the data in a graph so that we can do some simple analysis. We have the bike and computers already, just need something to do the data collection (and allow us to spin).

Anyone have any idea what might be a good fit for that task? Not looking to break the bank - a few hundred dollars would fly. 1K+ won't. Your input is appreciated.

[ttusomeone]

Sounds like a computrainer is what you need - RacerMate.

However, they are over your budget new although you may be able to find one used or get a special rate because you are using it within a school. You'll be hard pressed to find any new power meter for less than $700 - you could check out StageONE though - they have one that goes in the crank arm. With this route you'd still have to get a trainer to put the bike on though, and then a bike computer capable of reading the Ant+ signal from the power meter.

[brassnipples]

Maybe someone will chime in with a trainer that does specifically what you need, but I'll describe my pseudo power setup which might be helpful if your budget is tight and you don't need high precision. I wasn't interested in a proper power meter, certainly not for how much they cost, but I did want to be able to objectively measure output during trainer sessions. So I worked on making a reproducible trainer rig that controlled as many variables as possible and it has since grown into it's own little scientific experiment.

The trainer has to be one that isn't affected by heat so fluid trainers are out but wind, mag, and friction will all work, I went with the 1up USA trainer which isn't perfect but seems to be close enough. I picked up a trainer specific tire because they last a lot longer and don't heat up as much as normal rubber so those two things don't affect the numbers as much. To measure speed I grabbed a Wahoo fitness speed/cadence sensor that wirelessly sends data to an ANT+ USB stick and I use Golden Cheetah to track the data. GC is great, it's free and has tons of different ways to view the data that it collects including a variety of live graphs to watch speed, heartrate, cadence and it if it is aware of your trainers power curve it will display a power estimate based off the speed.

I came up with a procedure to make sure the tension is always the same on the rear wheel, set the tire pressure to a specific psi, and make sure the tire and roller are nice and clean to make sure the setup is consistant from workout to workout. Then to validate I spin the wheel up to 25MPH and time how long it takes to spin down to 0.

The power you get out of GC is not going to be accurate in the absolute sense but it should be pretty close in the relative sense. So it may not be correct when it says you're producing 100 watts (though it should be in the ballpark) but it should be fairly accurate estimating how much more or less power you are producing relative to the rest of the workout.

The wattage numbers that GC is estimating is about where I would expect my power to be so it seems to be in the ballpark in the absolute sense but more importantly the consistent setup lets me track how my output changes from session to session.

So if no one else chimes in with a viable packaged solution maybe this will help. If nothing else it's actually a pretty good exercise in setting up an experiment to control variables. I've been pondering starting up a super geeky thread to see if anyone with a power meter could try to reproduce a 'standardized' setup and we could try to validate the virtual power numbers, then those of us without power meters could get a more accurate sense of our absolute power.

[PissedOffCil]

A used PowerTap might fit your needs, it happens I'm trying to sell mine. You could couple it with an ANT+ stick and use GC as stated above or an ANT+ dongle for iPhone/iPad and whatever app suits your needs.
Powertap - Open Pro rear wheel and accessories - RoadBikeREVIEW.Com

Another alternative is to use the SportTracks Trainer Power plugin that maps speed to power for many different trainers. It works OK depending on setup and the trainer/tire used as brassnipple stated.

[WR304]

The Elite Arion Digital rollers might be what you're looking for. Ant+ £495gbp / $770 USD and available soon:

Arion digital - Elite /en

Elite Arion Digital Parabolic Resist Rollers | Buy Online | ChainReactionCycles.com

You could either output its information straight to Goldencheetah or have something like a Garmin Edge 500 set to record the data before downloading it onto a PC.

I saw it mentioned on the DC Rainmaker site where he was planning on reviewing it in the next few weeks.

[spsoon]

Kurt Kinetics publishes a power vs speed function for their Road Machine trainers. So if you can measure speed (with a simple bike computer), you can calculate power.

[AztecRider]

Physics teacher looking to spend a little bit of the possible leftover end of the year science budget.

I'm trying to develop a lab whereby students pedal on a bike at different cadences and we measure power, heart rate, speed, etc. - whatever we can, but definitely power. I'd like to be able to save the data in a graph so that we can do some simple analysis. We have the bike and computers already, just need something to do the data collection (and allow us to spin).

Anyone have any idea what might be a good fit for that task? Not looking to break the bank - a few hundred dollars would fly. 1K+ won't. Your input is appreciated.

Hi Tbaier,

I created a device to measure power off a Kinetic trainer and display it and graph it on a laptop. Maybe I can help you. I am at work now, in the evening I'll send you a picture.

[desnaephoto]

As a physics and chemistry teacher, I'm interested in this project. I wonder if TrainerRoad would be willing to work with you on this. They already have the ability to do what you are wanting -- get the data, export to the computer or simply have the kids transfer it to their computer.

[dhinten]

Here's the formula for the Kurt Kinetic conversion from speed to Power.
Watts= 5.2448 * mph+0.01968*mph^3

Would seem to be a pretty reasonable measure of relative power even if the absolute values may be a bit off. Somewhere on the web I saw of comparison of this with a powertap and they tracked each other real well.

[spsoon]

Analyzing potential sources of error is a good classroom exercise in any case. For example, plugging your measured speed into that formula only gives you the steady-state power at that speed.