Ifor's orienteering training log


Page 1 of 2  > >>

Jan 3, 2012
Category: Website
Posted by: ifor
Forum added
Jan 2, 2012
Category: Website
Posted by: ifor
OK upgraded the site CMS system just now to 1.10.2.  hope everything still works.
Oct 30, 2011
Category: Website
Posted by: ifor
Added a blog section to the site.  I need to sort out the templates a bit as things are not looking so good.
CMS - 1.10.3 - Hyacinthe


Password (again)*
email address*

Enter the text from the image:
This is a captcha-picture. It is used to prevent mass-access by robots. (see: www.captcha.net)

I have already completed the registration form, but did not receive an email, can you please re-send it Click Here
Apps » IpBike » How to set up Virtual Power to match a published power curve?

Started by LisaB Mar 20 2021, 22:24
Posts: 5
Mar 20 2021, 22:24

Hi, I am trying to set up Virtual Power for a stationary bike where the manufacturer has published full power curve data that the bike was calibrated to during manufacturing. I'm using a speed sensor and no other sensors.

It's a magnetic resistance bike, so there are no gears, just different levels of magnetic resistance. Fortunately, I've had the bike for 5 years and know from experience that I almost always ride it at level 3, and I'm satisfied setting up IPBike for this resistance level only. Which means that I have only a single power curve that I want to model on IPBike like this:

Speed Avg Power
5mph 10.1W
6mph 12.4W
7mph 15.9W
8mph 21.9W
9mph 35.6W

It seems there are two approaches I could try: (1) tweaking the inputs to the IPBike virtual power calculation to try to replicate this curve, or (2) using the User Expression facility to set up my own calculations. I'm a programmer by occupation, so I'm sure I can come up with something, but right now not sure which direction to go in . . .

Any suggestions? Thanks!

Last edit: LisaB Mar 20 2021, 22:44

Posts: 5
Mar 21 2021, 17:54

I thought about it more and think I can probably fit a curve pretty closely to the data points with power as a function of speed. I don't think I've done curve fitting since college but I think excel has a way to find the coefficients pretty easily. Then I *think* I can just define a user defined variable named my_vp with this function . . .

Not sure which of the speed variables to use, instantaneous or a moving average. I'm new to looking at cycling data and have been surprised how wild and and jagged the raw sensor speed looks when I'm pedaling along at what feels like a steady pace.

Previous page: Forum