recommended trainer profile for rollers w/mag unit?

[flahutewannabe]

I'm using a SportsCrafters aluminum rollers with their previously-offered mag resistance unit. I believe the unit is identical to the CycleOps model. What trainer should I use, or if there is not one to match this setup how would I go about using a PP indoors with this setup?

Also, I'm thinking the resistance curve may change based on whether the rollers are used in "native" mode or with a fork stand?

Thanks for the help. A post-SuperBowl response is quite acceptable

[flahutewannabe]

Ok, did some digging myself since now the Superbowl is past:)

Here is a link to the Sportscrafters power curve for their rollers w/mag unit at setting 5 https://www.sportcrafters.com/blog/virt ... rs-rollers and the curve itself.

Is there a trainer setting in Isaac that matches this curve (green linear)? Thanks.

[flahutewannabe]

Still trying to figure this out. In Isaac there are two profiles that appear similar to the green line above (straight line vs a curve) but with different slopes. They are Cycleops mag (low) and Cycleops: mag (high). [ Sorry I don't know how to grab the graphic for each to post in this thread.]

How can I adjust one of them to best match the green line above, or is there a different trainer profile to use?

Thanks for the help.

[klhutch52@gmail.com]

I hate to see this question go unanswered for over a year so here is what I would do. I would manually "digitize" the Sportscrafters graphs. For example, measure the distance on your computer screen or a paper copy of the graph above between 10 mph and 40 mph on the bottom scale. I got 146.39 mm. From this you can compute the horizontal scale factor:

SF = (Speed2 - Speed1) / (distance) = (40 -10)/146.39 = 0.205 mph/mm

Now for each of the three curves on the graph measure the distance from the left axis to the point where each curve crosses each power line. Actually for the two linear curves, ZRO-100 Mag 0 and ZRO-100 Mag 5 it is only necessary to measure two points on each curve. For the ZRO-110 curve you can measure the distance to each of the 7 power line crossings. Using the scale factor calculated above multiply each distance by the scale factor and then add the value at the left side of the graph (10 mph in this case) to get a table of power versus mph for each curve. Put these numbers into an Excel spreadsheet and graph them. Add a trendline to each plot and enable the "Display Equation On Chart" checkbox. Set the two trendlines for the linear curves to a linear trendline. Set the trendline for the ZRO-110 curve to Polynomial and set the order to 3.

The trendline equations on the charts now give you the coefficients you want. You may have to format the equation display by right clicking on it to get the right number of decimal places. Here's what I got from the curves above:

ZRO-100 Mag 0 : A = -14.49 B = 7.86
ZRO-100 Mag 5 : A = -70.23 B = 15.64
ZRO-110 : A = -25.32 B = 15.71 C = -0.6997 D = 0.02306

In each case the A coefficient is the constant term, the B coefficient is the linear term, the C coefficient is the squared term, and the D coefficient is the cubed term in the Excel trendline equations. For the two linear curves C and D are zero of course. These results are reasonably close but the limited numeric precision that Isaac supports for the coefficients does limit the accuracy especially for the last line with the cubic equation.

This post would be a lot more instructive with pictures but you have to upload the pictures you want to use on this forum to a photo hosting service and the post links to them. I used to use Photobucket for that but they went to a pay to play scheme and they want way too much money. I have never bothered to look for an alternative.

[flahutewannabe]

Wow, talk about dead-thread revival. Thanks so much @klhutch52. I really appreciate the detailed response. While I think I follow the logic, this is pretty far beyond what I'm willing to tackle given my pay grade/capabilities. I was hoping Velocomp might have an easier approach to recommending a "pretty close" trainer from those in the database.

But I do appreciate your thoughtfulness in finding the thread and going into detail here.

[Velocomp]

Wow, talk about dead-thread revival. Thanks so much @klhutch52. I really appreciate the detailed response. While I think I follow the logic, this is pretty far beyond what I'm willing to tackle given my pay grade/capabilities. I was hoping Velocomp might have an easier approach to recommending a "pretty close" trainer from those in the database.

But I do appreciate your thoughtfulness in finding the thread and going into detail here.

If you have a speed/power curve for your trainer, you can compare its curve to curves presented in Isaac. Just pick a trainer whose curve is close to yours.