How to reset data for CP chart in Activities tab

[Jordan Stanway]

Been training with GC and virtual power on a CycleOps Fluid 2 for several weeks, but recently got a power meter. Now I'm trying to figure out how to reset the filled background CP curve in the Activities tab to start over using only data from the power meter.

I'd prefer not to delete all the earlier workouts using virtual power from my history.

The power meter (P1 pedals) is measuring substantially lower power (1/3 to 1/2) vs. the model* for the Fluid 2, even though I've had the resistance knob on the trainer tightened to the torque limit.

* (I'm guessing the virtual power model is speed==>power. If so, where did the measurements come from to make that model? Being off by a factor of 2-3 seems like a lot to me, so it makes me wonder whether there's something wrong with my trainer or my power meter.)

[Ale Martinez]

You can configure the chart to start from a certain date using date ranges or exclude certain activities using filters, see the wiki

[Nigel Van de Velde]

If you made sure your setup (tyre pressure, tyre clamping, ...) was similar for all your virtual power rides, you might check the power-speed curve for your trainer setup with your powermeter.

Afterwards you might edit the virtual power with the corresponding actual power.

Some excel and the data editor in goldencheetah is all you need.

Kind regards,

Nigel

[Jordan Stanway]

I can get to the "Chart Settings" menu now and I've figured out how to select the proper date range. The little pop up (labeled "More..." in the upper left of the chart) wasn't showing before, I think because of the color scheme I was using. Actually ran across it as I was figuring out how to add a few fields to my training view.

Thanks for the suggestion.

For anyone else who finds this post and was having trouble locating the pertinent section of the wiki, here's what worked for me:

https://github.com/GoldenCheetah/GoldenCheetah/wiki/UG_ChartTypes_Trends#critical-mean-maximal

[Jordan Stanway]

That's a good suggestion, and I'll try to do a batch adjustment after a few more rides when I have enough data to do a regression for the power-speed curve. Honestly I'm a bit surprised that the model GC ships with would be so far off.

Searching the source on github, it seems the model for the Fluid2 trainer is:

    case 3 : // Cyclops Fluid 2
        {
        double mph = rtData.getSpeed() * MILES_PER_KM;
        // using the algorithm from:
        // http://thebikegeek.blogspot.com/2009/12/while-we-wait-for-better-and-better.html
        rtData.setWatts((0.0115*(mph*mph*mph)) - ((0.0137)*(mph*mph)) + ((8.9788)*(mph)));
        }

[Jan de Visser]

I have used a Fluid2 trainer for about 6 years and a powermeter for the last 5
of those. My experience (anectodal, so buyer beware :-) is that the first 15
minutes or so of a session are basically throwaway; while the trainer warms up
the power needed to reach a certain speed increases. When the thing is warm,
and for me that seems to be at about 15 minutes, it becomes fairly stable.

I don't know how accurate the model is though. I suspect it may depend on the
individual trainer and maybe even conditions.

I don't trust power curves. I would just use your powermeter and build a
history of real data, and just keep the old, speed only, data for
entertainment value.

[Jan de Visser]

I have used a Fluid2 trainer for about 6 years and a powermeter for the last 5
of those. My experience (anectodal, so buyer beware :-) is that the first 15
minutes or so of a session are basically throwaway; while the trainer warms up
the power needed to reach a certain speed increases. When the thing is warm,
and for me that seems to be at about 15 minutes, it becomes fairly stable.

I don't know how accurate the model is though. I suspect it may depend on the
individual trainer and maybe even conditions.

I don't trust power curves. I would just use your powermeter and build a
history of real data, and just keep the old, speed only, data for
entertainment value.


On Monday, January 16, 2017 10:01:40 PM EST Jordan Stanway wrote: