FIXED: Descend grade limit question

[Deb]

I ride veloreality just for fun and fitness and don't worry a lot about my detailed stats, but yesterday's ride was so far off from what I expected that I have to ask for clarification on how the descent grade limit setting works.

The user's guide says:

"Descent grade limit" does not let trainers that are capable of motoring to simulate grades below certain limit but the video still moves with the speed you would be moving if the grade was unclamped.

I have a computrainer, so it doesn't simulate motoring down grade, but I thought that my setting (-2% grade limit) would let me do some pedaling work on the downhill side but still have the effect described above, that is, the speed would be as though I were on the actual grade.

Last night I rode Grand Alps Part 5, Col de la Bonette which includes a long downhill of about 13.5 miles. On the descent, the on screen speed display never dropped below 30mph and ranged from low 30s to 40+, so I should have been able to easily finish the descent in less than half an hour.
It actually took me about 50 minutes and Strava says that my average speed on the descent was 16.7mph.

I wasn't riding terribly hard so I guess that time and speed are plausible for a -2% grade, but I thought that setting was just to give some resistance and the original video grade would still be in effect. I have noticed this on descents before, but this one was so long it became very obvious that there is a big difference between the virtual speed displayed and the distance covered.

Can you please clarify? Is this working as expected?

[admin]

First check if you have "Eliminate descends" checked in your grade reduction settings. This would cause what you've described. Other then that "Descend Grade Limit" limits minimal road grade being sent to trainer. How your trainer will interpret that particular -2% is entirely up to internal trainer logic. Your on screen speed does not depend on that clamped grade and the software always calculate your speed based on real road grade and input power (well some other parameters too but that's irrelevant).
Just to make sure I test rode the same descend and it took under 30 min with 100W input power. All as expected, and here is a link to Strava: https://www.strava.com/activities/1305557401

[Deb]

OK, I ran a few tests with some interesting results.

First I verified my "Eliminate Descends" checkbox. It was definitely not checked.
Next I created a 2 mile custom ride from a section of the Col de la Bonette descent.

For each test I tried to maintain a steady 100 watts. From your description, that should be the only thing that matters, yet I get very different results with test mode compared to a live ride.

Test #1:
I set ftp to 100 watts and ran the custom descent ride in Test mode. The results were as you would expect.
https://www.strava.com/activities/1305757153
moving time: 3:30
average speed: 34.2
average power: 100

Test #2:
Live ride of the same section. I double checked that eliminate descents is not checked.
https://www.strava.com/activities/1306296980
moving time: 6:12
average speed: 19.3
average power: 99

Test #3:
My normal display settings are different from the defaults which I used with the test demo, so I did another live test with default display options just to see if it affected anything, but it did not.

Before this one I selected the eliminate descents option and restarted v_ride. It was still checked. Then I unchecked eliminate descents and restarted again and it was still unchecked. Results were basically identical to the previous ride.

https://www.strava.com/activities/1306303644
moving time: 6:02
average speed: 19.9
average power: 102

Test #4
This time I selected the eliminate descents checkbox. I immediately noticed a difference in that the current speed displayed was reduced to the expected no-descent speed -- around 15mph. In all the other tests the current speed display showed 30+ mph during the ride.

https://www.strava.com/activities/1306308339
moving time: 7:59
average speed: 15
average power: 104

Thoughts?

[admin]

I just repeated my last test https://www.strava.com/activities/1305557401 of your original descend -2% descend grade limit but this time on a real trainer ( Lynx )trying to keep my legs in check

The results were as expected, I posted rides on Strava and Garmin Connect:

Strava - https://www.strava.com/activities/1306394913
Garmin connect - https://connect.garmin.com/modern/activity/2370328739

So I am not sure what causing it in your case, also I can not really check it on Computrainer as we no longer have access to it. So if you want to investigate any further contact me for live support as explained in https://veloreality.com/forum/viewtopic.php?f=2&t=94.

[admin]

I looked more closely at the source code for trainer drivers in our software and it seems that the one for Computrainer may cause something like you've described. But as I've already said I no longer have access to Computrainer and because of that will not touch that code unless I have somebody online to test it for me interactively. So if you are willing/want that offer of online support should fix the thingy.

[Deb]

Confirming that descending with computrainer is fixed in custom v_ride version provided during live support (thanks and always excellent support!!) BUT this version has a nasty issue in that if I stop pedaling I cannot get started again without shutting down v_ride and restarting the software once or twice.

If I stop, the gray "Start Pedaling", etc. bar appears at the top of the screen as usual, but when I start again the bar doesn't go away and the screen just keeps flashing. Restarting v_ride just once seemed to reset the ride to the beginning even when I clicked to continue, but after another try it was at the right place and I could resume.

I was out of time today to get back to you via Skype and I won't be around tomorrow. If you have any idea about this please let me know if you want to try another time to coordinate on testing.

[admin]

Catch me on skype at your convenience