cagnulein / qdomyos-zwift

Zwift bridge for smart treadmills and bike/cyclette
https://www.qzfitness.com/
GNU General Public License v3.0
418 stars 116 forks source link

Echelon Row-S Metrics in qdomyos match echelon app but not reality #448

Closed DoctorNode closed 2 years ago

DoctorNode commented 3 years ago

Hi there,

I love the app and thank you for your work to develop and improve. I have had an echelon row-s with integrated tablet for a little over a month and it's been great. However, echelons metrics don't match reality. I was hopeful that qdomyos would provide truer results but it seems it aims to match echelons metrics at this point. Since echelon seems reluctant to answer support I hope to take a more active role in finding a solution.

I have a background in engineering with some hobbyist programming and would like to help improve the accuracy of the readings, if possible.

The primary problem is that wattage and distance metrics are dissonant with one another and neither seem to correlate to real-world based on experience with other rowers. My aim is to have wattage reflect real-world effort so that other parameters can be calculated to better relate to other ergometers as well as real effort. I acknowledge that I do not understand the limitations that we are working with, with respect to machine sensors, etc, and I hope for others to provide their more knowledgable opinions.

Current implementation in qdomyos is to determine wattage from a table lookup using SPM and resistance, which parallels readings on the echelon rower. The evidence that this is wrong is that a full stroke and a pick drill with just arms reports the same wattage so long as resistance and SPM are the same. This demonstrates that it doesn't take into account stroke-length or stroke-time and doesn't represent reality.

Before I can start speculating and working towards a solution I need to better understand the context and constraints.

Primarily, what information is sent out from the machine?

From the code I've deduced the following: Output from machine:

Input to machine:

I will review the echelon rower manual to see if I can better understand what sensors the machine has.

Thank you for your time, assistance and perspectives.

cagnulein commented 3 years ago

Hi @DoctorNode thanks for your feedbacks! Yes the wattage metrics are coming from the echelon ones, so we can of course improve it! But we need a way to measure the wattage. Yes the metrics that we got from the rower are: Resistance Distance Speed Time

but we can double check in the debug log if there are more info that currently maybe i'm not using. Would you like to share the debug log from QZ so we can try to study again them?

DoctorNode commented 3 years ago

debug-Wed_Oct_6_18_21_20_2021.log

Thank you for your response @cagnulein, I have attached a log from today. I don't fully understand which lines indicate the information being sent from the machine.

(not of primary concern, but my wahoo tickr was intermittent during this session)

DoctorNode commented 3 years ago

I just reviewed the manual, and it looks like the only sense data that is being gotten is the SPM. So that's a bummer, looks like a limitation of the machine and there's nothing to be done without adding sensors to the machine.

Thanks for your time, but I think this can be closed.

cagnulein commented 3 years ago

@DoctorNode the frame with the metrics are these for example

<< f0 d1 11 03 4e 02 00 00 01 2b 00 27 00 01 ec 00 04 ed 00 48 9e"

bytes 0-1 are the header bytes 2-3-4 are the elasped bytes 11 is the cadence bytes 15-16-17 are the distance

so the others could be other metrics. If you want you can try to understand them and i will add them. So maybe there still some hope :)

DoctorNode commented 3 years ago

Thanks for the encouragement. I will process and review some of the other bytes and see what I can find.

Cheers!

stale[bot] commented 2 years ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.