Left/Right pedal balance

[Alan Benstead]

I have installed the latest snapshot build (Feb 5th) and use the Balance Potencia activity chart to view L/R balance from my Assioma pedals. My left leg is stronger than the right due to an old injury. All the data looked correct until recently, now suddenly it shows my right leg as being stronger than my left, I believe this is after installing the latest snapshot build. Data in the Wahoo Elemnt app continues to show my left leg is the strongest. The values between the Elemnt app and GC look correct just swapped.

Regards Alan

[José Luis García Pallero]

El sáb, 6 feb 2021 a las 17:50, 'Alan Benstead' via
golden-cheetah-users (<golden-che...@googlegroups.com>)
escribió:

>
> I have installed the latest snapshot build (Feb 5th) and use the Balance Potencia activity chart to view L/R balance from my Assioma pedals. My left leg is stronger than the right due to an old injury. All the data looked correct until recently, now suddenly it shows my right leg as being stronger than my left, I believe this is after installing the latest snapshot build. Data in the Wahoo Elemnt app continues to show my left leg is the strongest. The values between the Elemnt app and GC look correct just swapped.

Take into account that a L/R balance of 50%/50% can not mean a perfect
balance, and a L/R of 52%/48% can not mean an unbalanced pedaling. The
L/R balance stored by the cyclocomputers if the net contribution of
the stroke during all the circle, so the net power is the sum of the
positive and the negative power (the negative is generated in the
recovery phase, almost all people generates negative power at this
phase). This fact is stored in the torque effectiveness metric. So an
unbalances stroke can also means a perfect 50%/50% balance in the
first phase of the stroke (from 0º to 180º), but a disbalance in the
power absortion in the recovery phase (from 180º to 360º)

Then, for a study of imbalance in the power generation the metrics GPR
(gross power released) and GPA (gross power absorbed) must be used
instead of solely the L/R balance value provided by the cyclocomputer.
This is better explained in
https://sparecycles.blog/2018/07/05/investigating-power-asymmetry/

The graphics Balance-Potencia and Balance-Potencia-Global show not
only the L/R balance as it appears in the file from the cyclocomputer,
but also the GPR and GPA, so you can see the amount of power you
generate with each leg, but also the amount of power yours legs
absorb. Please, download the last version of the charts from the
CloudDB

> Regards Alan
>
> --
> _______________________________________________
> Golden-Cheetah-Users mailing list
> golden-che...@googlegroups.com
> http://groups.google.com/group/golden-cheetah-users?hl=en
> ---
> You received this message because you are subscribed to the Google Groups "golden-cheetah-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to golden-cheetah-u...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/golden-cheetah-users/292a9ea1-b23a-4220-bda7-bcc647dc10aan%40googlegroups.com.



--
*****************************************
José Luis García Pallero
jgpa...@gmail.com
(o<
/ / \
V_/_
Use Debian GNU/Linux and enjoy!
*****************************************

[Alan Benstead]

You state that:

"The graphics Balance-Potencia and Balance-Potencia-Global show not

only the L/R balance as it appears in the file from the cyclocomputer, "

That was the case in my past rides, the Balance(L/R), Efectividad(L/R) and Fluidez(L/R) all match the data on my cycle computer. At the beginning of February I updated to the latest snapshot build, since that point the left and right data seem to be mixed up. The data for my right leg in the cycle computer matches the data shown for my left leg in Balance-Potencia and vice versa. I downloaded the latest version from CloudDB and it makes no difference.

[José Luis García Pallero]

For me works withoit problems. If you want, post yout FIT file here
and I'll load in my GC and I'll say you the numbers I can see

El sáb, 6 feb 2021 a las 19:49, 'Alan Benstead' via

> To view this discussion on the web visit https://groups.google.com/d/msgid/golden-cheetah-users/22809f86-30fd-408a-87b4-bb56c517dcccn%40googlegroups.com.

[Alan Benstead]

Here is a small .FIT file from my cool down ride this morning and a screenshot from the Elemnt app on the iPhone.  The Effectividad and Fluidez data seems correct when rounded, the Balance seems reversed to me.

[José Luis García Pallero]

El sáb, 6 feb 2021 a las 20:22, 'Alan Benstead' via

golden-cheetah-users (<golden-che...@googlegroups.com>)
escribió:
>
>
>

> Here is a small .FIT file from my cool down ride this morning and a screenshot from the Elemnt app on the iPhone. The Effectividad and Fluidez data seems correct when rounded, the Balance seems reversed to me.

Mmm, I can see also the reversed order in the L/R balance, but also
with the GC build of the February 1st.

The L/R balance is extracted as

np.asarray(GC.series(GC.SERIES_LRBALANCE))

so the version of GC doesn't matter at this point and I can assure
I've not changed the code at this point. It seems to me a problem with
the file generated by the Wahoo. Did you updated the firmware in the
last days?

>
> --
> _______________________________________________
> Golden-Cheetah-Users mailing list
> golden-che...@googlegroups.com
> http://groups.google.com/group/golden-cheetah-users?hl=en
> ---
> You received this message because you are subscribed to the Google Groups "golden-cheetah-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to golden-cheetah-u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/golden-cheetah-users/8360af02-0253-413b-94a1-c01ede89be81n%40googlegroups.com.

[Alan Benstead]

The last Wahoo update I did was early January, I have ridden many times since then with no issues. I always have something in the order of 5-10% split in my L/R balance with the left leg being higher. That was always the case up until a ride I did on Feb 2nd, since then it shows my right leg as being higher. I think I updated GC on Feb 2nd to the Feb 1st release. Today I updated it to the Feb 5th release but that made no difference. Something peculiar is going on.

[Alan Benstead]

Looking at the Raw Data in the Details tab, I am beginning to think the issue might be with the import of the data into GC

[José Luis García Pallero]

El sáb, 6 feb 2021 a las 21:16, 'Alan Benstead' via

golden-cheetah-users (<golden-che...@googlegroups.com>)
escribió:
>

> Looking at the Raw Data in the Details tab, I am beginning to think the issue might be with the import of the data into GC

But with my Garmin files all works ok. Probably the problem is in the
Wahoo side. In the Summary pane, in Metrics section, which values show
you for the Left/Right balance (%) field?

You can also reload files (delete the original and load again) prior
to the last Wahoo firmware update and check the results

> To view this discussion on the web visit https://groups.google.com/d/msgid/golden-cheetah-users/8e29e6ba-96bd-423d-8f52-c55bcb2dcbb1n%40googlegroups.com.

[Alan Benstead]

I added the Left/Right Balance (%) to the Summary pane and the values match what I see in Balance-Potencia. My rides this time of year are on Zwift, I record the power pedals with a Wahoo Bolt as Zwift cannot record Left/Right power. I then merge the two activities together to produce one activity file.

I just went back to a file I recorded January 20th with the Wahoo and imported that into GC. The data should match the activity that I already had for that day, but again the Left and Right are reversed. I have attached two screenshots of the Raw Data, one from the Wahoo file I just imported into GC and the original combined file from the ride on January 20th. You can see that Right and Left TE match but the Left/Right Balance is different. To me it seems that in GC pre Feb 1st the value was just imported into GC, now it seems to be 100 minus the imported value.

[Alan Benstead]

Not sure I included both screenshots, so here is the second

[José Luis García Pallero]

But the file from Wahoo and the combined one are different, so the
problem could be outside GC

El sáb, 6 feb 2021 a las 21:59, 'Alan Benstead' via

> To view this discussion on the web visit https://groups.google.com/d/msgid/golden-cheetah-users/02f53d37-79c3-49d8-bb0b-3f4190968eddn%40googlegroups.com.

[Alan Benstead]

It is the same file that was used to combine with the Zwift ride I did on January 20th. Why would it produce different raw data from the same file? I would like to go back to a pre Feb 1st snapshot build to test, but I do not see a way that i can do that.

[José Luis García Pallero]

El sáb, 6 feb 2021 a las 22:29, 'Alan Benstead' via

golden-cheetah-users (<golden-che...@googlegroups.com>)
escribió:
>

> It is the same file that was used to combine with the Zwift ride I did on January 20th. Why would it produce different raw data from the same file? I would like to go back to a pre Feb 1st snapshot build to test, but I do not see a way that i can do that.

You could use the build of June 2020:
https://github.com/GoldenCheetah/GoldenCheetah/releases/tag/v3.6-DEV2006

> To view this discussion on the web visit https://groups.google.com/d/msgid/golden-cheetah-users/94051bf7-83c4-4782-b154-d3bc1d329916n%40googlegroups.com.

[Alan Benstead]

I found a backup GC install file from January 6th. I have the old GC and the new snapshot executable installed. I take the SAME Wahoo file and import it into both versions of GC and they produce different Raw Data, the problem is obviously in the February snapshot builds.

[José Luis García Pallero]

I've installed the release 3.6 from June and I can confirm your
problem: your file is loaded with a 55%/45% L/R balance, but with the
latest snapshot (and with the previous one, February 3rd) the values
are 45%/55%. In my case there are no problems with fit files from
Garmin. Maybe you could open a bug report in github

El sáb, 6 feb 2021 a las 22:48, 'Alan Benstead' via

> To view this discussion on the web visit https://groups.google.com/d/msgid/golden-cheetah-users/b19c02d0-c976-44b4-94d2-ef7574256529n%40googlegroups.com.

[Alan Benstead]

I opened a Github account but i cannot see where to report a bug. Is there somewhere in this site to report bugs? Or do you think someone from the development team will look at this conversation?

[Alan Benstead]

Found a place on Github to report the bug

[Rui_B]

I've discovered the problem a different way... in 3.6 Dev2101, I designed a w/o for Train mode, and do the w/o on rollers using SRM Exakt pedals, capturing the data on SRM PC8, which I then upload to GC... GC also captures the data directly via Train mode... I usually compare the file before discarding one of them... This is where I noticed the difference with the balance on reverse side of the axis

:... this is the GC file:

This is the SRM file: (its capturing a bit more data before the w/o began.)

[Ale Martinez]

El miércoles, 24 de febrero de 2021 a la(s) 03:27:23 UTC-3, Rui_B escribió:

I've discovered the problem a different way... in 3.6 Dev2101, I designed a w/o for Train mode, and do the w/o on rollers using SRM Exakt pedals, capturing the data on SRM PC8, which I then upload to GC... GC also captures the data directly via Train mode... I usually compare the file before discarding one of them... This is where I noticed the difference with the balance on reverse side of the axis

Can you attach the fit files to see if this is the same case?

[Rui_B]

Hi Ale 

I've enclosed csv files from GC Train capture and SRM for the same w/o ( source was from GC on the day of the w/o)

Also I've enclosed the fit file from SRM pc8... ( sorry I've already deleted the GC Train captured file so I can't generate a .fit file from that data - I kept the SRM version)

[Rui_B]

hmmm. having problems uploading the files

I'll retry in a hr or so...

[Rui_B]

Ale, I keep receiving a posting error every time I try to send a file... is there another way I can send them to you?

[Ale Martinez]

El lunes, 1 de marzo de 2021 a la(s) 01:05:14 UTC-3, Rui_B escribió:

Ale, I keep receiving a posting error every time I try to send a file...

Zipped attachments usually work, the alternative is to upload the files to Google Drive, Dropbox or similar and post a public link.

[Rui_B]

ok... here it goes... fingers crossed (all three files)

[Alan Benstead]

Hi Ale,

I noticed with the latest snapshot build I am no longer getting Left/Right pedal balance data in the raw data.. As we discussed before, the issue is that Wahoo are not following the correct specification for transmitting the data. I sent a message to Wahoo after our previous discussion, but just received a response that they will pass my message onto the developers. Wahoo are not responsive like you guys! I have almost zero confidence that they will fix the issue. I would like to keep using the latest snapshot builds, is there a work around that I can implement so that I can keep getting the Left/Right pedal balance data? Or do I have to revert and stick with the previous snapshot build and continue to use my python fix to swap the left and right data?

[Ale Martinez]

El sábado, 20 de marzo de 2021 a la(s) 13:05:52 UTC-3, alan.b...@yahoo.com escribió:

Hi Ale,

I noticed with the latest snapshot build I am no longer getting Left/Right pedal balance data in the raw data..

Yes, I warned you this will happen in https://github.com/GoldenCheetah/GoldenCheetah/issues/3819#issuecomment-778857638

 

As we discussed before, the issue is that Wahoo are not following the correct specification for transmitting the data. I sent a message to Wahoo after our previous discussion, but just received a response that they will pass my message onto the developers. Wahoo are not responsive like you guys! I have almost zero confidence that they will fix the issue. I would like to keep using the latest snapshot builds, is there a work around that I can implement so that I can keep getting the Left/Right pedal balance data?

Yes, there is a simple workaround, but it requires to build from source.

 

Or do I have to revert and stick with the previous snapshot build and continue to use my python fix to swap the left and right data?

This would be simpler. 

[Alan Benstead]

I know you warned me it would happen. I have tried to get Wahoo to look at the issue.... but they do not seem interested, probably because it displays okay with their own app. It would be nice if they responded as quick as GC to issues, I raised the issue with them 5 weeks ago!

I will look into what I have to do to build from the source code. Once/if I get to the point where i can build, I will ask you which piece of the code I need to modify.

Thanks as always for your help

[Alan Benstead]

It would be nice if GC were to add an temporary option in a configuration file to allow users of a Wahoo device to continue to get L/R pedal balance until Wahoo eventually fix the problem. I understand that you want to adhere to the FIT file specification. I would assume a lot of GC users are now not getting L/R balance when using a Wahoo head unit or maybe I am one of the few that actually looks at that data.

[Ale Martinez]

El sábado, 20 de marzo de 2021 a la(s) 19:06:03 UTC-3, alan.b...@yahoo.com escribió:

It would be nice if GC were to add an temporary option in a configuration file to allow users of a Wahoo device to continue to get L/R pedal balance until Wahoo eventually fix the problem. I understand that you want to adhere to the FIT file specification. I would assume a lot of GC users are now not getting L/R balance when using a Wahoo head unit or maybe I am one of the few that actually looks at that data.

Wahoo users can import to GC v3.5 release in the "Wahoo way" without any special configuration: https://github.com/GoldenCheetah/GoldenCheetah/releases/tag/V3.5

GC v3.6 is still pre-release, so there is ample time for Wahoo to fix their FIT files to adhere to FIT file specification before release, and in the meantime both versions can be used in parallel and data synced or exported/imported from one to the other.

[Alan Benstead]

okay, I understand. I didn't realize you could run both versions in parallel and sync the data.

[marcen]

I am trying to tell the support of wahoo the problem.
Unfortunately I could not find a definition for the L/R balance. So if only the right leg is used, is R = 100% or R= 0%?Does anyone know where to find this?

@amtriathlon what program did you use to convert the FIT file, in the link https://github.com/GoldenCheetah/GoldenCheetah/issues/3819
I tried it with FitCSVTool from https://developer.garmin.com/fit/overview/ but unfortunately not found the error code 0x80 in the file.

Thanks for the help.

[Ale Martinez]

El domingo, 27 de junio de 2021 a la(s) 04:57:31 UTC-3, marcen escribió:

I am trying to tell the support of wahoo the problem.
Unfortunately I could not find a definition for the L/R balance. So if only the right leg is used, is R = 100% or R= 0%?Does anyone know where to find this?

@amtriathlon what program did you use to convert the FIT file, in the link https://github.com/GoldenCheetah/GoldenCheetah/issues/3819
I tried it with FitCSVTool from https://developer.garmin.com/fit/overview/ but unfortunately not found the error code 0x80 in the file.

You can find L/R definition in profile.xls file when you download FIT SDK