Unable to connect Kurt Kinetic smart control

[Anil Soni]

I have KInetic smart control trainer and Garmin Ant+ mini usb stick . I cannot connect my trainer with golden cheetah it says unable to open ant+ device . Using on Window 10 

Please help i have tried everything n=but nothing has worked till now

[Ale Martinez (Please don't email or cc me)]

El miércoles, 30 de septiembre de 2020, 13:30:53 (UTC-3), Anil Soni escribió:

I have KInetic smart control trainer and Garmin Ant+ mini usb stick . I cannot connect my trainer with golden cheetah it says unable to open ant+ device . Using on Window 10 

Please help i have tried everything n=but nothing has worked till now

According to https://www.dcrainmaker.com/2016/09/kinetic-smart-control-trainers.html when these trainer were introduced, they don’t support standard ANT+/FE-C or BTLE protocols but their own proprietary version, and GC has no special support for them. 

[Anil Soni]

Okk So there is no way to connect KInetic trainer 

Thanks for help

[Eric Christoffersen]

Nice dicussion here:

https://www.dcrainmaker.com/2016/09/kinetic-smart-control-trainers.html 

I found the inride2 pod control here on github:

https://github.com/kinetic-fit/sensors-swift-kinetic

But I'm not sure if their smart control trainer protocol is public. If someone could get access to it then support might get added to gc...

[Eric Christoffersen]

OK, you might be in luck.

Yesterday my neighbor says she wants to do indoor training. It made me cringe when she talked about a subscription and I figured it might be nice for her to be able to use my old inride with golden cheetah. I had a go today hooking up the kurt github sources to gc. Kurt code is really nicely organized and easy to use so virtual ride using inride is now working for me on a windows 10 laptop, reporting power, cadence and speed.

The kurt sources above also include the protocols for communicating and controlling the kurt smart_control devices, so I hooked that up at the same time, but I can't test it because I don't have that device.

I haven't yet done any of the calibration stuff. Once I get it working I'll submit a change that will support the inride and try to support smart_control also.

[Anil Soni]

Nice to hear from you 

Thanks Now I will be waiting for next update

[Eric Christoffersen]

Well... calibration took me longer than it should have but here is a pull request:

https://github.com/GoldenCheetah/GoldenCheetah/pull/3636

After 3 hours of failing to calibrate with my knobby tired mtb I bucked up and got my road bicycle from the garage. Calibration actually works fine. I rigged up some notifcations on inride path about calibration state but they're pretty awful so hopefully someone will step up and do ui magic. I think best experience for anyone who wants to use a kurt device they should calibrate it using their phone.

As is: if you toggle calibration mode and pedal up to 35kph and let it coast the calibration will stick, but it won't tell you much.

I coded the kurt_smart control support in the same way as the inride. I think it will work but really it needs someone to test it.

The inride protocol requires the device system id. Normally that is acquired from device address but that isn't available on mac.

Mac: Bluetooth on mac doesn't have a device address, or doesn't give apps access to it... or something like that. So there is a convention that on mac people use the deviceUUID instead. The kurt inride protocol requires the actual systemid which is defined by the address.. Someone should test it and see if the current mac uuid code works... I write code to query the systemid from the device characteristics but its slow. Maybe mac user can get it going.

Lastly. OMG bluetooth is a pain. There are neighbors with ipads, kids headphones, phones, laptops, etc, etc. Unbelievably how many devices come and go while I was debugging. I got a bunch of av in the qt bluetooth code while it was sniffing ipads. So... turn off as many bluetooth devices as you can, and use an extension to bring the bluetooth as close as you can to the inride pod.

To connect with a windows 10 pc that has bluetooth:

-turn ON the laptops bluetooth

-plug in your usb ble device on a long usb cable and put it right by the inride pod. This massively helps the signal.

-do not connect your inride pod to windows, leave it not connected

-create a new training device: select bluetooth 4.0, you know the one that already mentions kk inride...

-leave everything 'autoconnect on startup', click next

-click next to skip virtual power

-give it a name and click finish

Now to use it: 

-select your new training device

-select a workout

-click the power button, it will turn yellow

-now... pedal... keep pedalling slowly until you start seeing watt readings that aren't zero. congrats! I found htis takes between 0 and 20 seconds. Sometimes it doesn't connect then toggle the power button and try again. It always works the second time. :)

-click play button to start your workout

-click the wrench thing to start calibration mode

-pedal up to 35kph, then stop pedalling and coast to a stop

-click the wrench again to leave calibration mode

If your roller was not too tight and not too loose, and you did get to 35kph, your pod is now calibrated.

[Anil Soni]

I will connect today night and test kinetic smart control but i am little confused how to use github . If you can guide download from GitHub it will be great

[Ale Martinez (Please don't email or cc me)]

El lunes, 19 de octubre de 2020, 8:30:17 (UTC-3), Anil Soni escribió:

I will connect today night and test kinetic smart control but i am little confused how to use github . If you can guide download from GitHub it will be great

For each Pull Request on GitHub, like Eric's one, an automatic build is generated on AppVeyor, the corresponding installer for Windows can be downloaded from Artifacts tab following the AppVeyor link on "Checks" section, in this case: https://ci.appveyor.com/project/Joern-R/goldencheetah-knhd8/builds/35809525/artifacts

The main limitation of PR builds is they don't include the keys to access online services s.t. CloudDB or Strava, but this shouldn't impede to test this new feature.

[Anil Soni]

Thanks a lot u r awesome, It was working , I connected through bluetooth  and tested for few minute i will connect again tomorrow and test for few Km Ride

Thanks again for adding support for kinetic 

[Eric Christoffersen]

It is a miracle!

A few things that would be good for you to try, since you're the only test:

Your device has multiple modes: simulation, erg, intensity and level. I don't know how to reach the last two from gc but could you try to use the first two?

- simulation mode: choose a punchy previous activity, lots of steep varied hills. Smooth the altitude of the activity (using fix gps), then export it as json, then load that json as a workout. That will run your trainer in slope mode and show your progress on a map. You should feel the trainer adjust its resistance as slope increases and decreases. Does it work? Note that slope should match the smoothed file which is not necessarily reality. You can skip forward/back 1km with the ffwd/rev buttons.

- erg mode: use any old ergfile workout, are watts being fixed by trainer?

Try a virtual ride and let us know how it goes. I really like this umbrail pass ride form alucard:

http://rlv-alucard.de/ch_umbrailpass.htm

Eric

[Eric Christoffersen]

I updated the change to support calibration on the smart_control units, can you give calibration a try?

Inride calibration feels really solid. Inride needs more feedback because you must get the roller tightness correct before calibration can succeed.

win10 x64 artifact with latest pull is here:

https://ci.appveyor.com/project/Joern-R/goldencheetah-knhd8/builds/35843420/artifacts

Man its great to have these rolling builds.

[Anil Soni]

Sure i will try tomorrow morning and update you

[Anil Soni]

I done training with *.fit file i had from my igpsort it was able to control kinetic smart control . I have attached screenshot for reference . I forgot for calibration i will update that soon 

[Eric Christoffersen]

Besides calibration, can you please also try an erg file? ErgFile should put trainer into erg mode and will test that your trainer will correctly hold fixed power.

Here is text for a short ergfile to use as a trainer test, save in file as *.erg, load as workout. Should give you quick definite power steps every 15 seconds for first minute, then a ramp from 1 minute to 1:30.

[COURSE HEADER]

FTP=300

SOURCE=HomeMade

VERSION=2.0

UNITS=ENGLISH

DESCRIPTION=

FILE NAME=HomeMadeSmartTest

MINUTES  WATTS

[END COURSE HEADER]

[COURSE DATA]

0 100

0.25 100

0.25 300

0.5 300

0.5 100

0.75 100

0.75 300

1.0 300

1.0 100

1.5 300

1.5 100

2.0 100

[END COURSE DATA]