Can ORM be used to correct low wattage on FTMS water rower? #170
-
|
Hey team! I have an FTMS water rower, and looking into @laberning's Sportstech wrx700, it seems the monitor is basically the same (Attached pictures). Although the model does provide connectivity with all Applications (EXR, Kinomap, ErgZone, etc.), the watts readings seem too low/inaccurate, which impacts metrics like split times, calories, FTP calculations, and so on. To fix this problem, I initially looked to the QZ app (https://github.com/cagnulein/qdomyos-zwift) to connect my rower to an old Android, add a watts multiplier, and call it a day. However, that is still a work in progress, as the app does not fully recognize my rower. Roberto has been helping me a lot with this. In the meantime, I wanted to settle for a more permanent/rowing-oriented solution. I have a Raspberry Model 3B lying around (I can check the specs to see if there are limitations to this model, and don't mind buying a more powerful one if needed). Still, my coding skills went as far as creating a PiHole and an Open Media Vault server for fun (all via tutorials, so I have near zero experience). Maybe this could be an issue if I need to add a 2x multiplier to the watts output. Do you happen to know if using ORM could help me with the watts' inaccuracy? Does it require much coding experience, or do you believe I could give it a go without breaking the rower? I have an 11" external monitor that I used with the Raspberry in the past. Would it work to output the information screen (as seen in the screenshot from the readme file)?
|
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 30 replies
-
It should work on a Pi3B without issues, especially as waterrowers do not produce much data.
If you can follow a tutorial, you can install ORM. See https://github.com/JaapvanEkris/openrowingmonitor/blob/main/docs/installation.md for the latest version, so you can decide if it is doable. Only "coding" needed is to modify the settings in the config file. Please note that the version found at https://github.com/JaapvanEkris/openrowingmonitor/tree/main is much more robust and a lot easier to configure (and it has a lot more features). So I'd advice to start with that version.
Nah, you can do a lot better. ORM can calculate your power in a decent way, once provided with the inertia of the moving water. That is a setting, and you can set it by trial and error. As a bonus your distance and pace will be correct as well, allowing the use of EXR, Garmin, etc.
Yep. These monitors on these machines are a total mess from a metrics perspective. The indicated watts are not a fair representation in any way. Doubling these numbers might reduce the issue, but they are bad estimates to begin with.
Not really. When you can follow the installation manual, your basically done. You have to change some setting, but we even have a manual for that (see https://github.com/JaapvanEkris/openrowingmonitor/blob/main/docs/rower_settings.md). And might these manuals fail you, we are here to help.
Yeah, here it becomes more technical. Breaking the rower: No. Modifying it: probably. Depending on how the original monitor measured progress, it might be a bit more work. Key is that you want to measure the speed of the impellor, and not the handle or seat. If you look at other discussions, you'll see that it might be more or less work.
Yeah. No problem. With the new version you can even display three rows of data. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you @JaapvanEkris! I decided to give it a go - just installed everything following steps from https://github.com/JaapvanEkris/openrowingmonitor/blob/main/docs/installation.md and it seems I can already access it from my notebook (it's not connected to the rower yet). Can you confirm that these logs are working as they should so I can proceed with connecting the GPIO pins to the sensors? 1 - sudo systemctl status openrowingmonitor 2 - sudo journalctl -u openrowingmonitor 3 - sudo systemctl status webbrowserkiosk 
|
Beta Was this translation helpful? Give feedback.
-
|
It seems to look for the ANT+ stick, so disabling that in the settings should resolve that (please note, do this both for the ANT+ mode and the heartrate monitor). But it looks OK to me. The xinit errors are odd, but the GUI has started so it probably is not critical :) |
Beta Was this translation helpful? Give feedback.
-
|
Thank you for all the help so far @JaapvanEkris! So now I've started with the hardware part: I've attached photos of my sensor and what I believe are the holes in which the magnets are placed. I've disconnected the sensor cables from the rower monitor and attached them to the Raspberry GPIO on pins 17 and GND as per the installation docs. So far, I've been rowing for a while and have not seen any data pop up when accessing the web page from my phone. Can you guys help me out if I'm missing something? Sensor on top of the wheel Possibly the Magnet? Cables connected to the sensor Cables connected to the Raspberry (I've tried shifting them too, but nothing happened) |
Beta Was this translation helpful? Give feedback.
-
This is good news as this most probably means that the sensor detects the impeller shaft rotation (unless its not and the stroke is detected via the seat and the sensor only cares about whether the handle is moving or not). This is a cylindrical reed switch, the magnet can be either on disk I marked facing outwards, or Or it is possible that its on the bottom. The picture you sent does not seem to be a magnet. rather a screw. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the feedback @Abasz! And I just discovered that you are right, the rowing monitor is powered by two AA batteries and turns off to save power if there is no readings captured by the sensor. To do a test with minimal damage I moved only the cylindrical reed switch to see if it would wake up the monitor and it did! The impeller rotated as well and both the handle and the seat were not moved so I know there’s no signal coming from them. I believe there’s probably only one magnet as it needed to almost fully rotate to wake up. Do you know what are the next steps I should take to have the raspberry pi read the sensor? rower_video.mp4 |
Beta Was this translation helpful? Give feedback.
-
|
You need to configure ORM in the config file. Please refer to the docs: https://github.com/JaapvanEkris/openrowingmonitor/blob/main/docs/installation.md#rower-settings You can start with another water rower profile and change that to your needs: https://github.com/JaapvanEkris/openrowingmonitor/blob/main/docs/Supported_Rowers.md there is a comprehensive guide to the settings options available: |
Beta Was this translation helpful? Give feedback.
-
|
I wanted to test with the same settings from the Sportstech wrx700 (hardware looks really similar to mine) but in the guide I could only follow until the gpio connection. Could you help me through the commands to select this profile in the config file? |
Beta Was this translation helpful? Give feedback.
-
|
You need to edit the config.js file in the config folder. But this is explained in the settings guide: https://github.com/JaapvanEkris/openrowingmonitor/blob/main/docs/rower_settings.md#why-we-need-rower-specific-settings Please read through the guide, it should answer 95% of your questions and if you have specific questions or you get stuck we are here to help. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks @Abasz! So, I just finished editing the config/config.js file using the rower setup as an example. Everything seems to be set from the configuration side (the file below is to confirm). But for some reason, the rower still doesn't recognize the sensor data. I could see once the time and total distance being recorded but when I tried again, it didn't move from 0. Is there anything else I might be missing? |
Beta Was this translation helpful? Give feedback.
-
|
I've been doing a few changes and the graph from the raw file got slightly better! (even though it's indeed few datapoints) But for some reason ORM is still missing the next drive section and thus throws everything together as one big stroke by the end of the workout - so the impulses are sorted out but I can't for the life of me figure out how to make the algorithm capture the individual strokes. I just took a screen shot of how it looks like when rowing, anything from a 1min to 5min rowing workout is always packed into one big stroke, is there anything I'm missing @Abasz?
|
Beta Was this translation helpful? Give feedback.
-
|
Please copy your current full settings here |
Beta Was this translation helpful? Give feedback.
-
|
This is the config file - some of the changes were based on our exchanges and others I've been testing for a better curve - but nothing so far allowed ORM to distinguish between the strokes: |
Beta Was this translation helpful? Give feedback.
-
|
Please move to V1Beta_Architecture_Revision branch (use 'use strict'
/*
Open Rowing Monitor, https://github.com/laberning/openrowingmonitor
You can modify this file to configure Open Rowing Monitor to your needs.
This file should be placed in the 'config' folder of Open Rowing Monitor.
All available configuration parameters are visible in config/config.default.js
To modify a parameter, copy it to this file and modify the value.
Changes to this file are persisted when you update to new versions.
*/
// eslint-disable-next-line no-unused-vars
import rowerProfiles from './rowerProfiles.js'
export default {
loglevel: {
default: 'debug',
RowingEngine: 'debug'
},
createRowingDataFiles: true,
rowerSettings: Object.assign({
...rowerProfiles.Sportstech_WRX700,
numOfImpulsesPerRevolution: 1, // only on magnet on my rower
minimumTimeBetweenImpulses: rowerProfiles.Sportstech_WRX700.minimumTimeBetweenImpulses * 2, // setting these liberaly to get more datapoints
maximumTimeBetweenImpulses: rowerProfiles.Sportstech_WRX700.maximumTimeBetweenImpulses * 2, // setting these liberaly to get more datapoints
minimumForceBeforeStroke: 0, // This is the minimum force that has to be on the handle before ORM considers it a stroke
minimumRecoverySlope: 0, // The minimal inclination of the currentDt's before it is considered a recovery. When set to 0, it will look for a pure increase/decrease
autoAdjustDragFactor: true, // When your machine's power and speed readings are too volatile it is wise to turn it off
autoAdjustRecoverySlope: false, // For this to work, autoAdjustDragFactor MUST be set to true
flywheelInertia: 0.75 // The moment of inertia of the flywheel kg*m^2
}),
shutdownCommand: 'halt'
}With this raw file: 2024-11-02_16-04-02_raw.csv.gz I get 7 strokes. Your inertia is obviously wrong but I cannot enter that instead of you, try rowing and see what fits. So there is plenty of room for improvement I spent literally 10 minutes with it only. |
Beta Was this translation helpful? Give feedback.
-
|
After some rowing and testing, I think I found a setup I'm good with! @Abasz, thank you very much for all the instructions and the patience; sometimes it takes a while, but I think I finally figured out how most things work. ORM is a fantastic product that improves the rowing experience a lot, so I just wanted to thank everyone involved! Here is how the initial setup looks. I'm working on a cleaner/permanent setup for the future and studying the possibility of adding another magnet to the rower to improve readings. I'm not entirely sure how to do it yet as there aren't many sources, but I'm very happy with how it looks so far!
|
Beta Was this translation helpful? Give feedback.
-
|
Thanks, once you are done we are happy to receive the config for your rower with a specific raw delta time file so others can benefit from the already worked out settings. on adding a magnet: it is not that complicated, but it depends on the tools you have. Most straight foward way to do is to 3d print a bracket with space for 2-3 magnets and mount it on the part of the shaft where its now. alternatively you can drill a small hole and insert the magnet there (but the later may safer from improper magnet placement that can cause measurement errors. |
Beta Was this translation helpful? Give feedback.
-
You are welcome. Hope to see you on the virtual EXR waters soon! |
Beta Was this translation helpful? Give feedback.
Please move to V1Beta_Architecture_Revision branch (use
updateopenrowingmonitor.sh -b V1Beta-Architecture_Revisionand use these settings: