Azuritber chassis modification

mbatorowicz

Member
Hello, if I want to change the mower software from azurite to azuritber, do I have to rebuild the chassis so that the drive wheels are at the rear?
I would like to add that I wanted to change the software to azuritber to use IMU and mowing lane by lane.
 
Hello, if I want to change the mower software from azurite to azuritber, do I have to rebuild the chassis so that the drive wheels are at the rear?
I would like to add that I wanted to change the software to azuritber to use IMU and mowing lane by lane.

No Sir,
there is nothing to rebuild nor to change.
You can use the software with a rear wheeler and front wheeler chassis.
UWE

PS. Good decision!
 
I have GY-521 and GY9250 modules. The GY9250 may have problems. It is appropriate to use the GY-521. Is this one module enough or is one more needed? Which firmware to choose?

azuritber.png
 
The GY-273 compass still needs to be connected - is it optional? Currently, it is very difficult to get the original HMC5883L. Does the software work with QMC5883L?
 
Compass is optional and don't need to be connected.
Only useful to have the lane always oriented in the same "true North" direction .
Without compass the lane is correctly oriented from "true station" (heading 0 deg) but after 10 or 20 minutes of mowing the gyro/accel drift change the true "station" reference heading But it's not a problem.

You can select between QMC and HMC into mower.h line 109.
Select QMC if you have GY-273, but again compass is not really useful and not accurate.
 
Patern is LANE strange behavior.
I downloaded the AzuritBer master mower. I installed the GY-521, and calibrated it in the pfod app. The mower behaves strangely as in the video. It is similar also in RAND mode.

 
Patern is LANE strange behavior.
I downloaded the AzuritBer master mower. I installed the GY-521, and calibrated it in the pfod app. The mower behaves strangely as in the video. It is similar also in RAND mode.

switch off your sonar and bumper for troubleshooting purpose.
Check the perimeter setting. is on coil selected? Is the perimeter polarity correct?
Uwe
 
Everything was running on Azurit Firmware today. Today I decided to switch to AzuritBer. I turned off the sonar. The Perimeter Wire is OK as the mower returns nicely to the Charging Station.
 
Everything was running on Azurit Firmware today. Today I decided to switch to AzuritBer. I turned off the sonar. The Perimeter Wire is OK as the mower returns nicely to the Charging Station.
If you have trouble to understand all behaviour, you can test the last version locate in branch Test-for-Display.

This is the last version with a new Console in the Arduremote main menu under Test ODO
Screenshot_20210627-100253_ArduRemote.jpg

You can see here all what the mower do and why it stop and reverse etc.... ,It's a real time screen.
Screenshot_20210627-100321_ArduRemote.jpg

Do not hesitate to share a video so i can tell you if setting is OK.
On your first video there is some pause on the movement (Possible SpeedOdoMinimum is not enought).
 
I use the fpod app because Arduremote probably only works via BT and currently I have ESP-015 connected with esp-link firmware. Logs are unreadable to me. From the charts it seems to me that the problem is in the perimeter. I don't know what the recommended values should be. I changed the wire connectors. The circuit has a resistance of 12 Ohm (10 Ohm resistor + 2 Ohm wire). The cable is connected in several places and insulated. Can poor insulation quality affect the signal. My mower housing is made of dibond. The bumper made of Dibond disrupted the operation of the receiver, so I replaced it with one made of HDPE. I just noticed the Statuses displayed in the application: POUTREV, POUTSTOP even though the mower is in the middle of the perimeter.
208312779_4195896373810428_965221951246457438_n.jpg206691633_345065640353222_1570374721482145736_n.jpg
207563408_499736187947547_7148081703540779042_n.jpgScreenshot_2021-06-27-13-09-59-192_au.com.forward.pfodApp.jpg
Screenshot_2021-06-27-13-10-13-963_au.com.forward.pfodApp.jpg
 
Logs are unreadable to me.
What do you mean ?
Have you tested the last version with console ? The log are now replace by a readable text console (into github branch with screen)

YES issue is in the perimeter noise these four peak are not OK and mower stop/reverse etc in the area without finding the wire.

Capture d’écran 2021-06-27 140643.png

Try the same test but with mowing disabled into setting now (force mowing off)
 
Previous images were taken with the mower very close to the wire. Switching off the engine did not change anything.
I took the present images when the mower was 6-7 meters from the perimeter, in the center of the area. Are these values too low? I think I need to build a new PERIMETER SENDER to test. Maybe the MC33926 driver has overheated. Can you somehow check if the perimeter sender is working?
Screenshot_2021-06-27-19-56-52-937_au.com.forward.pfodApp.jpgScreenshot_2021-06-27-19-59-36-950_au.com.forward.pfodApp.jpg
 
At 7 meters of the wire value is normally near -250 / -300
It's difficult to test the sender .What is the total length of the wire
Try to test with the receiver at 20 CM in front of mower to limit motor noise.
Do you have the same result with AZURIT ?
Maybe test other receiver coil.

Here ESP32 sender link:
as Bernhard said…….the value is woobling around -250 and -280 at 7m distance of the wire.
I can not see the installation spot of your receiver in your video of your rover. It looks like the receiver is underneath the bumper or chassis?
My second ( new) rover have also a dibond chassis like yours. The first installation of my perimeter wire was very very close to the chassis and I had a very weak wire value. With 10cm in front of the chassis it is much better. Just an idea.
 
The perimeter wire problem has been resolved. I don't really know how but it works now. I connected a new wire for testing and the smag and mag values were correct. After plugging in the old cable, it also worked - strange thing. I suspect that the temperature in the housing with the electronics was too high because I carried out these tests with the housing open. In addition, I reduced the voltage on the DC / DC converter from 12 volt to 6 volt and it's OK.
There were other problems that can be seen in the video.
- Often the mower does the IMU calibration.
- The track lines are not parallel to each other.
- How to reduce the time gaps between turning and driving?
Yesterday I also connected the RPi so that I could control the mower with home assistant because I use it at home.
- How much will RPi reduce the mowing time (power consumption), has someone checked it?
- Is there a description of the MQTT topics that I can control the mower?
 
I suspect that the temperature in the housing with the electronics was too high

I am not sure i use a DTH22 in my mower and stop everything at 70 Degres without perimeter issue.

Often the mower does the IMU calibration.

It's into setting IMU Delay between 2 calib


The track lines are not parallel to each other.

You need to adjust the opposite value for each 3 main lane
See setting By Lane
For long lane >8 meters it's difficult to have a perfect result.

- How to reduce the time gaps between turning and driving?

It's into motor setting : increase Speed ODO Minimum


How much will RPi reduce the mowing time (power consumption), has someone checked it?

I never see a difference in mowing duration with or without Pi
 
I need to change the high grass management into Bylane mowing to avoid spirale mowing.
I need to copy what i did in the Robomow firmware:
Mower stop roll 180 deg and mow with only half distance between lane.


What append at 5'57 ?

for mqtt and home assistant i need to take time to add this in the WIKI but:
you have all the file into Github under test file / mqtt / yaml
Copy the ardumower.yaml into a package folder.

For MQTT topic see :
 
Oben