Bitcraze flow breakout board with PX4

Discussions about all things Bitcraze
Post Reply
krayola
Beginner
Posts: 3
Joined: Tue Sep 08, 2020 11:34 am

Bitcraze flow breakout board with PX4

Post by krayola »

Hello,
I'm trying to launch the your flow breakout board on the pixhawk4 system, There a few tutorials on how to physically connect the flow board tot he PX4, but basically nothing for troubleshooting, I'm not able to get the drone to hold the position, using this bitcraze flow breakout board.

I've been able to start up both of the necessary sensors, but the drone still doesn't seem to be holding the position, I've toggled all of the necessary parameters via PX4 that should result in indoors position hold, but nothing happens, and when inspecting the logfiles from the drone, seems like the drone is detecting correct movements with IMU sensor, but the output from the flow breakout sensors seems to be pretty much a flat line, on the graph, and if I enhance the scale of the optical sensor output by about 100x, the graph seems more like it should.

So I do know for sure that the distance sensor on the flow breakout board is functional and starting, the output from the distance sensor is correct, which could be messing up the calculations when not outputting the correct distance, but not in this case.

The quality output from the flow breakout also seems a bit jumpy, but still in the correct zone, I've also tried powering the flow breakout board from a separate power source, not the PX4(due to an article on this subject viewtopic.php?f=2&t=4345&p=19832&hilit=pixhawk#p19832 )

Not sure what else I've haven't tried already, perhaps someone can point me in the correct way, the PX4 community isn't helping much regarding this topic also, therefore, I'm asking here.
tobias
Bitcraze
Posts: 2339
Joined: Mon Jan 28, 2013 7:17 pm
Location: Sweden

Re: Bitcraze flow breakout board with PX4

Post by tobias »

I guess it would be good to rule out any hardware defect. Have you been able to rule that out? E.g. by testing it with an Arduino or similar? In the other thread a comment is:
so, I think PMW3901 should be strictly powered under 5V.
Top
The Flow breakout contains voltage regulators that should work well up to 5.5V. However maybe the 5V he supplied was noisy and that effected the sensors.

We are not that experienced with PX4 so can't provide much help there, sorry.
krayola
Beginner
Posts: 3
Joined: Tue Sep 08, 2020 11:34 am

Re: Bitcraze flow breakout board with PX4

Post by krayola »

Hey Tobias,
Thanks for the reply.

Yes, I even created a custom regulator outputting 3.7V using LM317 regulator that is powered from the battery, I'm getting a stable 3.7V therefore, I don't think it's the case.

The integrated range sensor seems to be loosing sight mid-flight, I'm getting a lot of maximum values outputted in the log file, and at the same time the quality of the pmw3901 is dropping accordingly, so at the moment I'm trying to connect a terarange one lidar, to see if that helps with actual distance determination during flight, that is essential for the optical position hold.

The HW seems to be fine, I'm able to start up both sensors via the PX4 CLI, even a custom FW was compiled in order to start up the vl53l0x sensor on the boot-up process, since the PX4 original latest firmware was lacking this feature, after which via the analyzer tool, all of the sensors are outputting what seems to be like sensible values.

I'm hoping that someone has done it in the sense of actually enabling the optical position hold during flight to the point where the drone truly was holding the position, and can perhaps help with troubleshooting and the parametrical setup for the pixhawk4_v5 setup.
tobias
Bitcraze
Posts: 2339
Joined: Mon Jan 28, 2013 7:17 pm
Location: Sweden

Re: Bitcraze flow breakout board with PX4

Post by tobias »

The range sensor, VL53L0X on the flow breakout board is not that great at distances over 1m, maybe that is one of the problems?
krayola
Beginner
Posts: 3
Joined: Tue Sep 08, 2020 11:34 am

Re: Bitcraze flow breakout board with PX4

Post by krayola »

Hey,
I've even tried flying @ rather low height(below 1m), but that didn't cut it either, so this is the reason I'm working on "teraranger one" lidar implementation. Will let you know how that goes, but at the moment figuring out the HW aspect of the whole teraranger lidar venture.

But does that mean that crazyflie drones can fly only up till 1m of height?
tobias
Bitcraze
Posts: 2339
Joined: Mon Jan 28, 2013 7:17 pm
Location: Sweden

Re: Bitcraze flow breakout board with PX4

Post by tobias »

But does that mean that crazyflie drones can fly only up till 1m of height?
With the flow deck v1 the recommendation is to stay below ~1.5m but the newer sensor VL53L1x goes to 3-4 meters (flow deck v2). It is also possible to fuse with the barometer, nothing we are doing though (yet).
Post Reply