Radio Calibration

Confirmed, putting in issue on github.
https://github.com/bluerobotics/ardusub/issues/20
The issue seems to be on qgc side, changing ARMING_CHECK through mavproxy works as expected.
-Jacob

It seems that the “Load Parameters” function will force the ARMING_CHECK to disabled if you have previously saved your parameters prior to installing the more recent versions of QGC.

Regards,

TCIII AVD

Hi Team,

Just to second the above issue with the arming check. I had to use a prior version of QGC to set the arming check to disable as well. The newest version does not save the arming check parameter mods correctly…apparently.

This is only the case when using the pixhawk, my navio+ does not have this issue…it has others LOL hence the pixhawk…unfortunately.

Jim

We finally have everything working. We did a fresh installation of QGC on a new laptop and now we can disable the arming check. It seems that on my other laptop it still keeps certain settings in the registry even when I remove the software.

We have bought a Logitech F310 controller to make sure the problem wasn’t our Xbox One controller. This atleast made the joystick and RC calibration a lot smoother. We still had to change the RCmapping for Forward and Lateral movement manually.

We discovered that we had to change the direction of all motors for the pixhawk to work correctly. We did some wet testing and everything seems to work perfectly. Thanks for your input in this.

One thing that’s not working perfectly is the following. We are currently powering the ROV with a 15V power supply with sense wires going to the ROV to compensate for the voltage drop due to tether length. This works fine but the power supply sometimes drops out when we run multiple thrusters simulteneously. This mainly appears when we change heading at full power. Is there a way to change the maximum output power for heading changes? I have tried changing the min and max PWM output for channel 4 to 1400 and 1600. This doesn’t seem to have much effect.

Any help in this would be much appeciated. This week we will try to go outside to a lake to do some real testing. We will use the battery packs for this so this should be no problem. It would be nice though to have everything working with the power supply.

Hi Klaas,

Great! Hopefully the QGC developers fixed that bug. We raised an issue to them a few weeks ago and they are very responsive to fixing things like that.

Regarding the RCMAP parameters, we have recently hardcoded those in the software so that you don’t need to set them any more. I would recommend updating your ArduSub firmware to the most recent version to get those changes.

Regarding the power output, I would recommend changing the ATC_RAT_YAW_… PID parameters to get a more gentle response from the control system. That should help out.

How long is the tether? It’s impressive that you’re able to get a 15V supply to work at all!

-Rusty

Hi Rusty,

We downloaded the most recent Arudsub firmware a few days ago so I was assuming we had the most up to date version. What do you mean by hardcoded? Is the RC calibration removed in this case?

Regarding the Yaw power output, I will try this tomorrow. Our power supply is probably not powerfull enough to have multiple thrusters running at full speed simulteneously. The power supply we are using can be adjusted between 12 and 30V (Home. | Delta-Elektronika.nl). We have set it to 15V with sense wires going down the tether. We are currently using about 75m of tether (two rugidized CAT5 cables and two power wires bundled to one tether) and the output voltage at the power supply when running two thrusters is about 21V (6V power drop over 75m)

Hopefully we can do some outdoor testing this week.

Klaas

Klaas,

Okay, you should have the most recent version then. We are bypassing the RCMAP parameters so that they don’t actually have any effect anymore. See here: ardusub/radio.cpp at master · bluerobotics/ardusub · GitHub

That power supply should work for gentle power loads but you’ll get big voltage drops if you use too much power. Cool setup!

-Rusty

Hi,

We are having problems manually controlling the motors with an XBox controller.

Possible issues: Pixhawk and Mavlink not talking, because Pixhawk is not sending commands to ESC. We also tried startup in ESC calibration mode and reuploading the px4 file. We changed the gains and were able to calibrate the Radio/Xbox controller, but the motors don’t respond at all. The ESCs make a sound when we arm them and turn on the battery. It’s only when we press the switch down and it changes to flashing red that the motors spin for a quick moment. Is there something we are missing, like how to switch between flight modes?

We tried both connecting the pixhawk to the raspi and connecting it directly to a PC/Mac. The ESCs were connected to the Main out ports. At this point we’ve tried many of the solutions in the forum posts.

Thanks in advance!


@Alicia,

Could you save your parameters settings file and post it here? I think that will help us determine what might be happening. It sounds like when you press the arm button, the motors spin momentarily and then stop?

-Rusty

Here it is! Sorry for not attaching it before. And yes, but I think it happens when we press the button to disarm the motors since the light goes from solid to blinking red.

failedmotorteststxt (16.6 KB)

Alicia,

Do you get any warnings or info in the “Info View”? The info view can be seen by clicking on the “horn” icon in the top bar when connected to the ROV.

-Rusty

No, we didn’t have any errors in the info view.

Okay, can you open the “Mavlink Inspector” from the Widgets menu and then look at the servo output message and see what values are being sent to the motors?

The values were changing within the 1000 - 2000 range in the Mavlink inspector. I can’t attach a picture because this morning we plugged the power module into the pixhawk for the first time and the pixhawk got fried :confused: We probably won’t be needing more help for a while haha

Yikes! That’s not good! Did you hook it up normally?

With those values the thrusters should be moving. Let’s make sure your ESCs and power are hooked up correctly. Do you have a picture of that or can you describe your ESC and power setup?

-Rusty

Yup, we’re chatting with 3d robotics support to figure that out.

EDIT: We looked at this page LEDs Meaning — Copter documentation and the main LED was flashing blue, don’t know if that was an issue as well?



@Alicia,

Is the Power Module a genuine 3DR Power Module or a clone?

Many of the clone Power Modules did not include the 5 vdc regulator that the 3DR version included and actually supplied the full battery voltage to the Pixhawk instead of the expected regulated 5 vdc.

Regards,

TCIII AVD

TCIII,

We measured the output and it does give 5V, and it is a 3DR power module. We are wondering if the Pixhawk was somehow defective and that’s why the motors didn’t run, or if it had to do with the parameters, flight modes, etc.

Alicia

Update: Our issue was our pixhawk, our new one works fine with the power module.
After realizing we needed to arm both the xbox controller and the safety switch, the ESCs calibrated properly and the motors run.

@Klaas Visser,

Hi Klaas, I’m also Dutch, and interested in your external power supply setup.

Can you tell me more about it?