One prop, two chunks of GPS mast so far!

[crayfellow]

You had to move it up to get it to engage the arming?

Typically its the other way around - requiring lowering the trim to get it low enough to kick in.

No, arm and disarm behaved as expected. Throttle at 0, yaw right armed, yaw left disarmed. So I could arm, give it a teeny bit of throttle to make sure the motors would spin, then down to 0 and disarm to make sure the motors would stop. This isn't a perfect test of course, but it seems promising. The only difference from "correct" behavior is having the motors spin when armed. For that I believe I need to adjust MOT_SPIN_ARMED, but I feel like I should do another quick flight first to see if it behaves and spins down the motors upon disarm, before changing anything else.

I also will need to check the minimum throttle (something greater than whatever MOT_SPIN_ARMED ends up being).

You said that the throttle shows the correct reading in mission planner? And the taranis stock cal has been done?

Throttle shows correct, 1100us at minimum, and Taranis weight and offset were adjusted on the channel in order to get there. I just used trial and error until the Mission Planner range were where they needed to be.

 

[Motopreserve]

I'm not sure weight was the way to do it. Typically you would just move the end points and the offset until it was correct. Not usre there is any downside to the weight being changed - someone else will have to weigh in on that.

If the motors are spinning when armed, that should be adjustable with the parameter you mentioned. Some FCs do it automatically (DJI, Xaircraft etc) and some do not (multiwii). Takes a bit of getting used to - but if set correctly, I actually now prefer to see them working, but at such a slow sin they will never provide enough thrust to do anything whacky.

 

[crayfellow]

I'm not sure weight was the way to do it. Typically you would just move the end points and the offset until it was correct. Not usre there is any downside to the weight being changed - someone else will have to weigh in on that..

just my inexperience, that was the way I was able to affect the endpoints in the SERVO screen. Where are the endpoints affected, can that be changed without affecting other models in memory? I'd definitely rather have somewhere more precise to strictly define 1100-1900 vs. weight and offset, that is simply the way I found to do it.

 

[crayfellow]

If the motors are spinning when armed, that should be adjustable with the parameter you mentioned. Some FCs do it automatically (DJI, Xaircraft etc) and some do not (multiwii). Takes a bit of getting used to - but if set correctly, I actually now prefer to see them working, but at such a slow sin they will never provide enough thrust to do anything whacky.

They are not spinning when armed, that is why I gave a tiny bit of throttle to make sure they would, then that they would stop when disarmed.

I had turned it off on the mini quad, but I appreciate it on the tri and think it is useful on these big ones as well, even with the Pixhawk confirmation tone.

APM may well do it "automatically" too, but there are so many parameters at play here that I may be working against it until I get things right where they need to be. I know APM::copter 3.3 will be another big step in features, but I won't be running that on here until it's released.

 

[Motopreserve]

In the computer you do it under the "limits" tab. It will not effect anything else for the model - and you would only be changing the values for the throttle. On the taranis itself, it's the servos page. The default is -100, +100. The center is default to 1500. You can change it there, and there is an offset value as well.

 

[crayfellow]

In the computer you do it under the "limits" tab. It will not effect anything else for the model - and you would only be changing the values for the throttle. On the taranis itself, it's the servos page. The default is -100, +100. The center is default to 1500. You can change it there, and there is an offset value as well.

I see - yeah I have never used the computer UI since I was curious to see if I could get by using only the Taranis UI. I will try doing it from Servos, thanks!

 

[crayfellow]

@Motopreserve OK I set the weight back to 100 and offset back to 0. In "Servos" I altered the endpoints until Mission Planner showed 1100-1900 range. I left the midpoint at 1500. Re-ran RC Calibration. I will do a test flight with these settings and report back.

 

[Motopreserve]

Glad you found it. It's good to be able to do all that you need on the radio - for being at the field. But in my experience, using the computer is what makes the taranis a step above the rest.

Good luck!

 

[crayfellow]

another day, another crash. Today's casualty is 2 450mm x 16mm carbon tubes as I landed hard on two legs.

Wow - scary - I took off in Loiter and held it there for awhile. And mind you, it was windy, somewhere around 10mph but I don't have the luxury of choosing ideal testing periods. It was holding pretty well so I got my orientation and tried a bit of pitch and roll to make sure the channels didn't need to be reversed. Then a bit of yaw. Then I played with the throttle a bit but it seemed as if it wanted to climb no matter what I did at that point. This is not the classic Pixhawk RTL-when-GPS-lost because I have failsafe disabled for now, and this is a very large field with the widest LoS to sats possible for that matter. I went so far as to drop throttle to 0, and it kept climbing!

So I had a moment of lucidity and flipped it into Stabilize. It fell out of the sky, and I managed to catch it with enough throttle that it didn't crash into a million pieces. But by then I was so nervous that I didn't have the mental facilities to get it in a stable hover before landing, so it landed with some lateral velocity and broke the legs.

I do believe when I landed and disarmed that the motors stopped, as I had no "motors spinning when disarmed, yaw right speeds up the motors!" insanity this time.

I will need to pull the logs and learn to read them as it seems like all the APM people are busy helping Solo customers these days. If anyone has any pointers to help me decipher why it climbed and would not come down, I'm all ears.

Certainly I am no Loiter mode or Pixhawk expert, but on the tri it will gently descend, then land when I move the throttle toward 0.

 

[crayfellow]

and for the record, I'm not theorizing it was trying to fly away, maybe some aspect of Loiter I don't understand required that it climb to 25m altitude or some preset I don't know to adjust, yet.

 

[Motopreserve]

That seems damn scary.

One suggestion unrelated to the Loiter issue: if you are testing the pitch and roll etc while in the air - meaning you're not 100% confident in the setup, it may be best to have someone hold it above their head for you while you gently run through all the stick commands. Or you could tie it down tight to the ground - just enough slack to get light under throttle, and test that way.

Good luck with the logs.

 

[crayfellow]

Yeah. That's one big issue is I am always testing alone (not ideal I know, don't really have a choice). I won't bring my kids with this one until I'm 100% confident in it. Tying it down would work, though.

I don't think fiddling with pitch/roll/yaw caused the climb. The wind blew hard right then and, I dunno, tweaked on the GPS/compass mast maybe?? The weird part for me is that, in the moment at least, it seemed to refuse to come down, and I didn't want to lose the primary rig for my entire research effort.

I will regroup, put on some much shorter 315mm legs, strap the water bottle back on, and maybe just try stabilize for awhile until I know what happened.

 

[Motopreserve]

No. Definitely no kids!

Tie downs are a good way to go if you're on your own.

Doesn't seem like it was related. The GPS mast should be rock solid - make sure you rule that out as a suspect. Regardless, if you lowered the throttle to zero then it should have reacted and come down - and fast! Regaining in stabilize mode is clearly the thread to follow.

 

[Mactadpole]

I have had a similar problem in the past and for the life of me can't remember exactly what it was. I believe it has to do with mid-throttle calibration (parameter) and the difference in position of the throttle to that. You might need to adjust the dead band around center throttle too. I would continue to fly with the dummy weight for sure. Start digging into the logs, they can answer a ton of questions or at least help you think you know what happened and try to trouble shoot from there.

Definitely stick with flying in stabilize mode first and get the throttle mid value tuned. Once your happy with that work on altitude hold.

Its pretty scary when you think your big bird is getting ready to fly away. I can't recommend the Garmin Astro dog tracker setup enough. I would have spent hours looking for the downed copter last week compared to walking straight to it. At least fly with a ground station so you can track it that way for as long as the telemetry stays connected.

 

[crayfellow]

I know, I need one of those for sure. Here I am thinking I'm doing a quick hover test/reality check and I'm right on the verge of a loss!


Sent from my iPhone using Tapatalk

 

[Motopreserve]

I assumed that you still had the weight on. I dont think that it's the best idea to rip around without the wight the craft was designed for. I know it's possible - but knowing the thought and care you put into the design of the power system - you are definitely not going to be near 50% throttle for hover at your specs.

Best to eliminate at least that variable of the equation that you know for sure will effect handling - to make it easier to track the issue.

 

[crayfellow]

Weight was on at all times. That is, until the incident. One of the tie wraps snapped.


Sent from my iPhone using Tapatalk

 

[Mactadpole]

Weight was on at all times. That is, until the incident. One of the tie wraps snapped.


Sent from my iPhone using Tapatalk

Ok, I had interpreted previous statement as you were flying with it off. Sorry. Have you measured the gimbal setup with camera and matched the water bottle weight to that. When you get around to auto tune I would do it without the gimbal, etc. and just do it with the bottle as close to the weight of gimbal and camera as possible. I have found that doing autotune 2-3 times seems to do a better job of getting close to desired PID's with these bigger copters. For some reason that first autotune always seems to set the PID's to high for my liking.

 

[Motopreserve]

Sorry. Got confused there as well.

Try to make sure the water isnt moving/sloshing when you do the auto tune, that would likely throw it off.

 

[Mactadpole]

I've also constructed a lego container with lead fishing weights inside (built so they don't move around) as dummy weight before too!