small weight vs heavy weight

[gionag]

Hello guys, today is a great day for my aq

just for testing, i put the Aq on a small quad 57cm motor-to-motor, Tmotor MT2208-16 1100kv, graupner E-prop 8x5.
6.6 r37, tmotor 18amp esc with Simonk fw latest revision. 3s 2200mah. PPM. 770gr AUW.
The calibso was done to another frame (hexa).

The performance was outstanding... really, never seen something like that. no pid tuning at all, motor mixer 100/50/50/100 (as wiki...). perfect stable hover, perfect pos/alt hold, perfect missions.

Now the question.

why that things performs so much well on a small quad and when i bring an heavy weight hexa in the equation this brain performs so poor to make me thinking where the 400 euro spent ? i think the pid and other things obviously make a huge difference. but, in almost a month i have done all sort of tuning i can't get results not near of what i achieved today ?

also statistically i seen the Aq performs much better in small situations rather than something bigger and heavier.

what do you think about ?

[bn999]

It is all about tuning parameters. AQ was originally designed on smaller (< 2Kg) platforms and most of the default parameters reflect this. This does not mean the board will not work on larger machines, but they will most certainly not fly as well using the default parameters.

My suggestion is to find someone who has a similarly size craft with similar motors & props and flying well. Use their settings as a starting point for tuning.

When I say tuning, I mostly mean attitude PID's. This should be done in manual mode exclusively until it is perfect. Then some of the navigation PIDs might need tuning, but usually not. Nav PID's can only be tuned in autonomous modes.

[gionag]

[teramax]

gionag, i am a little confused, you try to sell the controller and continue testing?

But i must congratulate you for the finding we discuss now a long time. If you stay under 2kg, rig flys great with near to default PID, abouve 2kg it will but only with sophisticated tuning.
What shows us that tuning is a very hard job if you dont hit a lucky strike.

Bill, could you provide us with some sort of guide, in example: if i use a 2kg rig that fly realy well and i rase weight to 3 or 4kg which parameters i have to watch, raise or lower.
Just something for us to compare.

I know that there are hardly 2 exact same copters under the AQ out there but a referenz would be fine, thought.
Thanks
Marcus

[gionag]

[HeliHenkie]

Hi all,

Just like Gionag I'm considering selling my AQ board. I've literally spent weeks trying to get it to work, with only minor success. My quadcopter has an AUW of 2.2 kg and flies very well with the original KK board and the Naza controller. In the process I replaced all my motors, props and other parts by top of the line material hoping that it would reduce vibrations. With no progression I sense that I'm losing my interest and yesterday my wife could only barely prevent me from ritually burning the board

At this moment alt/pos hold is working, more or less. RTH is working, sometimes.
I suspect this is caused by vibrations, but without a guideline I'm not sure what to aim for. Especially for the x and y axis. I've been trying to tune the PID's but without the hoped for result. In the wiki a lot of material is posted about PID tuning, but after reading I have more questions than answers. It would be nice to have a step by step guideline for PID tuning. It appears I don't care for the 'Art of PID tuning' but want a more scientific approach.

I'd like other AQ users with a quad or hexacopter over 2 kg AUW and success with alt/pos hold and RTH to post their PID settings and log files of their flight. This gives me the possibility to analyze and try other, less random, changes in my setup. Furthermore I'd like to see some examples of flight vibration data on all axes. So I know what to aim for.

Hope someone can help.

Best regards,

Jan Willem a.k.a. HeliHenkie
The Netherlands

[Max]

Here's my first AQ ship, Y6 at ~2.4kg, default PIDs: viewtopic.php?f=27&t=887

Onboard video from same Y6, same settings, but loaded with cam, gimbal, bigger 4S batts, in heavy wind, 3.4kg: viewtopic.php?f=27&t=887&start=10#p4322

I don't think weight has much to do with it. Vibrations, definitely. I later tried to upgrade the same Y6 frame with bigger motors, and that made it un-flyable. viewtopic.php?f=24&t=1453. I'm going with a new frame for my heavy lifter.

Jan, you could always post your graphs/logs and ask for feedback. Sometimes comparing to some else's data can be misleading. I know we still have to finish a wiki page regarding vibration issues, but frankly it's hard to define what is "good" vs "bad."

-Max

[HeliHenkie]

Max, thanks for your reply. I must say your answer is not very hopefull. You state vibration is a cause of many problems, but there does not seem to be a clear correlation between vibration data and performance. So apparently it's possible to fly a multicopter with bad vibration data. This makes ruling out problems more a question of good luck than systematic tuning.

This weekend I will make a flight log and wait for the comments on that.

Best regards,

Jan Willem

[Max]

Jan, It is hard to present empirical data because there is no controlled test environment. To really compare your logs to mine, we'd have to be performing the same test in the same way (eg. with the MR suspended in a particular way a certain distance from the ground, etc). Even a hover test is hard to replicate because the MR could be moving around anyway for other reasons (like poor PID settings, wind, etc). Ground tests are OK but a lot of vibrations are introduced from the prop wash, so again this is only useful as a relative comparison (meaning to compare your own tests with each other).

I've also used other tools to analyze vibration, like my smartphone with a "seismograph" app. For example, strap the phone onto each arm and run the motor up (just the one on that arm) and then compare the results... one might find a poorly balanced motor/prop combo or other issues. This can sometimes be quicker than making an AQ log for each motor and analyzing them.

You should definitely log all your flights, and please do start a thread and ask for feedback.

Cheers,
-Max

[Kisssys]

Jan,

Vibration is a big problem, you might have less vibration than the next guy but if it's at the right frequency it can cause lots of problems. Some people mount a little mirror on the boom and reflect a lazer off the mirror onto the wall to get a more visual representation of the vibration levels.

When I think I have one bad motor out of the bunch, I tie down the craft and remove all the esc lead's, power up the board and start logging. Then in a pattern I can remember I power up each motor individually( servo tester ) to hover rpm for a short period.

I then look at the logs to see if any motor is worse than the others. Then you can start the prop swap, engine swap, esc, tighten bolts etc. to isolate the problem.

I just flew my Big Octo without it's normal load and it jumped all over the place. It seems like being very light can be good as the motor's are creating less vibration but bad in that a very very small change in power can cause it to over react. The point is their is nothing out their flying that is more dependent on the perfect combination of parts than a multirotor.