XAircraft SuperX

[fltundra]

Ok, thanks! What's your UAW? Mine's at about 14.5 lbs.

That's a lot of weight for those motors, even with 6S power. Triple 15's is to big. I think you will see massive improvement with 14's.
While your changing the props check your CG and make sure the motors are perfectly level.
My AUW 2360g. on a 4S7000mah.

 

[Motopreserve]

Damn. Sorry you're having these weird issues. Always scary when a stable platform starts to act kooky.

There must be some type of center point setting in Futaba. I set my gain knob by firing it up on the bench with props off - set it so the motors are spinning (black box recording). Turn knob slow from center to right. Pause. Right back to center. Pause. Same on the left. Then when you look in black box you'll watch the "g" box and see how it reacted. Then you can reset end points for less travel etc. Redo it until you have the gain knob acting the way you like.

As Tundra said, setting center point (offset in Taranis) higher will allow you to begin at Kloner's suggested +20%.

 

[fltundra]

KDE doesn't even spec the 515's with 15's, let alone triple 15's. 14's on a 6s, on a hex is 23.1 lbs thrust at 62%.

 

[SamaraMedia]

I didn't expect this rig to get this heavy, first time I weighed it today, 6525 grams. Now that is with 3 5000mah 6s batteries for weight, testing by using one at a time, only about 4 minutes per batt before low level kicks in. My hope is that when I switch over to some 8-10,000 mah batteries and a 3 axis gimbal I can get the weight down some. I thought these motors would have more oomph that the 2814's but maybe with each motor weighing more etc...things almost balance out weight to power ratio. Such a balancing act and I seem to be tipping all over the place lately

"There must be some type of center point setting in Futaba. I set my gain knob by firing it up on the bench with props off - set it so the motors are spinning (black box recording). Turn knob slow from center to right. Pause. Right back to center. Pause. Same on the left. Then when you look in black box you'll watch the "g" box and see how it reacted. Then you can reset end points for less travel etc. Redo it until you have the gain knob acting the way you like."

So you run the hex with the SuperX connected to your PC to view the logs, never knew you could do that. I'll have to try that.

 

[SamaraMedia]

KDE doesn't even spec the 515's with 15's, let alone triple 15's. 14's on a 6s, on a hex is 23.1 lbs thrust at 62%.

Yeah, I was using 14" on that setup and it was wonderful on the UAP 800 size hex I built. Wanted something transportable and able to carry a GH4 cam so went bigger.Since the DJI S900 runs 4014's and this frame is based on that (sort of) I thought the 4012's would be a good fit, I may be wrong...

KDE thrust ratings always seem so high in relation to size, that is a lot of thrust for a 2814.

 

[SamaraMedia]

KDE 4012 thrust at 6s 50% is 1270 grams, 2814 - 515's are 1360 grams at 50%, that just seems so outragious.

 

[SamaraMedia]

Damn. Sorry you're having these weird issues. Always scary when a stable platform starts to act kooky.

There must be some type of center point setting in Futaba. I set my gain knob by firing it up on the bench with props off - set it so the motors are spinning (black box recording). Turn knob slow from center to right. Pause. Right back to center. Pause. Same on the left. Then when you look in black box you'll watch the "g" box and see how it reacted. Then you can reset end points for less travel etc. Redo it until you have the gain knob acting the way you like.

As Tundra said, setting center point (offset in Taranis) higher will allow you to begin at Kloner's suggested +20%.

I misinterpreted your post, thought you had the SuperX connected to your PC while on the bench to see g knob, read it a second time and figured out what you meant. That's kind of what I was trying to do this morning in the wind.

 

[Motopreserve]

Right. No need for wind.

Do it on the bench since you'll probably need to go back and forth a bit. But unfortunately still have to connect/disconnect to computer each time. I have mine set so it really only goes a bit negative, and up to about +100%. Also, watch carefully in black of to make sure the g knob is not skipping around like crazy. My tests before following Kloner's advice showed the values jump from positive to negative and back, all over the place - deposits only turning the knob in the positive direction. It doesn't like wide end points.

 

[SamaraMedia]

Well, I'm packing it in for the night, week, month, year, who knows...

Latest bench tests have frustrated me to my wits (and there ain't much left) end. Tried as Moto suggested and ran the hex on the bench with no props taking note as to position of the g knob then looking at the logs to see what's happening. First I had to figure out how to get the g knob to show +20% at the center point, I think I was able to to so using the sub-trim on that servo. What I did find is that getting +20% on the Futaba servo menu does not translate to +20% on the g knob in the Black Box, much higher. So after the first test I lowered the sub-trim so that is shows about +6% on the servo menu and that translated to about 35% on the g knob. Not sure if this is the correct way to do this but it's what I came up with at this point.

What really frustrates me is that the motors seem to have no real control from the throttle input, especially in GPS or ATTI, I know those two modes have a very smooth response curve but it seems like there is no more speed available past the 65% point on the throttle stick. Motors seem to be all up at that point, a little better response in that regard in Manual mode but only slightly. Looking at the log report it looks like the motors are at 95-99%, now that could be attributed to not having any props on and getting no resistance, but the charts are looking worse than ever.

The motors and esc's were purchased used here but I can't say that is the issue, probably more likely I'm missing something in my wiring, radio setup or something else but I've hit a wall that has me down for the count right now. Too much to process, one step forward, three steps back.

Here's the latest log if anyone is interested.

Thanks for all you help guys!

http://log.xaircraft.cn/#2015_06_09_075900.739_94.1.02.07 and http://log.xaircraft.cn/#2015_06_09_075900.739_95.1.01.09

 

[Motopreserve]

You are on the right path. The numbers in your trims will not translate exactly. You just need to raise it slightly to get to +20%. If you're at 6, do 3 or 4. That should be close. Then change the end points. Maybe try only -30 & +30 and see where that allows the g knob to travel to.

You can't judge the motor output on the bench when in ATTI/GPS modes. Remember, those modes are looking to do some work that is not possible without props, and on a bench . Manual would be the only mode that would act "normal" in that situation.

Take a step back. Best thing you can do when you're frustrated. You'll come back fresh and hit it with a new-found calm.

 

[fltundra]

You are on the right path. The numbers in your trims will not translate exactly. You just need to raise it slightly to get to +20%. If you're at 6, do 3 or 4. That should be close. Then change the end points. Maybe try only -30 & +30 and see where that allows the g knob to travel to.

You can't judge the motor output on the bench when in ATTI/GPS modes. Remember, those modes are looking to do some work that is not possible without props, and on a bench . Manual would be the only mode that would act "normal" in that situation.

Take a step back. Best thing you can do when you're frustrated. You'll come back fresh and hit it with a new-found calm.

It still won't act normal in manual mode either. Gyro's still come into play. Best way to check, with props REMOVED in manual mode. Set the throttle at 50% and lift her up and tilt her on all axises. Whatever motors you tilt toward the ground will rev trying to level itself.

 

[SamaraMedia]

Thanks guys! g'night.

 

[SamaraMedia]

Ok, new day new attitude

Balanced my 14" Maytech CF props, revisited my motor alignment using iPhone app (used it the first time too but put the phone on my island counter top to zero out the bubble level but after watching kloner's video some pages back I decided to check again. SleepyC also recommended this method but I failed to understand his simple instructions) All motors are within in 0.0 & 0.2 although with the cantilevered arms it can be tricky to get them properly level.

Went to the test field and burned three batteries and trying GPS and ATTI modes moving the G knob slightly to see where the hex liked it, sounded better in the air than previous tests so that could be motor alignment.

Problem now is the damn Black Box upload, been trying for over a half hour with now success. Used the app fltundra provided link to but that will not allow me to share the results like the BB. Why is that still a Beta version? Can't they get something to work 95% of the time? There is much to be said for some other FC firmware that allows for local viewing instead of a server in the Motherland...

 

[Motopreserve]

It's a beta because I believe it was created by someone outside the company, no?

Anyway, sounds like you're moving in the right direction. The motors being level will help tremendously. If the vibes are low - you can then focus on your motor outputs. Were they better in this last test???

 

[SamaraMedia]

Ok, made a few screen shots of the three test flights, first one just hovering in GPS and turning g knob, 2nd hovering in GPS and some forward flight, 3rd ATTI mode, hovering and flying around field a little while trying a couple different g settings. Looking at the logs in the app it seems like it likes somewhere around the 52-60% for g gain. FC setting were 1.0, 1.0, 0.9 yaw, 0.9 height. Going to try all at 1.0 except height next time to see if there is difference.

One thing that I've begun to consider as a potential pitfall of this frame design is having the FC on the top deck and all the weight, batteries and gimbal/camera below deck may cause too much of a swaying for the CG.

We'll cross that bridge when we get comfortable enough with gains and get the gimbal setup...

 

[Motopreserve]

It's seems to me that you are close. I would think it worth checking with all gains up .1 in software of your knob was booting it 60%. I find that if my craft is tuned and flying well, I have the knob centered at 20%, and I barely ever move it from there unless the wind is a *****.

Vibes are also lower, but may be able to hone in a little tighter. Not sure you should divert your attentions on gain tuning for now - with spikes of .8 but averaging .4 or so.

I have my superx up on top of the Vulcan before and I'm not sure it really adversely effected it.

 

[SamaraMedia]

Moto, are you saying to go up.1 on all gains to start - roll 1.1?

Vibes may be present most of the time due to the folding properties of the frame, I'll take .4 with a few spikes for now. May be that's when the hex is adjusting to some g knob tuning but I will keep an eye on it. Thanks

 

[Motopreserve]

Yep. If it's flying best at +60%, maybe the main gains should be up more - and then knob can be back at starting point (+20% if you were able to get that sorted). Then you can adjust from there.

It would at least be a good test. I say this because the knob can be flaky if not set exactly right for your radio. Better to nail it with the main gains - tweak with the knob.

 

[fltundra]

Ok, new day new attitude

Balanced my 14" Maytech CF props, revisited my motor alignment using iPhone app (used it the first time too but put the phone on my island counter top to zero out the bubble level but after watching kloner's video some pages back I decided to check again. SleepyC also recommended this method but I failed to understand his simple instructions) All motors are within in 0.0 & 0.2 although with the cantilevered arms it can be tricky to get them properly level.

Went to the test field and burned three batteries and trying GPS and ATTI modes moving the G knob slightly to see where the hex liked it, sounded better in the air than previous tests so that could be motor alignment.

Problem now is the damn Black Box upload, been trying for over a half hour with now success. Used the app fltundra provided link to but that will not allow me to share the results like the BB. Why is that still a Beta version? Can't they get something to work 95% of the time? There is much to be said for some other FC firmware that allows for local viewing instead of a server in the Motherland...

What is your current draw in hover now?

It's seems to me that you are close. I would think it worth checking with all gains up .1 in software of your knob was booting it 60%. I find that if my craft is tuned and flying well, I have the knob centered at 20%, and I barely ever move it from there unless the wind is a *****.

Vibes are also lower, but may be able to hone in a little tighter. Not sure you should divert your attentions on gain tuning for now - with spikes of .8 but averaging .4 or so.

I have my superx up on top of the Vulcan before and I'm not sure it really adversely effected it.

Moto, are you saying to go up.1 on all gains to start - roll 1.1?

Vibes may be present most of the time due to the folding properties of the frame, I'll take .4 with a few spikes for now. May be that's when the hex is adjusting to some g knob tuning but I will keep an eye on it. Thanks

Moto is right, I was going to suggest the same thing. One thing, .9 is kinda low and might cause tipping issue on takeoff. If I remember right default gains out of the box were 1.1 straight across.

 

[SamaraMedia]

I emailed Patrick at KDE in regard to my motor/esc/prop setup and here's what he had to say:

"For sure, the 15" triple-propeller edition is not excessive for the KDE4012XF motor, and is quite popular in the market. This will be more efficient for you than running KDE2814XFs, which are more designed for 14" and less propellers - gearing more towards 500-class and smaller frames. The Tarot X6 is a larger-frame, and best to get into the 15" and larger propeller category for the proper performance range.

The smaller motors (2814 515kv), which capable of higher-thrusts, have less efficiency and less thermal mass for the larger-system that are designed to carry high-loads for extended durations. Looking at the Tarot X6 design, I would recommend running this with the triple-propeller setup, on either the KDE4012XF-400 motors or the KDE4014XF-380 motors (slightly larger and even higher-efficiency on the larger propellers) - matched to the KDEXF-UAS55 ESCs for and excellent and dependable system."

My last three test flights were with the 14" Maytech props, software gains all at 1.0 and gain knob I believe about 16-20% and center. Still can't upload to BlackBox so using the app charts look pretty good. @fltundra - not sure what my amp draw is since I can't see it anywhere on the app info and can't upload to XAircraft.

Haven't had a chance to do much else but making some progress. I will likely go back to the 15" triple props and test again.