MultiWii Flip 1.5 Help

[constantmotion]

I recently received a Flip 1.5 from RTF Quads for my ElectoHub build. This is my first MultiWii board and I'm not new to multirotors. I setup it all up correct and calibrated everything. When I lift off it wants to pitch forward and to the right a pretty good bit. I have recalibrated ACC several times...making sure the quad was perfectly level. In the MultiConfi software it shows pitch and roll at 0 when the quad is sitting level. I have also tried to retrim the ACC with no effect.

I'm not sure if it has anything to do with this issue but the artificial horizon in the software shows that my roll is reverse but graphic of the quad is correct....quad rolls left, graphics shows rolling left but the artificial horizon rolls right.

 

[Craftydan]

take a moment to think about the instrument from the quad's perspective . . . if you lean your head to the left, which way does the horizon move?

 

[constantmotion]

take a moment to think about the instrument from the quad's perspective . . . if you lean your head to the left, which way does the horizon move?

The horizon tilts down to the left. It should be down to the right. I'll post a screenshot or video of the GUI this afternoon

 

[Craftydan]

If you lean your head to the left, the left side of the horizon rises from your head's viewpoint, while form your sholder's viewpoint, your left eye goes down.

Sounds like what you're describing the MWC is showing (and I've seen it do before), which may feel backwards, but is correct.

 

[Craftydan]

BTW, as far as the unbalanced takeoff . . . this doesn't soundd like an ACC calibration issue, and an ESC calibration issue should be balanced out by the gyro correction. two potential sources:

- Sloppy ESCs -- cheap ESCs with a poor ROM will add *a lot* of slop in the control loops, making some a little faster to respond and some slower. on a strong change in thrust the control board will have trouble keeping the gyro rate errors down if it doesn't have responisve contorl form the ESCs.

- Poor PID gains on the gyros --if the control loops aren't tuned for high enough gains, it simply can't respond strongly enough to keep the craft under control during strong changes in thrust.

Once a boom is way off angle, the Gyro's will know something happened and cancel out the unbalance (eventually), but have no way of knowing how far to go back -- a little bit, the I gain can remember, but too much and the error becomes lost.

 

[constantmotion]

I new it wasn't the ESCs or Motors. These components are donors from other working quad. Just to double check, I revered my motors and ESC..moved front motors to the rear and rear to the from. I get the same characteristics pulling hard to the front right. I did recalibrate the ESCs and the motor do start at exactly the same time. I did notice that in Manual Mode it did fly a little normal...PIDs could be adjusted for better flight but it'll mostly be flown in self level mode. There has been too much rave about the controller being able to fly pretty good out of the box and I'm not seeing that at all. There has to be something that is missing.

Here is a video showing the artificial horizon.


I did see something interesting with the motors powered up in the GUI. When I pitched and rolled the quad really only the front right and back left motors were changing speeds. The front left and back right were not sensitive to the gyro at all. Also the RTF Wiki says "sitting on your desk the ACC numbers should be approx 0/0/250". Mine is 0/0/500 when freshly calibrated.

Here's a vid

 

[Craftydan]

Hate to break it to you . . . the instrument is right. I think where you're getting wrapped around the axle on that is the screen isn't moving -- if you tilted with the copter (I don't recommend you do this, you'll drop and break your monitor) you'd see the horizon stays aligned with the real horizon as the screen tilts. again this is the copter's perspective on the world as the left boom tilts down, it's view of the horizon is the left side rises.



Back to your off balanced launch . . .

- so you've moved parts around and that didn't help. So it's not that easy.

- you've set the PID's for acro mode --is it more or less easy to place it in an attitude and hands off it stays there without changing attitude? not the same spot in the sky, but if I lean it back to the left and let go does it stay leaned back to the left? If it can do that your Gyros are dialed in, if not, they need a little more tweeking. The better they are, the more likely the board will fight this unbalanced takeoff.

- if that still doesn't help check for warps in your booms and mounts -- it's the natural out of trim (mechanical or electrical) the contorlboard is fighting in the first few seconds of flight, and after that it should have them under control


All that being said . . . I'm not a fan of multiwii's self level -- I've never gotten it to fly as I like it, and on my boards it has had a strong tendency to "forget" level in FFF. Others may be able to help tune that, but all I've gotten out of that process are broken props.

The flip boards are very nice and multiwii is a powerful box of tools, but I'm not in the "it's plug-n-play" camp. it's surprisingly close out of the box, but tuned? no. easy to tune the last little bit? In a plane-Jane quad, maybe, but oddball cases like the dead-cat, it might just be easier to start with a kk2.

Keep at it -- What you have can fly, and it can fly well . . . hopefully well enough for your satisfaction.