Flip 1.5 - Motors won't spin up at the same time, and low RPM

[makattack]

My guess, one or more of the following reasons:

1) Vibrations causes the IMU sensors to go crazy -- how do you mount the flip1.5? Are your props/motors balanced individually and dynamically?
2) Failsafe is somehow triggered even momentarily. Do you have failsafe set? How is it set? On the TX/RX side, or on the flip1.5 with the failsafe configuration? If on the flip1.5, what are your config.h parameters for failsafe?

 

[jyunte]

My guess, one or more of the following reasons:

1) Vibrations causes the IMU sensors to go crazy -- how do you mount the flip1.5? Are your props/motors balanced individually and dynamically?

I balanced each prop, including the hub. The FLIP 1.5 is mounted identically to the way it's done in the Electrohub build video, with a strip of 3M VHB tape.

2) Failsafe is somehow triggered even momentarily. Do you have failsafe set? How is it set? On the TX/RX side, or on the flip1.5 with the failsafe configuration? If on the flip1.5, what are your config.h parameters for failsafe?

I will have to look into this. I have made no changes to the FLIP 1.5 v 2.3 Sketch file that is provided on the RTFQuads.com website. Unfortunately, I'm working long hours over the next couple of days and won't have time to look, but like I said, it's the same as the file on RTFQuads.com.

 

[makattack]

I balanced each prop, including the hub. The FLIP 1.5 is mounted identically to the way it's done in the Electrohub build video, with a strip of 3M VHB tape.

I will have to look into this. I have made no changes to the FLIP 1.5 v 2.3 Sketch file that is provided on the RTFQuads.com website. Unfortunately, I'm working long hours over the next couple of days and won't have time to look, but like I said, it's the same as the file on RTFQuads.com.

What wasn't mentioned on the flitetest build video for the electrohub/flip1.5 build, but is talked about in other videos, is dynamic balancing:

http://flitetest.com/articles/Laser_Balancing_Props

Doing the static balancing is great and a necessary first step, but after doing that, I took those balanced props and motors, and put them together, and balanced them together -- in my case, I just rotated the prop on the motor until I found the smoothest combination. I didn't use a laser, but I used the vibration meter that multiwiiconf shows at the bottom of the screen. I used a servo tester to run each motor up, but had the tricopter securely held down and the props clear of obstructions.

As for failsafe, you have two choices: The easiest, if your TX/RX supports it is to setup failsafe on that so that it cuts throttle or stops sending a PWM / PPM signal. Next, is setting failsafe on the flightboard, but that kind of depends on the ability of the RX/TX combo you have for setting failsafe to cut PWM/PPM signal.

I personally have both set, as I use two different kinds of failsafe types.

With my frsky X4R/SB RX, I set a custom failsafe mix to put it in horizon mode (hopefully to right it up) and then cut throttle. On the flip1.5, I just have it cut throttle -- the default is to sort of "autoland" with a minimum throttle run over a time constant. I set the throttle to zero and the time constant to zero as well.

Here's my modified config.h:

    #define FAILSAFE                                  // uncomment  to activate the failsafe function
    #define FAILSAFE_DELAY     10                // Guard time for failsafe activation after signal lost. 1 step = 0.1sec - 1sec in example
    #define FAILSAFE_OFF_DELAY 0           // Time for Landing before motors stop in 0.1sec. 1 step = 0.1sec - 20sec in example
    #define FAILSAFE_THROTTLE  (MINTHROTTLE - 100)    // (*) Throttle level used for landing - may be relative to MINTHROTTLE - as in this case
    #define FAILSAFE_DETECT_TRESHOLD  985

 

[jyunte]

Just to be certain that we're using the same terms, since I'm new to quads with a "brain"... for me, as someone who comes from fixed wing flying, failsafe is something that happens when the receiver stops receiving a signal from the radio -- having flown out of range, or the Tx died, etc, right? Or is it an "Oh Crap" button/switch that you activate if something goes wrong?

With my frsky X4R/SB RX, I set a custom failsafe mix to put it in horizon mode (hopefully to right it up) and then cut throttle.

Is that the same as just flipping a pre-assigned switch to put it into horizon mode (I have a 3-position switch, for Acro, Angle and Horizon modes) and then chopping the throttle?

On the flip1.5, I just have it cut throttle -- the default is to sort of "autoland" with a minimum throttle run over a time constant. I set the throttle to zero and the time constant to zero as well.

So, basically, with your setup, it would simply kill the throttle, and drop out of the sky?

What wasn't mentioned on the flitetest build video for the electrohub/flip1.5 build, but is talked about in other videos, is dynamic balancing:

http://flitetest.com/articles/Laser_Balancing_Props

I've seen that video, and I don't see how anything they did served to balance the motors. The laser showed that the motor and prop, as a system, was improperly balanced, but adding/removing tape from the prop did not balance the motor, or the prop, it balanced the system as a whole. If a new prop is needed, you couldn't simply add a new balanced prop, you'd have to do the whole laser balancing process again, which doesn't seem like an easy thing to do in the field.

Is there a way to balance the motor itself? I've seen a few Youtube videos where an iPhone is used to measure the vibration, and tape is added to the motor, but I don't have an iPhone! That still seems a bit hit and miss, based on the resolution of the iPhone's sensor. I guess I could use the vibration sensor on the MultiWiiConf screen, like you said? I assume the vibration sensor is the white box with the squiggly lines?

 

[nilsen]

With the Naze32 and cleanflight, if your PID's are way off and there are some vibrations my miniquad sometimes just starts flying away as the vibrations and incorrect PID's have a cascading effect and the only way to get it back is to cut the throttle and then bounce the throttle to try and land it relatively well without it crashing.
Since cleanflight/baseflight are very similar to MultiWii you could be having a similar issue?
I would try and drop the PID values to some lower defaults and try and fly, then you can start your next task of tuning