• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

MultiWii Flip 1.5 Airplane Flight Stabilizer

#1
Hey everybody!

I managed to get a Flip 1.5 that I purchased a while back from RTFQ to work as a flight stabilizer. I'm no programmer and I fumble around at best in arduino sketch files. None the less, I managed to make it work.


Let me first say that I originally tried to get a similar setup working using a KK2 board Flashed with OpenAero2 and had only minor success, the board is far too sensitive to vibration and it would test fine on the bench but as soon as throttle was applied the controls would oscillate terribly. I have found vibration dampeners and adding weight to the board (thanks to another user) helped dramatically but all that can't compare to the light weight smooth function I have found with the MultiWii board.

I did not try to set this up using PWM inputs from the receiver. I used a FrSky D4R-II setup in PPM mode, it will do 8 ch output but the Flip 1.5 will only accept 6 incoming channels. Aileron, Elevator, Throttle, Rudder, Aux1, Aux2 - both the Aux channels were setup to control flight modes.

I did lots of google searching for a "How To" and didn't find one, I hope to refine this information to eventually turn it in to an article but I'm afraid it will still only benefit a limited audience using PPM Receivers.

I started here: http://fotoflygarn.blogspot.com/2012/03/multiwii-airplane.html

To clarify some things I felt were not clear...

1.) The Rx will receive power from the MultiWii board. You don't need a separate BEC. You connect your ESC and Servos to the assigned pins and the Rx in PPM mode connects to the Throttle IN and will receive power from the FC.

2.) There are 2 wing servo outputs, HOWEVER, everything will work just fine if you run your aileron servos off of a Y cable.

3.) Just when you're freaking out because your motor won't spin up but everything seems to be connected properly you'll need to ARM the board to get any throttle output (just like you did on your quad). When you start freaking out again because it won't ARM; you need to have 100% Rates setup on throttle and rudder to activate the Arming function. then it will work ;)

4.) For Taranis users, if you are using a PPM Rx like the D4R-II which is a 4ch PWM Rx, make sure you change the channel range on your radio from 1-4 to 1-8 or you will not get any output for channels 5 and 6 and thus have another freak out as to why you can't change flight modes.

5.) I adjusted the Gyro filters to 20hz eliminate the servo feed back from the motor vibration it helped but wasn't perfect, until I fly it I won't know how this will effect the controls. there is a Low Pass Filter in the sketch file. It was necessary to uncomment that line as well. with the LPF enabled I may be able to adjust the gyro filter up (not sure if it's even beneficial). If flight performance is poor that will be my first adjustment.

6.) MultiWii config is slightly misleading when configuring your servos. If gyro moves servo in the wrong direction correct it in Multiwii config if the Transmitter moves the servo in the wrong direction reverse it in the transmitter.

7.) When choosing a flight mode GYRO (stabilize) mode is not on the list. This is the option that is selected if NOTHING else is checked in MultiWiiconf. To program my panic button I used the SH switch for PANIC and SG for flight modes. Passthrough is selected in Low position, nothing selected for MID (which is actually gyro mode) and Horizon mode for High. ANGLE mode is selected when SH is in High position. I don't know how the board will assign priority if two modes are selected at the same time SO.... I used a special function on the Taranis Radio that when SH is placed in high position it will override CH5 (SG) to 0 (mid position, gyro mode, nothing selected) Forcing Angle mode to be the only ACTIVE mode. SH is spring loaded and will not stay on with out constant pressure, which makes it perfect for a panic button.

My pinout for all controls on my FLIP 1.5 from RTFQ is as follows:

D2: PPM Input from RX - remember DO NOT POWER YOUR RX DIRECTLY OR WITH A UBEC, it will get power from the board
D12: Wing1 - I assume this is the Left aileron I'll know for sure when I get this hooked up to my spitfire and will update.
D11: Wing2 - Right Aileron?
D10: Elevator
D9: Throttle
D3: Rudder

I currently have the BloodyWonder's solo aileron servo plugged in to D12


HERE is the sketch file for Arduino. View attachment MultiWii_Airplane.zip NOTE!!! This sketch file includes enabling the Baro/Mag on my board you will need to comment those lines out or you will get an error when uploading. As far as I have found the baro/mag are useless on an airplane with out a GPS.

I dunno if it's completely necessary but I used MultiWiiConf 2.4 (vs 2.3) to configure all my settings on the computer.


Lastly your MultiWiiConf program allows you to adjust all the PID settings just like your quad. All i have done was make the necessary changes to get the airplane servos/motor to work. I'm running Pauls stock PID settings (i'm not sure if this is a good or bad thing yet) I imagine you could tune your air frame similarly to a MultiRotor and start with I, D at zero and turn up P until you get oscillations. again... needs flight testing before I go that far. I throttled up the motor as I changed the filter settings and eventually got ZERO servo feed back as the throttle was Ramped up from min to max. for the time being I'm satisfied.

Let me know what questions you have and I'll try to improve this information as I'm sure there will be plenty of people upgrading to the 32bit FC boards as they gain in popularity, keep that Flip board out of the trash and setup an Autolevel trainer for a family member.

There is additional information about using a GPS, this could be even cooler, as it can provide PH and RTH functions, as well as some failsafe functions.

- Jes

PS along with my Naze32 I have a Sparky board coming. this is something else that there is information on but not a good start to finish How to. I hope to cover that as well.

PSS - Once i get my flight in with the FC installed I'll post a follow up video and the official success or failure of using the MWC 1.5 as a flight stabilizer.

PSSSS - as soon as I get some time I'll get some screen shots and try to place them throughout this post to help some things make more sense.
 
Last edited:
#2
Ok just a quick update. It flew. Unfortunately my hat cam video was too low on my head and I looked like I was tracking an invisible jet. The performance was sluggish. The winds were 15 mph with gusts of probably around 20 this may have just been too much for the whole setup abut it was good to test it I suppose. I have some wobbles but only when pushing hard in to the wind. I had great cross wind performance where often the wing would get blown over the FC kept the plane tracking nice and straight.

I'll do some PID tuning and see how it goes. I plan to tune it similar to a quad but I expect this I will be tedious task. I'll try to post any relevant updates.

- Jes
 
#4
RCG Forum for MultiWii airplane: http://www.rcgroups.com/forums/showthread.php?t=2078096
PatrickE's MultiWii Airplane setup page: http://fotoflygarn.blogspot.com/2012/03/multiwii-airplane.html


I have since reviewed the the pagse above and found some useful info. Both places seem to be focusing more about getting GPS and RTH configured on a multiwii, supposedly you can attach a GPS to the IC2 port on the board. I'm looking for the stabilization features and autolevel (for a trainer and a panic button), I have a sparky board I'll be working on RTH and GPS ... but that a different thread.

So I did several more test flights, i brought my computer to the field and made several adjustments to the PID settings to try to improve the sluggishness of the plane. I really didn't have any luck. Lower P values only made it slower to respond and seemed to further limit the amount of throw in the controls. Too high of P and it would just rock even more. there was no sweet spot to be found.

Tuning a flight controller seems to be just as much of an art as a science. There are so many variables, if the first parameter isn't right the effects will snowball to each additional parameter... that being said, I'm reading that I should not be changing the Low pass filter and stick to the default 256 setting. and I should also refrain from using the gyro smoothing, the methods the flight controller uses can cause the rocking problem and the sluggishness that I have been experiencing. It was highly recommended to use a reliable method of vibration dampening so I will add that to the system before I test any further.

In addition for flying wings and smaller aircraft it seems that reducing the the settings in half improves the flight characteristics. But MANY people are reporting good results with Bixler type/size planes using the stock PID values from the firmware.

I just ordered some HK orange foam which seems to give really good results for vibration dampening for video systems and flight controllers, once I have that installed I'll do some more testing and report back.

- Jes
 
#5
IT WORKS!!

OK.. Last update to this thread.

I was over thinking the configuration .. I ended up disabling the low pass filter setting which defaults the board to 256Hz I disabled the gyro smoothing also. I cleared the EEPROM on the board and loaded the new sketch file leaving the default PID settings and hit the field. IT WORKED GREAT!!!

sorry for the poor quality I pushed it to youtube from my phone right after recording, I'll do a proper edit and post in a new thread, along with the sketch file hopefully sooner than later.


- Jes
 
#9
Sketch File update...

View attachment MultiWii_FlyingWing.zip

that's the updated sketch file I'm using, I had decent success on the bloody wonder however I had a decrease in throws when enabling stabilized modes. I counteracted this by using the same switch to also increase the rates on the throws it helps but not perfect. I asked around and likely the issue is in the PID settings, supposedly you can tune the plane just like a multirotor, starting with P and so on. I have yet to try this, recommended is a Bluetooth adapter to avoid opening/closing the air frame for every adjustment. anyway be sure your 3+ mistakes high and keep your flight mode switch easily accessible, my first trial runs with Open Aero were terrible, enabling stabilized modes would send the plane in to a nose dive. Just be ready for the unexpected and to switch back to manual/Pass-through if needed.

I've since transferred this to a Baby Blender and it did not perform as well, in fact it was unflyable in stabilized modes. there was so much wobble in the air frame and control response was slow/sluggish. obviously each air frame will have it's own specific tuning needs.

On a side note... I've picked up a taulabs board and transferred/converted the multiwii settings to the taulabs PID settings and to no surprise I'm still seeing a decrease in throw. I've got a Bluetooth module finally and I'm going to try to tune the taulabs board, if it works well I'll post my procedure/results/settings (hopefully wont take 2 months this time).

as of right now if you are cruising around or FPV flying my setup has been awesome, but if you're in need of aerobatics with that locked in stabilization it still needs to have the bugs worked out. I'm certain it's a setting i just haven't identified it yet... probably should look in to sluggish multirotor controls and try some fix for that likely this same thing....


anyway new position at work and a new baby has had me busy, going on vacation soon as well... Please keep this thread alive if anyone is trying it out.

- Jes
 

makattack

Winter is coming
Moderator
Mentor
#11
Yes, fixed wing flight controllers also have to be tuned in a similar way with multirotors -- PID's where you are essentially trying to get the P high enough so you have good control, but not so much that you get oscillations. In my experience the D value is less important, and is most useful for getting the P constant higher.

I've also found that with fixed wing, your airspeed and throttle makes a bigger difference. For example, if you tune at about 50% throttle right up to the edge of oscillations, but start flying close to 100% throttle, you might get more oscillations.

I've also found in my case, if I forgot to remove the pitot tube cover (to an airspeed sensor) it really confuses the controller and can give oscillations in any of the auto modes.