HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Jan 21, 2013, 08:04 PM
Registered User
Palo Alto, CA, USA
Joined Nov 2000
1,019 Posts
Mini-HowTo
VTOL-Trainer Build Thread

Congratulations R.D. And thanks Lee for the video

Here is the proof of concept:

It's a Quad... it's a Plane ... It's SuperSchizoQuadPlane ! (3 min 37 sec)





And here is the prototype thanks to Jason for the video:

Ran D. St. Clair's VTOL-Trainer (3 min 24 sec)
kasra is offline Find More Posts by kasra
RCG Plus Member
Last edited by kasra; Mar 30, 2013 at 11:29 PM.
Reply With Quote
Sign up now
to remove ads between posts
Old Jan 21, 2013, 08:50 PM
Will fly for food
ilektron's Avatar
United States, CA, Carlsbad
Joined May 2011
532 Posts
Why dont more people build something like this
ilektron is offline Find More Posts by ilektron
Reply With Quote
Old Jan 22, 2013, 12:33 AM
Registered User
VTOLicious's Avatar
Austria
Joined Oct 2007
281 Posts
Quote:
Originally Posted by kasra View Post
Congratulations R.D. And thanks Lee for the video

http://youtu.be/8HQT4XsZBRo
That's pretty awesome, good job! Multiwii+VTOLfirmware?

Regards, Michael
VTOLicious is offline Find More Posts by VTOLicious
Reply With Quote
Old Jan 22, 2013, 08:07 AM
Winging it Ò>
leadfeather's Avatar
Joined May 2006
9,653 Posts
Great job!

Can you share the build details please?
leadfeather is offline Find More Posts by leadfeather
Reply With Quote
Old Jan 22, 2013, 10:44 AM
vtol fan
nadine's Avatar
Joined Jul 2009
627 Posts
Whoooaa....nice plane
nadine is offline Find More Posts by nadine
Reply With Quote
Old Jan 23, 2013, 07:31 PM
Registered User
Ran D. St. Clair's Avatar
United States, CA, Sunnyvale
Joined Dec 2006
917 Posts
Quad-Plane Details

Since you asked for the details...

I am using all 7 channels of my DX7. The plane has 4 small servo’s on rudder, elevator, and both ailerons. It also has 2 HS-85MG servo’s driving the motor tilt on each of the left and right motor pods. The motor mounts are made from 1/16” thick aluminum from the hardware store (no fancy alloy or heat treatment). The tilt bearings are simple 1/4” brass bushings with nylon washers on each side. The motor mounts are the only thing that is a little difficult to make by hand, but they would be a snap to make in production so hopefully one of the hobby distributors will pick up the idea and run with it. The linkages driving the tilt motors are simple pushrods with a standard ball link on the motor end. The motors tilt 90 degrees and the tilt servo’s rotate about 100 degrees. There’s a tiny bit of slop in the tilt mechanism, but it doesn’t seem to hurt anything. There is however a mechanical stop to prevent over rotation and prop strikes on the motor pods. The point of rotation is roughly through the CG of the motor/prop assembly to minimize any shock loads on the servo’s due to a bad landing, etc.

Roll, pitch, and yaw come out of the RX and go straight into the Eagletree Guardian, and then come out of the Guardian and go into Y cables. One side of the Y goes to the ailerons, elevator, and rudder servo’s respectively. The Y cable for the ailerons is a 3Y because it drives 2 aileron servo’s. The other side of the Y goes to the KK2 flight controller board from HobbyKing. Throttle goes directly from the RX into the KK2 board. The Gear channel from the RX goes into the Guardian to select the flight mode. Two separate RX channels drive the two motor tilt servos.

The KK2 board demands a fixed polarity on roll, pitch, and yaw so it is necessary to give it what it wants via the transmitter. If that is the wrong polarity for the aileron, elevator, and rudder servo’s then you will need to use some sort of in line servo reverser. In this case I was able to arrange the servo’s mechanically to avoid this.

The TX is set in “Acro” or airplane mode. The wing type has 2 flap servo’s and one aileron servo. The flaps (motor tilt) is controlled by the 3 position flap switch. The down position is “hover mode” with the motors pointing straight up, or down in the case of the rear motors. The middle position is “slow flight mode” and is programmed to give the motors about 10 degrees of forward tilt. The up position drives the motors to point straight ahead, or straight back in the case of the rear motors, which are pushers. The flaps are set up with a 3 second full travel delay to allow for a smooth transition, and also to reduce any loads on the servo’s due to gyroscopic precession. Driving the 2 flap servo’s with 2 TX channels is necessary because it allows for separate endpoint adjustments to compensate for slight differences in linkages or servo centering.

The Guardian is set to maximum gain on all axes. It is programmed for “2D” mode when in a hover or slow flight mode. In full forward flight mode the Guardian is driven to “pass through” mode, so it is like it isn’t even there. The Guardian mode is controlled by the “Gear” output from the RX. Inside the TX flaps are mixed to Gear. In flight the Gear switch on the TX is always up. The Gear switch is only used to switch the Guardian from 2D mode to 3D mode on the ground as part of a level set initiation sequence. 3D mode is never used in flight.

The KK2 board takes the roll, pitch, and yaw signals from the Guardian and mixes them as needed to drive the 4 motors. It drives 4 each Phoenix-10 motor controllers which drive 4 each E-flite EFLM1200 motors with a KV rating of 1080 RPM/Volt. I am using HD-9050X3 propellers, 9” diameter, 5” pitch, 3 blade propellers from GWS, two clockwise, and two counter clockwise. The motors are rated at 9A continuous and 11A burst current. They probably draw a little more than that at FT so some different choices might be better. In a hover, however, they only draw about 5A each, and in forward flight, even at full throttle the amps should be less. I have never noticed that either the motors or speed controllers are getting excessively warm. The speed controllers are mounted 1/4” above the foam with free air all around. The Phoenix motor controllers are set to fixed endpoints but nothing unusual otherwise.

The KK2 has lots of detailed parameters to be set. The first step is to use the “Receiver Test” mode to set the TX sub trim to zero and the end points to +100 and –100 as seen by the KK2. The KK2 is set for a standard X quad copter motor layout though the parameters of the mixer are tweaked for optimum performance. The throttle offset is set for each motor individually so they all start evenly at just above minimum throttle. The throttle gain for each motor is set by actual hover testing to eliminate any drift in roll, pitch, or yaw. It is also set to make sure that all motors reach full throttle at near full stick deflection. This last part is done in forward flight mode to make sure the Guardian isn’t messing with any of the signals.

Both the Guardian and KK2 board are contributing to stability in the hover and slow flight mode. In forward flight mode we don’t want any stability, but we can’t entirely switch the KK2 board off so we set the stability from the KK2 board as low as possible for acceptable hover stability. The KK2 is set to about 20 “P” and zero “I” on all axes. The guardian effectively provides both “I” and “P” in “2D” mode which is their term for “auto-level” mode. The stick setting menu in the KK2 is set to 60 on both roll and pitch which is about twice the default value. This effectively amplifies the inputs from the Guardian by about 2 X. The yaw is set to 100 for even more amplification, and the throttle is set to 90 which is nominal if I recall correctly. All of this is intended to maximize the stability from the Guardian and minimize the stability from the KK2. That way when we are in forward flight and the Guardian is in “pass through” mode the KK2 is providing minimal stability, because ideally we want none at that point.

In forward flight the KK2 is still providing unwanted roll, pitch and yaw mixing to the motors. When you give up elevator, for example, the forward motors speed up and the rear motors slow down. Other than some unwanted drag this has no real effect on the airplane because all 4 motors are pulling/pushing forward at that point. Likewise, rudder input has minimal effect because one motor on each side speeds up and the other slows down. Aileron is the only real problem. Left aileron causes both motors on the right to speed up and both motors on the left to slow down. This results in a powerful and unwanted yaw to the left, AKA “proverse” (the opposite of adverse) yaw. This is not the worst thing in the world because at least it is in the preferred direction, but it does make rolls look sloppy and is not ideal. It would be better to have the motors go to “throttle only” mode, but that would require some minor code changes inside the KK2. If this form of flight catches on I have no doubt that it will be a standard option in various flight controllers before long.

The Guardian is set to about 50% gain for the “step on the ball” or “anti slip” mode, or as I like to call it, “tail following mode”. One problem airplane pilots have when transitioning to quad-copters or helicopters is that they instinctive expect the plane to turn nose into the relative wind. A quad copter has no nose to speak of and will happily fly in any direction. The application of ailerons will simply cause a quad-copter to slip sideways. The “tail following mode” in the Guardian automatically applies rudder towards the low wing, effectively coordinating turns for you. As a result this plane can be flown with just ailerons in slow forward flight. The turns look better if you use rudder as well, but you don’t need to. In a hover the tail following mode has little effect because the tilt angles are small and not sustained. In forward flight the Guardian is in “pass through” mode so none of this applies.

The Guardian is not set to “heading hold” mode. The heading hold mode uses ailerons to try and hold heading on yaw. (Yes, it sounds odd but that is how it works). It works fine on an airplane, but would not work well in this application. The Guardian also has various other interesting modes and capabilities that I am not using and won’t go into here.

I am using a 2,200 mAh battery and flight times are about 5 minutes with 30% safety margin, longer if you manage the throttle and mostly fly in forward flight mode.

The net result of all this is that the plane is easy to fly in all flight modes. It is easy to hover and will hold itself level. All you have to do is manage altitude with throttle and compensate for any wind. In slow forward flight it flies like the most gentle of trainers, except that it can’t be stalled. It will still hover in slow flight mode, but with the nose about 10 degrees up. It is the hardest to fly in forward flight mode, but still not bad. This version has a ton of drag and is a bit heavy for its size. It has plenty of power, but the low pitch propellers will never allow it to be very fast. There is no reason a more streamlined version can’t be made, but it will still never be very fast, just more efficient.

My thanks to Lee for taking the video. I had not planned a video shoot and this was only about the 5th flight on a brand new plane. This project was designed and built in about 3 weeks, but that doesn’t count the 30 years of trying and mostly failing that came before. There were so many enabling technologies that had to emerge before such a thing could be relatively cheap and easy. I should also point out that I am not the first to turn a quad-copter into an airplane. These guys did a fantastic job.

http://www.youtube.com/watch?feature...QiaYLgIaI&NR=1

Ran D. St. Clair
rd_s_c@pacbell.net
Ran D. St. Clair is offline Find More Posts by Ran D. St. Clair
Reply With Quote
Old Jan 23, 2013, 07:50 PM
Registered User
Ran D. St. Clair's Avatar
United States, CA, Sunnyvale
Joined Dec 2006
917 Posts
Some pictures of the motor mounts

Some pictures of the motor mounts
Ran D. St. Clair is offline Find More Posts by Ran D. St. Clair
Reply With Quote
Old Jan 23, 2013, 08:03 PM
Registered User
Ran D. St. Clair's Avatar
United States, CA, Sunnyvale
Joined Dec 2006
917 Posts
Guardian and KK2 setup files

Guardian and KK2 setup files
Ran D. St. Clair is offline Find More Posts by Ran D. St. Clair
Reply With Quote
Old Jan 23, 2013, 08:04 PM
Retardedly intelligent
foam and tape's Avatar
beavercreek,ohio
Joined Jan 2007
3,854 Posts
Very nice!!!!!! You did what I couldn't do with the QTR. The KK board in forward flight wreaked all sorts of havoc on the airplane. Your design seems to work extremely well rather than fighting the mixing. Thanks for sharing and post more videos.
foam and tape is offline Find More Posts by foam and tape
Reply With Quote
Old Jan 24, 2013, 01:12 PM
Registered User
Ran D. St. Clair's Avatar
United States, CA, Sunnyvale
Joined Dec 2006
917 Posts
The next step

The next step is to turn this thing into something that anyone can do, and make it better of course.

I have ordered a Bixler kit, just the airframe to use as a platform. I will be documenting the design and share it, especially the motor mounts. The rest is all pretty easy as it is just a combination of off the shelf stuff. I am not a multirotor expert so some of you may want to chime in on recommended motors, motor controllers, etc. I know that there is a lot of brand loyalty regarding the different flight controller options and we already know that the Guardian + KK2 is not ideal. Some of you may have better/cheaper options to offer. Custom code could certainly be an improvement. The extra expense of the Guardian could easily be eliminated, especially if you don't care about "tail following" mode. Experienced helecopter pilots may not mind riding the rudder, but it's not very user friendly for those coming from a fixed wing perspective. I am guessing that implementing "tail following" mode in the flight controller code might be a bit more complicated than just turning off the motor mixing in "forward flight" mode.

My goal is to create a true VTOL trainer that everyone can enjoy. It has to be cheap, easy to build, and easy and fun to fly in all flight modes. I think it is entirely possible with current technology. If we can manage that, then Hobby King and others will soon offer it as an RTF and then everyone will have one.
Ran D. St. Clair is offline Find More Posts by Ran D. St. Clair
Reply With Quote
Old Jan 24, 2013, 04:24 PM
Registered User
Joined Oct 2004
2,810 Posts
I remember seeing a full size concept of the same layout some time ago, but can't find it anymore. Those are some very classy motor mounts, btw
Brandano is offline Find More Posts by Brandano
Reply With Quote
Old Jan 24, 2013, 04:25 PM
Registered User
VTOLicious's Avatar
Austria
Joined Oct 2007
281 Posts
Quote:
Originally Posted by Ran D. St. Clair View Post
The next step is to turn this thing into something that anyone can do, and make it better of course...

My goal is to create a true VTOL trainer that everyone can enjoy. It has to be cheap, easy to build, and easy and fun to fly in all flight modes. I think it is entirely possible with current technology. If we can manage that, then Hobby King and others will soon offer it as an RTF and then everyone will have one.
Great stuff! I'm ready to contribute.

I'm working on a similar project and have already designed and build a tilt mechanism that would fit your needs. Just tell me the distance from front to rear motor pivot and I will provide a 3D drawing and corresponding files for laser cutting of plywood parts. See pic attached!

Michael
VTOLicious is offline Find More Posts by VTOLicious
Last edited by VTOLicious; Jan 24, 2013 at 04:36 PM.
Reply With Quote
Old Jan 24, 2013, 04:27 PM
Registered User
United States, TX, Cypress
Joined Jan 2012
895 Posts
Subbed

Nice build
corsair2014 is offline Find More Posts by corsair2014
Reply With Quote
Old Jan 24, 2013, 04:30 PM
Registered User
VTOLicious's Avatar
Austria
Joined Oct 2007
281 Posts
Quote:
Originally Posted by Brandano View Post
I remember seeing a full size concept of the same layout some time ago, but can't find it anymore. Those are some very classy motor mounts, btw
Have a look here:http://www.rcgroups.com/forums/showp...&postcount=159
VTOLicious is offline Find More Posts by VTOLicious
Reply With Quote
Old Jan 24, 2013, 05:59 PM
Registered User
Ran D. St. Clair's Avatar
United States, CA, Sunnyvale
Joined Dec 2006
917 Posts
Quote:
Originally Posted by VTOLicious View Post
Great stuff! I'm ready to contribute.

I'm working on a similar project and have already designed and build a tilt mechanism that would fit your needs. Just tell me the distance from front to rear motor pivot and I will provide a 3D drawing and corresponding files for laser cutting of plywood parts. See pic attached!

Michael
VTOLicious, Your design places the motor at the end of a lever arm of an inch or two. In the event of any kind of shock, like a bad landing, it will pound back into the servo output. Depending on the strength of the servo output gears, weight of the motor, etc. The servo may not be able to take it. Your design does have the advantage of raising the prop farther away from the boom in hover mode so a prop strike on the boom is pretty much impossible. Unfortunately, in the back, where the motor has to rotate down your design will drop the propeller down into the grass. I am already cutting the grass with my rear prop in hover mode as it is. Dropping it down another inch or 3 would be a problem. You could just make the landing gear longer, but then it gets wobbly, and causes other design problems.

Some designs, like a tri-copter with the rear motor projecting forward off the vertical stabalizer, all the motors rotate up, so grass cutting would not be a problem.
Ran D. St. Clair is offline Find More Posts by Ran D. St. Clair
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Another Success Story for the Obama Administration. LcJ Life, The Universe, and Politics 77 Jan 20, 2013 12:05 AM
Discussion Another F35 trying to VTOL! capt VTOLs 13 Feb 07, 2010 12:59 PM
Discussion EDF vtol..most successful vortices VTOLs 4 Mar 24, 2009 10:44 AM
Discussion Yet another VTOL Bobnormal Foamies (Scratchbuilt) 29 Oct 27, 2008 05:50 AM
Vtol Plane Success vtolman Electric Plane Talk 34 Jan 11, 2006 01:01 PM