Split Elevon Wing Test

[JasonK]

This project is to test an idea for getting yaw control on a flying wing that has a single motor.

I am splitting the Elevons into inner/outer bits to try creating a differential drag situation that can be used for yaw control.

CG calculation
Cutting Template is attached

 

[JasonK]

Build Process & Notes

• Initial cutting
  • Realized that the top and bottom most bits need to 5mm or so remove from the front edges (not in the template) as after beveling the other layers, the front edge they are lining up with will have been moved back that much.
• Remove paper between the foam layers
  • I removed 1 side of every piece except the middle most layer, and templated the smaller layers on the bigger ones to remove all the overlapping paper, so all the gluing was foam to foam
  • Foam weight after removal of all paper but before gluing/beveling/triming: 72g
• Gluing layers together
  • I used White Gorilla Glue. I realized that this is a prototype, using hot glue for speed could have been a better choice for build speed. When laminating the layers together, working out from the middle layer, alternating top/bottom would have made clamping the layers easier while letting the glue dry.
• Iron Elevon joins
  • I only irons the joins, this is instead of bevel cutting and hot glue reinforcing, I didn't iron the other edges as this is a protoype (ironing also strengthens the edges)
• Cut out servo slots
  • If this gets past the prototype phase, these should be added to the plans and cut 1 or 2 layers deep (1 layer leaves the servo sticking out, 2 will over sink them. Trade off of dealing with the servo being deep or messing with airflow)
• Print out control horns and Motor mount
  • This could be done with hobby plywood or any other reasonable solution
• Install servos, control horns, motor mount
  • using hotglue instead of gorilla glue - had bad experience with gorilla glue getting inside the servo and destroying it.
• Cutout slots for ESC, wires, battery, etc and install
  • forget to take pictures of the process, thankfully it is mostly viable from the resulting image.
• Program up Flight modes and mixes to be able to test.

 

[JasonK]

Maiden
Took full up trim and almost full stick deflection to fly, need to figure out if it is a throws issue or a CG to forward issue. Both 'normal', where the elevons moved as one, and split, were there was an 80/20 split on movement [inner more elevator, outer more aileron], flew well. given the extreme level of elevator input, I don't feel the yaw test was valid as the inner servos were close or completely saturated, leaving no additional movement for the yaw input to effect it.

Second maiden flights of my split elevon wing attempt. I swapped out the ~101g battery with one that was about 40g bringing the CG back 3cm or so from the recommended situation. The wing was still very stable in pitch and didn't need nearly as much up-elevator to fly straight. Unfortunately it was rather draggy in the air and if I cut the throttle completely it would descend very fast. Definately has a fairly poor glide slope.

Unfortunately the yaw control didn't do anything useful. However the wing itself flew fairly well. The flags at the edge of the park were flying full and coming in over the southern trees (I noted at one point in the video which ones) and I am quite sure there is a band of significant turbulence in the area.

 

[JasonK]

Final thoughts:

• the 5 layer symmetrical kfm wing ended up very draggy (it did have the thickest part of the airfoil right at the front edge of the wing, instead of ~30% back which seems to be the best place for the thickest part), so it flew fairly slow and stable, but fell almost like a brick if left to glide.
• DTFB de-laminates much easier then ROSS brand foam and is basically the same weight. (the store I get the ROSS from was out and I restocked with DTFB, definitely easier to pull the paper off, which is useful when you need to, but disadvantageous when you don't want to)
• the yawing motion didn't work, or if it did, it was so minor as to not be useful/relevant
• when I dialed up the yaw mix %, it removed control from the plane (which makes sense, it saturated the servos)
• the kfm wing was really easy to build and was very sturdy
• I should have used an A-pack sized motor on 3s with the 5x3 prop instead of the 1306 3100kv motor that I used. that motor got hot the first time out with being at 100% throttle the whole time.
• I could try moving the battery farther back, it was still rather stable on pitch, so I think there might be more room to move the CG back, however I don't think this would help any with the yaw issue, just possibly getting a better flight for what it is.
• I probably could have got air brakes out of the split elevons, but didn't try it.

This is related to what Flite test did years back with trying to put rudders on wings

 

[L Edge]

This project is to test an idea for getting yaw control on a flying wing that has a single motor.

I am splitting the Elevons into inner/outer bits to try creating a differential drag situation that can be used for yaw control.

CG calculation
Cutting Template is attached

How are you controlling the amount of deflection for the differential drag?

 

[JasonK]

How are you controlling the amount of deflection for the differential drag?

mixing in my TX (I have an OpenTX so setting up all sorts of complicated mixes are easy).

I will setup multiple flight modes, so I will have one that is a 'traditional' config, so I can flip to that if something goes odd.

I do question if this has any chance of working, it is going to be a fly -> test, repeat to see if I can work out a mix setup that actually does what I want.

 

[L Edge]

Have you thought about using aileron differential to correct roll to yaw coupling and adverse yaw characteristics?

 

[JasonK]

Have you thought about using aileron differential to correct roll to yaw coupling and adverse yaw characteristics?

I have thought about all sorts of mixing... need to fly it, see what happens, then adjust.

 

[JasonK]

bump, second maiden outing and final thoughts added.

 

[Scotto]

Cool experiment. I thought it would yaw a lot with all that drag. Wonder why not. Could you share your program from the opentx?

 

[JasonK]

General

Name	Arrow
Model Image	Delta1.jpg
Throttle	Source: THR Trim idle only: No Warning: Yes Reversed: No
Trims	Step: Fine Display: On Change Extended: No
Center Beep	None
Switch Warnings	SA↑ SB↑ SC↑ SD↑ SE↑ SF↑
Pot Warnings	Mode: OFF
Other	Extended Limits: Yes Display Checklist: No Global Functions: Yes

Timers	Time	Switch	Countdown	Min.call	Persist
Tmr1	00:05:00	L02	Voice	Yes	Flight
Tmr2	00:00:00	OFF	Silent	No	OFF
Tmr3	00:00:00	OFF	Silent	No	OFF

Modules

Internal Radio System	Protocol: DIY Multiprotocol Module Channels: 1-16 Receiver: 0 Radio protocol: FrSky Subtype: D8 Option value: 0
External Radio Module	Protocol: OFF
Trainer port	Mode: Slave/Jack Channels: 1-8 Frame length: 22.5ms PPM delay: 300us Polarity: Positive

Flight modes

Flight mode	Switch	F.In	F.Out	TrmR	TrmE	TrmT	TrmA	Trm5	Trm6
FM0:Normal	----	0	0	0	0	0	0	0	-8
FM1:4WNoYaw	SB-	0	0	0	0	0	0	FM0	FM0
FM2:4WYaw	SB↓	0	0	FM1	FM1	FM1	FM1	FM0	FM0
FM3	----	0	0	FM0	FM0	FM0	FM0	FM0	FM0
FM4	----	0	0	FM0	FM0	FM0	FM0	FM0	FM0
FM5	----	0	0	FM0	FM0	FM0	FM0	FM0	FM0
FM6	----	0	0	FM0	FM0	FM0	FM0	FM0	FM0
FM7	----	0	0	FM0	FM0	FM0	FM0	FM0	FM0
FM8	----	0	0	FM0	FM0	FM0	FM0	FM0	FM0

Global vars	GV1	GV2	GV3	GV4	GV5	GV6	GV7	GV8	GV9
Name	A	B
Unit
Prec	0._	0._	0._	0._	0._	0._	0._	0._	0._
Min	-1024	-1024	-1024	-1024	-1024	-1024	-1024	-1024	-1024
Max	1024	1024	1024	1024	1024	1024	1024	1024	1024
Popup	N	N	N	N	N	N	N	N	N
Flight mode
FM0:Normal	50	50	0	0	0	0	0	0	0
FM1:4WNoYaw	80	20	FM0	FM0	FM0	FM0	FM0	FM0	FM0
FM2:4WYaw	FM1	FM1	FM0	FM0	FM0	FM0	FM0	FM0	FM0
FM3	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0
FM4	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0
FM5	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0
FM6	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0
FM7	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0
FM8	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0	FM0

Inputs

I1:Ail	Ail Weight(+100%) Expo(30%) Switch(SA↑) Ail Weight(+70%) Expo(30%) Switch(!SA↑)
I2:Ele	Ele Weight(+100%) Expo(30%) Switch(SA↑) Ele Weight(+70%) Expo(30%) Switch(!SA↑)
I3:Thr	Thr Weight(0%) Switch(SF↑) Offset(-100%) Thr Weight(+100%)
I4:Rud	Rud Weight(+100%) Expo(30%)

Mixers

CH1:ElAO	I1:Ail Weight(-GV1:A) [AilR] += I2:Ele Weight(+GV2:B) [Elev] += I4:Rud Weight(+GV3) Flight mode(FM2:4WYaw) Func(x>0) [YawL]
CH2:ElBO	I1:Ail Weight(-GV1:A) [AilR] += I2:Ele Weight(-GV2:B) [Elev] += I4:Rud Weight(+GV3) Flight mode(FM2:4WYaw) Func(x<0) [YawR]
CH3:Thr	I3:Thr Weight(+100%) [Motor]
CH4:ElAI	I1:Ail Weight(-GV2:B) [AilR] += I2:Ele Weight(+GV1:A) [Elev] += I4:Rud Weight(-GV4) Flight mode(FM2:4WYaw) Func(x>0) [YawL]
CH5:ELBI	I1:Ail Weight(-GV2:B) [AilR] += I2:Ele Weight(-GV1:A) [Elev] += I4:Rud Weight(-GV4) Flight mode(FM2:4WYaw) Func(x<0) [YawR]

Outputs

Channel	Subtrim	Min	Max	Direct	Curve	PPM	Linear
CH1:ElAO	+0.0%	-100.0%	+100.0%	NOR	----	1500	N
CH2:ElBO	+0.0%	-100.0%	+100.0%	NOR	----	1500	N
CH3:Thr	+0.0%	-100.0%	+100.0%	NOR	----	1500	N
CH4:ElAI	+0.0%	-100.0%	+100.0%	NOR	----	1500	N
CH5:ELBI	+0.0%	-100.0%	+100.0%	NOR	----	1500	N

Curves
Logical Switches

L1	Edge(SH↓, [0.5:0.5])
L2	( I3:Thr > -90 ) AND SF↓
L3	( Edge(SH↓, [0:0.5]) ) AND SC↑
L4	( Edge(SH↓, [0:0.5]) ) AND SC↓

Special Functions

SF1	L04 - Instant Trim (0)
SF2	L01 - Reset (Timer1)
SF3	SC↑ - Play Track (fm-7)
SF4	SC↓ - Play Track (fm-8)
SF5	SF↑ - Play Track (disarm)
SF6	SF↓ - Play Track (armed)
SF7	L03 - Play Value (TMR1)
SF8	SB↑ - Play Track (fm-1)
SF9	SB- - Play Track (fm-2)
SF10	SB↓ - Play Track (fm-3)
SF11	SA↑ - Play Track (fm-acr)
SF12	!SA↑ - Play Track (fm-nrm)
SF13	ON - Adjust GV3 (S1)
SF14	ON - Adjust GV4 (S2)

Telemetry

Protocol	FrSky S.PORT
RSSI Alarms	Low: < 39; Critical: < 42; Telemetry audio: Enable
Altimetry	Vario source: None
	Vario limits > Sink max: -10; Sink min: -0.5; Climb min: 0.5; Climb max: 10; Center silent: No
Top Bar	Volts source: None; Altitude source: None

 

[JasonK]

Cool experiment. I thought it would yaw a lot with all that drag. Wonder why not. Could you share your program from the opentx?

I started with something around +50 on S1 and S2... when that didn't work I turned them all the way up and tried again (which caused the crash in the video)

right here is where I turn up the rudder gains:

and it ended up crashing shortly after... the +100% on the rudder would have basically saturated the servos, on the left side.

 

[Scotto]

You could try using an iron to smooth it out and thin that front ~25%. Maybe less overall drag would make the differential more noticable.

 

[L Edge]

It would be interesting to see some flights when the wind(gusty) is blowing.

 

[JasonK]

It would be interesting to see some flights when the wind(gusty) is blowing.

you mean like every single flight in that video? the flags were full at that field... the weird red block should have been text pointing out when the FPV feed was facing the trees that the wind was coming from. The wind comes over the tree tops and causes quite a bit of turbulence ( which is basically always a given at this location... the only variation is how bad it is).

here is a picture I took to show what the wind was like:

 

[L Edge]

I tried the aileron differential(no programming needed) which generates uneven drag just on the wing and can be used only with elevons. I tried both positive and negative values with no such luck. Was hoping you could succeed.

You showed the Flitetest video above about adding rudders to the wing. It can be done a different way and this is my "Udder Rudder" which is centerlined and works well in wind and gusty conditions. By adding controllable yaw to the roll and pitch axis, it now makes a wing act like any other airplane. Landings with crosswinds is a breeze. That is 1/2 of the problem with wings.

Explaining "Udder Rudder"

https://forum.flitetest.com/index.p...-new-tools-to-enjoy-flying-in-the-wind.67067/
\

 

[JasonK]

I tried the aileron differential(no programming needed) which generates uneven drag just on the wing and can be used only with elevons. I tried both positive and negative values with no such luck. Was hoping you could succeed.

You showed the Flitetest video above about adding rudders to the wing. It can be done a different way and this is my "Udder Rudder" which is centerlined and works well in wind and gusty conditions. By adding controllable yaw to the roll and pitch axis, it now makes a wing act like any other airplane. Landings with crosswinds is a breeze. That is 1/2 of the problem with wings.

Explaining "Udder Rudder"

https://forum.flitetest.com/index.p...-new-tools-to-enjoy-flying-in-the-wind.67067/
\

aileron differential is mostly to prevent adverse yaw from rolling, I did expect this to get some yaw, but it could also be being over ridden by other aspects of the air-frame stabilizing it. without the split, you can't add drag without effecting roll/pitch.