PDA

View Full Version : Force on the yoke?



spitfire9
07-20-2007, 02:14 PM
I've been thinking about force feedback for a large Boeing style yoke.

I've come to the conclusion that it is not worth the effort,, seeing the ffb effects are minimal in Fs9,to say the least.

Please bare with me as I try to put my idea to words.;)

I was wondering,,if anyone has done anything like this.

Have some kind of mechanical interface , that somehow tightens the spring/hydraulic force on the yoke, dependent on airspeed.

IE : an electric servo motor that will spread the centering springs on the yoke. The faster the airspeed, the more the springs are spread/tightened.
This would simulate the heavier control feel at faster airspeeds and inversely, the sloppier lighter feel at the slow flight range.

Maybe,,, somehow,,, a simple interface with the airspeed value or a/s indicator that would drive or back up a servo motor to stetch or slack off the centering springs of the yoke????

I think this would be as good or better than a ffb stick or at least a lot easier to accomplish.

If anybody has tried something similar or has any ideas on this,, your input would be greatly appreciated.

Thanks.

Mike.Powell
07-20-2007, 04:44 PM
The topic has come up over the years. A few people has experimented with some degree of success.

"Bricklayer", who posts on ViperPit put a substantial amount of effort into adapting an automobile power steering unit for use as a control loader for a joystick. I believe it was targeted for use in a Spitfire simulator being built in the UK.

"Tigert", who posts on Avsim, built a control loader based on a pendulum. (IIRC, he used a sledge hammer as a ready-made pendulum). The yoke was mechanically connected to the pendulum. The more you moved the yoke, the farther the pendulum moved from vertical and the greater the perceived restoring force. He used a traveling nut to shift the pivot point of the pendulum. By using a motor on the threaded shaft supporting the nut, he could adjust the proportionality of the force to yoke movement. I think the results were mixed: nice force effect, but the motor turning the screw was too noisy or vibrated too much.


Tightening the centering springs intuitively seems like it will work. Unfortunately it does not. Each individual spring will have more tension, but it's cancelled by the opposing spring. Ultimately it's the spring constant that determines the force versus position function. If you can change the mechanical advantage of the spring configuration, you can change the F vs P function.

I did some wild hand waving about electric control load last year. I wrote about it here: http://www.mikesflightdeck.com/old_stuff.htm It's almost at the bottom of the page. You might find it interesting.

I'm a fan of control loading. Sadly, there are just too many things to work on. (To be very specific, my official, number one project according to the chief, is fixing the landing in front of the front door before someone falls through. Sigh!) Eventually I'd like to work on control loading. It's just a bit low on the list these days.

ivar hestnes
07-20-2007, 05:38 PM
I have an idea for this which is self centering and using only one spring. Cant explain in words. I will try to make a drawing and post here.

I plan to use this mechanism for my yoke and pedal setup. Have not built it yet but are very sure about it will work fine;)

Will try to draw a simple drawing tomorrow;)

Please kick my butt if I forget:D

ak49er
07-20-2007, 07:08 PM
Of course this may take some money , but what about stepper motors, (i.e. positioning motors, with or without variable frquency drive) They would act as servos for the AP, and would allow for "contol loading forces" by variable resistance to physical input, would also self center, and would react like CWS (control wheel steering) during manual maneuvers while on AP. By the way Ivar, could you send me plans for your cockpit framework?

NicD
07-20-2007, 07:11 PM
We're about 70% done with ours - details here: http://www.insitesolutions.com.au/737project/design_controls.htm (see link at bottom of text) and early pics here: http://www.insitesolutions.com.au/737project/construction_diary_04.htm

I agree with Mike - this is difficult to get right. We've got more work to do to add more general elevator load, but we won't be varying it with airspeed. Opposing springs and dampers is a black-art and seems to defy logic!

We've got the mechanism in place to move the trim rack (which changes the conrol load, allowing a trimmed column position) but haven't found a motor solution that gives smooth and non-noisy operation yet. In the too-hard-basket for now while we finish other stuff.

Westozy
07-20-2007, 09:26 PM
I have already sorted self centering on my yokes successfully with 2 screendoor closers but have another idea about creating adjustable forces. I am thinking about making a small 'hydraulic' cylinder with a ram, (imagine a bicycle pump that can also pull air). It will be connected to my pitch motion linkages. When the ram is being pulled, oil is forced through a tube looped back into the ram behind the chevron and vice versa when it is pushed, the flow reverses. this becomes a simple speed hydraulic controller frequently used in my industry. I'll draw one and post it here later. I can make it out of PVC or there maybe something already available to use. With a needle valve on the bypass tubing the speed becomes adjustable.
Now if you very clever PC tech dudes can make a servo controlled by IAS turn a small needle valve, you have a very simple variable column loading device. I was going to make my device for a fixed load, I never thought about variable/IAS - good idea!

Gwyn

kris
07-21-2007, 11:35 AM
I have built basic control wheel force feedback by ripping apart a Logitech USB force feedback racing car wheel - it auto-centres, shakes during turbulence and you can feel the landings! You can set the level of force, and it'll pull itself out of your hands if its too high! It cost me 15 GPB from Ebay!!!

You also get a set of pedals with another 2 axes, and a whle load of buttons to play with.

With regard to pitch feedback, my throttle quadrant will have motorised trim wheels, using DC motors (IOCards), so I presume I'll be able to use a similar setup for the control column - don't quote me on that though!!!

spitfire9
07-21-2007, 12:48 PM
Lots of great stuff to think about in all these replies.

I still have to go through all of it but it sure has me thinking.

Kris,,, how heavy is your yoke? Did you have to use bigger motors to drive the ffb ???

Westozy,, please keep us informed on your idea. It seems simple enough if you already have the fixed load device set up.

Mike,,, I have a friend that is a good electrical guy and he is already looking at a website that sells all sorts of electrical clutches and he seems to think that it is very do-able.

Thanks for all this guys and please keep the ideas flowing.

ivar hestnes
07-21-2007, 04:25 PM
As promised, I have made some drawings of my idea. This is only an idea and no calculations have been made. Hope you understand my drawing;)

I am not an autocad-expert so I prefere to draw by hand, sorry for that.

http://i31.photobucket.com/albums/c380/vrinsk/Centeringdevice.jpg

http://i31.photobucket.com/albums/c380/vrinsk/Tensionadjustment.jpg

kris
07-21-2007, 08:06 PM
Kris,,, how heavy is your yoke? Did you have to use bigger motors to drive the ffb ???



No, the forces from the motor in the steering wheel are huge! I have only mounted it on a test rig so far, but I don't think it'll all fit inside a control column - I'll have to use chains, pulleys etc. It should be able to drive dual linked yokes if my ideas work!

AndyT
07-21-2007, 11:52 PM
Now if you very clever PC tech dudes can make a servo controlled by IAS turn a small needle valve, you have a very simple variable column loading device.

That's actually pretty easy using a phidgets servo controller linked to the IAS as a percentage function.

Westozy
07-22-2007, 12:01 AM
This is what I am thinking of making...

Gwyn

AndyT
07-22-2007, 03:05 AM
Correct me if I'm wrong but would'nt that just load the yoke on one side? Up or down on the pitch? You need something that goes both directions like a 'Friction Loader'. That way, no matter which way you turn it you get a loading. If you have 2 small motorized rubber friction balls (1 for pitch and 1 for roll) and control how hard they push against the yoke with a small servo or DC motor. This way you could also have them 'fight' you for a return to center.

PaulEMB
07-22-2007, 04:06 AM
Gwyn,

This hydraulic idea should work fine - it was suggested to me by Dieter from Routech druing his visit a few weeks ago - he has built several sims and says this method is very realistic.
I am currently rebuilding my contols to fit the new Routech yokes, so will keep you posted.

Westozy
07-22-2007, 06:35 AM
Correct me if I'm wrong but would'nt that just load the yoke on one side? Up or down on the pitch? You need something that goes both directions like a 'Friction Loader'. That way, no matter which way you turn it you get a loading. If you have 2 small motorized rubber friction balls (1 for pitch and 1 for roll) and control how hard they push against the yoke with a small servo or DC motor. This way you could also have them 'fight' you for a return to center.

My device would provide resistance to movement in both directions, the sketch is almost a schematic. It is a basic damper.

Gwyn

spitfire9
07-30-2007, 02:44 AM
AndyT,, I'm not quite sure what you mean??

Could you please elaborate .

Sounds very interesting.

spitfire9
09-07-2007, 01:57 PM
Anything new on this concept ?

I'm going to start on the yoke part pretty soon and was wondering,, what kind of interface card could be used to run a small electric clutch with relation to airspeed ?

I was probably going to go with the FDT cards .

ve4anc
09-07-2007, 03:27 PM
Spit:

I'm back and into the action.

Last night while reviewing the new posts, I noticed the post about motion systems. There is a link to this pretty neat little item ... KT-5198 ... it is a bidirectional DC motor controller and handles fairly high current with some simple printed circuit board mods and heat sinking.

It would be perfect for taking the ffb motor voltage output of say .. a MS Force Feedback 2 joystick and amplifying its output to drive much larger DC motors which in turn are connected to the control column.

http://www.oceancontrols.com.au/motor_controller/dc_motor_controllers.htm

Will talk to you about this over the weekend, as we're past due for another talk about what's on the go.

All the best for now,
Lee