Joystick pitch trim force by mechanical means

Hi

I fly a Eurostar 3 axis microlight when I can afford the lessons, and keep my brain in gear with regards to procedures and RT using FSX. I've built pedals, a throttle and various rotary encoders and switches using a bodnar board. My joystick is a bog standard USB.

After doing an exercise involving power climbs and descents, I went back to the sim and found it difficult to follow the exercises due to the unrealistic pitch trim.

When flying, if you've got the trim wrong, you feel it in the stick, holding an out of trim Eurostar hurts your arm. correcting the trim brings the stick forces back to a balanced state. Trying to replicate this in the sim seems to need a force feedback joystick and some software. Perhaps my mechanical ideas will work - shoot me down if it won't work.

A gimbal for the stick with potentiometers for sensing. Beneath this, a channel, aligned in the aileron axis, where the bottom of the stick is free to move left and right. Compression springs in the channel give aileron self centering.

The channel is mounted on sliders in the pitch axis, with quite heavy springs to give pitch centering. Mounting the whole system on a cradle, moving in the pitch axis will allow the pitch centre point to be controlled.

Thus, if the stick is pulled back to climb, the bottom of the stick pushes the channel against the forward spring, giving a force on the stick. A trim control, moving the whole cradle forward will re-centre the forces, leaving the stick in the correct attitude with no force.

A quick and dirty mockup - https://www.dropbox.com/s/giw3httvdhh0nk9/IMAG0045.jpg

Tell me all the problems you can see :-)

Very interesting idea. One possible challenge would be the non linear force at either pitch extreme caused by the increased angle of the shaft in that channel. But I suspect there's a way to over come that. I was over thinking your idea (as I do many things) and trying to visualize adding aileron feedback and trim by mounting your pitch channel slide device on top of yet another that moves on the aileron axis...


Here's another approach. Have you seen these?

Attachment 7208Attachment 7209Attachment 7210

They are designs for a FFD yoke, flight column, and rudder pedals by Ian from Built for Fun.

http://www.youtube.com/watch?v=KlVkeJlQYXg

In the world of DIY Force feedback IMHO I think they are among the best. Not cheap mind you, but top shelf all the way. I am saving my pennies to build the yoke system. My sim pit build is using spring returns for now, but is designed so that it will accommodate the unit at a future date when I can afford it.

Hi Tom

I've seen the BFF design, but cost is the issue. or that much, I could complete my training and do it for real!

This morning, I built the channel with sliders, but the springs I have are just too strong, so it's had to go on hold until the new springs arrive.

Not too concerned about the angle change. The sequence I need to imprint on my brain is Power,Attitude,Trim, so the forces will be trimmed out as soon as the required attitude is reached.

Another bad picture of the new channel - https://www.dropbox.com/s/v5aks9sizz7pfe9/IMAG0046.jpg

The yellow spacers on the ends are merely to give tension until fitted into the cradle. The yellow spacers in the channel are old bits of kite spar to give a smooth slide over the aluminium tube.

The bit I still have to work out, is how to adjust the trim. Ideally it would be a lever at seat height. Mechanical connection may be difficult over such a long run, so I'm torn between a flexishaft on a ball screw, a motor on a ball screw, or a flexible control cable to a lever.

The joys of designing on the fly!

Don't know if you saw this latest FFB post, so I figured I'd pass it on.

http://www.mycockpit.org/forums/general-aviation-ga-builder-disccusion/25402-diy-force-feedback-yoke.html

It looks fairly simple to build, and far less costly than the BFF solution I shared above.

What really caught my attention in light of your design is that I can see how easy it would be to adapt Roland's idea to your mechanical linkages. Do an e-Bay hunt for a Side Winder, and a couple DC motors, add some pulleys to your yoke gimble shafts, give some creative thought motor mounts and you'd have some fairly inexpensive live FFB.

http://www.youtube.com/watch?v=WcIpXSaVEa4&list=UUPXwxOwrFnl4iRuGYPnDRUQ&index=1&feature=plcp

If I maintain my desktop sim after I finish my fuselage restoration, I might consider Roland's idea as the econo version of the BFF that's going in the fuselage.

Hi Tom,

Just a side note here:

Ian and I had lenghty discussions on Force Feedback systems. The main issue is how to achieve smooth control loading force at the higher force levels. My modest 5kg elevator force is reasonably smooth, but at max force the commutation effects can be felt. Keep in mind that my elevator motor is a high quality skewed rotor DC motor, which would be expensive when bought new. So motor choice is really key here.

Ian also did experiments with DC motors, only to find that at high forces, the coarseness of the commutators is really hard to solve even with expensive DC motors. Hence his decision to go to the brushless sinewave drive, which overcomes these issues.

As in all things in engineering, it depends on what you are willing to compromize.

Quote:

---

Originally Posted by Roland

Hi Tom,

Just a side note here:

Ian and I had lenghty discussions on Force Feedback systems. The main issue is how to achieve smooth control loading force at the higher force levels. My modest 5kg elevator force is reasonably smooth, but at max force the commutation effects can be felt. Keep in mind that my elevator motor is a high quality skewed rotor DC motor, which would be expensive when bought new. So motor choice is really key here.

Ian also did experiments with DC motors, only to find that at high forces, the coarseness of the commutators is really hard to solve even with expensive DC motors. Hence his decision to go to the brushless sinewave drive, which overcomes these issues.

As in all things in engineering, it depends on what you are willing to compromize.

---

Roland, thanks for the additional feedback. I am working on a Cessna 172 fuselage conversion (a link to my project website is in my signature line, below) to replace my current desktop sim and have Ian's system in mind for that build. At least when I can afford the parts that is. In the interim I have his drawings and have allowed for the space needed and mechanical interfaces for it. And, of course, I'm attempting to save up the money...

The original idea was to stop any further development on my desktop sim pit and divert all funding into my fuselage conversion. I planned to continuing using the desktop sim for prototyping and recreational sim'ing until the fuselage is done, or at least usable. It seems, though, that while the majority of my sim budget does go into the fuselage conversion some money keeps finding its way into the occasional bit or bobble for the desktop pit. I keep falsely rationalizing this as part of the prototyping effort, but deep down I know that I really want to keep both pits. I just hope my wife never finds that out :mrgreen:

If I do in fact keep the desktop pit then FFB would be a great addition to it, but a less costly solution would need to be used. Sacrificing a bit of fidelity due to the coarseness of the commutators under load, would be more than acceptable on the desktop pit if it helps to reduce the build cost. But, that's probably a long way off at this point so I have plenty of time to think about it and find a doghouse to live in after my wife hears this plan...

Again, thanks for the feedback, I have been aware of your sim project and website for some time now and enjoy stopping in from time to time to see your work and developments.

Tom G.

Hi Tom,

The desktop pit would definitely be suitable for a FFB addition. For the full blown cessna, keep in mind that the installed controls must be light in movement, as any mechanical friction needs to be overcome by the force system, and may reduce force fidelity.

Good luck with both sims!

Tom

Interesting video, but my budget/wife won't stretch that far :-(

The bit where he's testing the trim, he says "reducing the pull to zero", that's what I'm trying to do. Sort of a trimable centering spring.

The size of the springs will ultimately dictate the rest of the design. VW van side door striker plate springs look about right.

The trim control will initially have a Eurostar like lever, connected to another lever on the base board via a bicycle brake cable. The board lever will connect to the cradle with a pushrod. When I'm happy that the system replicates the Eurostar, I'll look into a motorised system.

Quote:

---

Originally Posted by Morticiaskeeper

Tom

Interesting video, but my budget/wife won't stretch that far :-(

The bit where he's testing the trim, he says "reducing the pull to zero", that's what I'm trying to do. Sort of a trimable centering spring.

The size of the springs will ultimately dictate the rest of the design. VW van side door striker plate springs look about right.

The trim control will initially have a Eurostar like lever, connected to another lever on the base board via a bicycle brake cable. The board lever will connect to the cradle with a pushrod. When I'm happy that the system replicates the Eurostar, I'll look into a motorised system.

---

You're not alone when it comes to the wife err uhm I mean budget constraints. That's why I won't be installing any FFB for quite some time...

Keep us posted on progress. More pic's as it comes together would be great. Others may be interested in your idea.

A quick update

Nothing has changed :-(

the new springs arrived, and looked fine, but were advertised at 8mm ID when they were in fact 7.54mmID. The aluminium slides I made are 7.96mm! I've now made brass slides from 7mm OD .45 wall tube. The spring pressures are better, and I've preloaded them so there is no slack in the system.

The local Mercedes Benz truck dealer let me dig out a ball joint with an 8mm thread. I found a piece of 25mm galvanised conduit for the joystick and bought an end cap. Trying to tap the end cap with an 8mm thread was going ok, until.... The tap broke :-(. Couldn't get another tap until this morning, but can't get the stub out, so ill have to wait until Monday to get another cap.

I started to drill the new steel gimbal frame, 4 angle brackets bolted together, but the drill batteries are not liking the cold workshop and die after 2 6mm holes.

The good news is: the weather looks like improving and I've got a lesson booked for Wednesday.

Another quick update.

Life has got in the way over the past few weeks, company move, job interview, now our town has flooded!

The flying lesson was great, we hired two eurostars with instructors, my wife in one and myself in the other. We did a formation take off and then some side by side photos, my wife then headed back to the airfield while I had a lesson on climbing and descending turns. The instructor got me doing all he radio calls and commented that I fly as if I have weekly lessons, even though its bi-monthly.

The joystick has a bit of a problem - the sliding carriage is a bit too slidey. Cable operation works, but to push against the springs forces, it will need to have two cables, a push/pull system with a cable at each end of the carriage.

Another picture, showing the stick held in place by my son, one cable on hashed brackets and the ball joint in the channel.

https://www.dropbox.com/s/jp0xr3w9wg...2013.30.46.jpg

I'm researching control loading schemes for use in my sim. The links to your sliding channel trim concept are broken. Could you please post the pictures to the forum? How did your idea work out?

Thanks

Alan/Denver

Quote:

---

Originally Posted by wa5zxu

I'm researching control loading schemes for use in my sim. The links to your sliding channel trim concept are broken. Could you please post the pictures to the forum? How did your idea work out?

Thanks

Alan/Denver

---

Alan

Sorry, the photos have gone, along with the computer :(

i couldn't get enough accuracy in the mechanism, the springs exerted a lot of twisting forces into a slack mechanism. If I could have built a more precise mechanism, I think a ballscrew would have done the job, but that's beyond my means.

i also tried the concept of elastic cord to provide the spring force with trim adjustment provided by winding the cord onto a roller. This had a bit of a problem with friction, but ultimately, the lack of precise mechanics, and the engineering tools to build it, meant that the project died.

ive also switched aircraft to an Icarus C42, with electric pitch trim!