Let's design a joystick...

[Wolf Rider]

How about it?

 

keep in mind though, this shouldn't become a this or that idea is better, or this or that idea is cr@p thread. It has been started with the best intention of getting good ideas out, so that a better joystick can be arrived at.

 

here's a thought for the "gymbal/ sensor" mechanism for the joystick itself.

 

It comprises the shaft of the joystick passing through a solid ball, which is held in place by two matching cups.

The cups are secured togaether with a double spring arrangement to supply a constant tension (self tensioning?) from the cups to ball. I had thought about a race but getting a race to work in 3dimensions may be a bit tricky.

 

At the tip of the lower shaft. which has passed through solid ball sits a "mouse type infrared/ laser sensor", which reads against another matching cup. This is similar in application to a mouse on a mouse pad, except as a cup it acts as a 3dimesional mat. Calibration shouldn't be an issue as the stick isn't repositionable such as mouse is on a pad (ie, the mouse can be picked up and positioned anywhere on the pad, whereas the stick is always fixed).

 

for tensioning an octagonal (8 springs mapped out on a frame, evenly spaced) tensioning spring arrangement could be mounted between the joystick shaft (underneath the handle)l and the solid ball. the solid ball is to supply a bit of "weight" to the stick.

*Alternatively attach some motors instead (perhaps 4/ maybe eight at 90degrees/ opposite to each other and have the opposing motors wired in a reverse phase to its opposite match, so a push/ pull configuration is achieved).

 

Oh... this idea I give freely to the community under public domain.

 

yes, the image is rough (and off the top), but I only have MSPaint to work with

 

 

 

*Late edit

Edited January 7, 2011 by Wolf Rider

[aelfewulf]

Great topic!

 

Here's are some ideas (or wishlist ;))

 

1. Variable stick tension via easy spring swap mechanism.

 

2. Different stick designs modelled after real aircraft, swappable (like Cougar/Warthog.

 

3. 2/3 clip-on palmrests of different thickness to cater to different hand sizes. Similar to what X65 has now, but it comes with only one.

 

4. Dual throttles with easy removable idle and afterburner detents.

 

5. Different throttle grip designs, modelled after real aircraft, swappable.

 

6. Wireless, similar to logitech's Unifying usb thing

 

And now for the really crazy idea:

 

MODULAR DESIGN.

 

Basically the idea is that u buy the basic throttle and stick base, choose between either force-sensing module or spring module, then finally choose the throttle grip and stick design. The force sensing/spring module shd be easily assembled into the stick base.

[y2kiah]

Good idea with the laser mouse sensor. One thing I think is that calibration could potentially be a big issue with this design. There is no feedback loop to directly correlate center stick position with the sensor's starting position, you just have to tell it that it is centered and it will have to believe you. This is a "dead reckoning" type positioning so it's possible that after some use, errors will build up and could cause the perceived center to creep. Or... it very well could turn out that the laser is so precise, error buildup is negligible and you would never be able to tell.

 

edit: just wanted to add that regardless of laser sensor accuracy, tracking the position in software will inevitably end up with a buildup of floating point error. This is a common problem when integrating floating point numbers in any scenario. Because if this, I think it would be worth looking into incorporating an origin sensor so the stick can automatically recalibrate itself.

Edited January 7, 2011 by y2kiah

[Frazer]

I'm in contact with the Paccus firm, designing hydraulic flightsim equipment.

Will follow this thread and offer them the most interesting wishes/ideas.

http://forums.eagle.ru/showthread.php?t=64741

[Wolf Rider]

A habit I developed (good/ bad doesn't matter) with my MS FFB stick was to rest my hand/ palm on the base itself, which gave for me, very good fine control when needed and didn't add extra weight onto the gimbal. It was also very comfortable. Of course when dogging the hand grip proper was held.

I find a lot of sticks out today don't have the same degree of comfort when resting the hand on the base.

[walker450]

Great topic!

 

One thing I thought of for the list above:

 

The ability to easily add an extension.

[power5]

Do not need the stick to extend beyond bottom of "ball". Just have the laser at the bottom point of ball. A few mm below the sensor would be the cup. Actually, in that instance it does not even need to have the laser at the bottom. It could be on the back, front, or anywhere else.

[Wolf Rider]

no it doesn't Power5, it (the rough sketch) it won't necessarily need to extend any great distance below the lower cup, but the lower cup is needed to mount the ball so it acts as a gimbal. The top and lower cups would have a cutout to allow the stick the range of movement, plus the stick would want easy servicing with minimal "pulling' it completely apart

[walker450]

I was talking about an extension that attaches to the grip, so it can be set on the ground for simpits and be center mounted.

[Wolf Rider]

That's a good thought... as well as the ability to easily rotate (offset) the grip

[power5]

The ball could be suspended on roller bearings like old mouse balls. Then the ability to tighten the pressure on the bearings would give more or less tightness to the stick. The ball should be suspended not on a pivot. Also, if you add motors to the roller bearings you can have FFB as well.

[metalnwood]

Balll bearings are not utilised in that way. You usually have a good fit on the bearing housing but don't try to squeeze them to create friction, in this case for pressure on the stick. As the ball bearings have little surface area it doesnt work except to try and bind the balls.

 

 

You would also need the FFB mechanism to attach to the stick as you would not be able to transfer the force through the bearings adequately.

 

Depending on the final design you would probably have the stick go right through the main ball joint and attach to the ffb motors below.

 

 

I like the idea of the thread, there are a few people who are up to machining the components.

[Feed]

Microsoft "precision" sticks already used the optical tracking idea. I like the idea of this thread, but part of the problem is that no one stick is going to satisfy everyone. I agree that a modular design concept is absolutely the best way to go, since components could be 'mixed and matched' for each individual user.

 

Want a force-feedback, long-stick with modern jet-fighter grip? No problem. You want a (nearly) motionless load-cell sidestick? No problem. You want a dual throttle or a heli collective with that? How about both?

[Wolf Rider]

there not being one solution to fit 100% of needs, is very true indeed in so many situations

[emenance]

Ill take anything that doesnt die or flash back at me!!!!!!!!

[mikoyan]

My idea would be to develop a force feed back stick that runs by a separata software that reads the pilot's input and then provide different stick forces by using data imported from the sim, at the same time the stick would transfer a corrected imput to the simulator.... That is my dream, that way for example; when you fly the f-15 it would be more realistic and you would not over g the airplane so easy. The software could be program to give you 1 g for each inch of stick travel.

[y2kiah]

I think the stick should use an open-source and widely adopted microcontroller with USB HID capability. Something like Arduino Uno for example. That way, if anyone's electronics go bad, it will be easy and cheap for them to replace the board. Also, the firmware should be an open source project so it is sure to be constantly improved by the community, will always be free, and will never go missing.

[Sokol1_br]

Very nice DIY one here:

 

http://forum.fliegende-schweine.de/showthread.php?tid=102

 

Sokol1_Br

[power5]

Thats pretty sweet. Add some motors to the pivots and you would have FFB ability.

[Succellus]

Well that basicly a UBX(UBN?) cougar with hall sensors.