Force feedback throttle

[SC-Maik]

Three brushless motors (one per pivot arm mounted on each shaft) keep pivot arms at zero position, around 135 degrees.
Everytime time user moves throttle around, motors are trying to return all three pivot arms at neutral position.

One switch to turn delta throttle into a normal throttle while crusing, locking two pivot arms together (arm 2 and 3) while leaving arm 1 (Y axis only) to move freely or even lock it on last position.
One pot to change motors' voltage and resistance.
Option to change resistance depending on arms' position with arduino programming.
No need for springs.

Thoughts?

[jeruw]

Sounds awesome! Unsure how that would affect cost, but sounds like it'd be a lot of fun to build. I wonder how smart the programming could be as well, so that the motors could not just provide resistance but also provide an assist with movement, so for example with a positive Z movement the motors would release resistance or even reverse to make raising the handle vertically effortless. Right now positive Z movement is difficult unless you've clamped the throttle to your table. You can't get much more positive Z than just whatever your springs give going back to neutral.

Would there be any added noise, risk to damaging the motors or excess heat using the motors in that way? Mock anything up yet?

[SC-Maik]

A friend gave me today a broken Logitech G25 force feedback wheel and I have permission to disassemble it.
I want to have a look of that motor, gears and possible belts used. And of course use that motor to mess around and try move one pivot arm.
Programming won't be an easy task but it will be fun to build indeed.

Added noise, probably - Risk of damaging motor, well they work like that in many joysticks/wheels (MS Sidewinder FF2, G25/27 to name some). I guess I will figure out soon.
I have to find motors that can stall without damaging shaft, gear etc. Servocity has some with titanium gears, dual ball bearings etc. Real challenge would be to find the one with the right torque/speed ratio.

My "normal" built with springs is 90% complete, it works above my expectations, so I want/need a new challenge.

[mike43110]

This makes me think stepper motors...

Though I would have no idea how to implement anything with that!

An open FFB implementation would be awesome, I would most certainly use it as it would do everything I wanted then!
No need for the throttle slider I was adding, just a mode switch somewhere easily accessible for "Throttle" mode - hell you could even have it work then as an actual throttle axis with some simple programming. Locking a single axis would be possible definitely. I would have to read through all that inverse kinematic code however >_< will have to see how exactly that would be done. You would be telling the motors that any deviation in the other linear axis is disallowed and they must be returned to 0 while keeping the free axis in place.

It would be interesting to try and use actual force feedback information - when it is made available in the space-sims we all love. I know vibration feedback through the ffb motors is possible as well. All of this makes for a very ambitious project but one that would be very fun once completed. I know the ffb motors can even be used for vibrational feedback. So the throttle could vibrate slightly when shot as an example. For the stress on the motors however it may be better to add small vibration motors into the handle (assuming that information can be used even!

Now if my parts could ever arrive...

[KStolp]

I have been working on FFB version of the Delta Throttle as well, currently waiting for parts to come in before printing my base.
I have ridged up a few different motor to test their draw of power along with the heat built up by both the motor and H Bridge for such a system, i don't currently see a heating issue. i also don't think Steppers would be a reasonable option for this, rather costly for a motor you will be forcing against the direction it wants to go most of the time. I am currently waiting for some gears to come in to do some gear ration tests and see how much torque and heat is built up when gearing it up.

[ai88]

What does everyone think about this: http://www.novint.com/index.php?option=com_content&view=article&id=39&Itemid=175?

Looks like it is a consumer version of this: http://www.eurohaptics.vision.ee.ethz.ch/2001/grange.pdf / http://www.forcedimension.com/downloads/specs/specsheet-delta.3.pdf.

Seems like a delta throttle turned on its side and someone did a bit of breakdown post http://qdot.livejournal.com/236133.html.

[mike43110]

Anybody made any physical progress on the building of an FFB version?

Sadly I am still waiting for all my parts for a vanilla delta so I can't even start playing with the idea myself.

If heating isn't an issue, 2 modes would be simple to do. one would be a constant return force which simply is a constant force which doesn't change with extension. The second would be the ramp force which is greater the further apart you are.

I was thinking on how to do SC-Maik's holding idea. That may be simpler than I thought. The motors will only react when there is a deflection away from 0 in any direction. The motors then just get a low reactionary force and you can then pull the arm around with minimal resistance. The motors then as it moves change the point they are aiming for to be where you leave the throttle.

This would form the basis of the controller for the more difficult throttle mimicing setup. That would do the same in the x axis but center the other axis. So you can use it as an ordinary throttle as well as simultaneously using it for strafing left/right up/down.

Sadly I don't have anything physical to work with. If I did I think I could quite easily get a working version. Optimising the PID values for the controllers would be a different story however... That would take a bit of time.

[KStolp]

After a long time of fighting with my printer and waiting for parts, I finally have a Physical Prototype working in motion only
i have not get it up to feed XYZ pos in the computer yet, that being the easier of the problems.

My Current working Version is running Brush Motors, they have been geared down so the RPM is lower.
Analog Hall Effects A1304's (dont have my part number in front of me at the moment might be a different one) with 2 Magnets circling each one

it turned out that after printing my Gear Spacing is off slightly so there are problems with the current version.

I am getting ready to print a Version 2 along with and Equivalent Version 2 but for Stepper Motors to see the pro's and con's of each. i have also gotten MLX90363's to test with version 2 see if they perform better.
should also add the they system is running all on 5V, but had to add 2A from a phone charger to drive the motors with enough force

I will try to get a PIC and VID up later for all to see

[aniron]

After a long time of fighting with my printer and waiting for parts, I finally have a Physical Prototype working in motion only
i have not get it up to feed XYZ pos in the computer yet, that being the easier of the problems.

My Current working Version is running Brush Motors, they have been geared down so the RPM is lower.
Analog Hall Effects A1304's (dont have my part number in front of me at the moment might be a different one) with 2 Magnets circling each one

it turned out that after printing my Gear Spacing is off slightly so there are problems with the current version.

I am getting ready to print a Version 2 along with and Equivalent Version 2 but for Stepper Motors to see the pro's and con's of each. i have also gotten MLX90363's to test with version 2 see if they perform better.
should also add the they system is running all on 5V, but had to add 2A from a phone charger to drive the motors with enough force

I will try to get a PIC and VID up later for all to see

Looking forward to seeing them!

Been fighting my own printer to get it to produce an actobotics-compatible baseplate prototype, as opposed to ripping the glass bed off halfway through and making a blobby mess of PLA.

[eieikuean]

The motors then as it moves change the point they are aiming for to be where you leave the throttle.
gclub royal