Instructions - Joystick Gimbal Build
HTML-код
- Опубликовано: 1 окт 2024
- Download the files for free at www.realrobots...
Support realrobots at / realrobots
1 x GE12C Spherical Bearing
2 x 0.7x5x25 tension springs.
2 x 14mm M3 bolts
2 x WH148 B10k 15mm potentiometers
1 x 4p Aircraft plug male & female GX16-4P
These are the instructions on how to build the realrobots.net joystick gimbal. Parts available on Thingiverse.
www.thingivers...
A few practice and he probably could sound like Morgan Freeman
I love this project. Simple and practical!
I just love a VERY well made design for 3d print. Congratulations and thank you! :)
aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
this is more fun to watch than I would have expected :)
Today i found out, You can buy Spherical Bearings. Thanks!
When i get my hands on a 3D printer, im gonna run some tests to see which is better/more beneficial/easier to make for a Flight Stick, Spherical Bearing gimbal or the more traditional style with two or more bearings. I'm absolutely not gonna touch the Single spring down the center type of gimbal design. Eugh, it works in a pinch but it's just kind of nasty.
Let us know how you go, I tried a few different designs with one axis rotating inside the other, I ended up going with this design because it was more compact. It also puts all the stress on that spherical bearing rather than on the rockers or potentiometers, so you don't need to put in rotary bearings for every axis.
@@jakewilkinson7815 I'd suggest going contact less if you're worried about about stress on components. Also, that should make it even more compact. No need for the extra space on the sides for Potentiometers if you use a Hall Effect Sensor. Or even a single Hall Effect sensor in the middle bottom, with a strong enough magnet. Could also use its counterpart, whose name escapes me right now, but its essentially a magnetic type sensor similar to the Hall Effect sensor that detects in the other axis. Ill come back to these comments when i get a Printer and some hardware. Also, use a modern, high bitrate Arduino or any kind of chip that has 12 - 18 Bit resolution.
@@gokuta333 Working on a yaw-axis grip and that'll have an AS5600 rotary magnetic sensor :-) Of course now I have to have rotary bearings.
Yeah the Arduino Pro Micro runs at 10-bit precision, usually would be enough but when you're using only a small segment of the pot it can be a little rough. I've got my software running on ESP32 with bluetooth now and that guy has 12-bit precision ADCs.
www.gitlab.com/realrobots/rr_configurator
I've had a few people ask me where to buy those spherical bearings, did you ever find a good source? I'm in Hong Kong so I get them shipped for a couple of bucks from over the border in Shenzhen but I'm not sure how available they are in the rest of the world.
@@real_robots Usually bearings have a model number that is almost universal. A skate board bearing is a 608zz. Do you know the model of the bearing that you used?
I absolutely love this design, I just gotta make one minor complaint: that spherical bearing is THIRTEEN DOLLARS! I could get a 20-pack of normal bearings for that price.... Although I'm currently trying to design a gimbal with normal bearings and it's not going well so far lol
Yeah I was worried the spherical bearing might be hard to get hold of, I live in Hong Kong so I get them online from China for about $3.50 each.
To be honest though you can just print a ball and socket joint that'll be fine, below is my first prototype that I just printed, wore in a little and packed with PTFE grease.
www.reddit.com/r/HotasDIY/comments/k3pwuf/trying_out_a_ball_and_socket_style_gimbal_this/
@@real_robots Okay, that definitely explains why you'd use one so casually! I would've gone that route too if I had found one that cheap. I'm hoping to simply adapt your spring back and pot mount designs into the four bearing gimbal I've designed.
Thanks for the reply! I'll look into giving that printed bearings try. I've been diving into custom controllers lately and your channel has been a huge help 🥰🥰
My issue with this design is that its softer in x and y and harder with diagonal movements. I cant figure a way to solve this problem as you will always be moving twice as many springs in diagonal. any help on this would be really appreciated.
I see potential weak areas as regards longevity.
The squared bottom end of the shaft, when sliding in the fitting slots of the two gimbal parts, will wear the sides and thus over time become looser and sloppy.
Using threaded screws as a pivot for the two pairs of spring arms has rather less area of contact than a smooth, un-threaded shaft, and this will accelerate wear on those little holes.
If those pots are in the $3 range each, expect to change them not infrequently if the stick gets a fair bit of use.
Generally, the wear characteristics of the plastic used will play a role. It should be maximally resistant against abrasion. A good example of such a plastic is found in CH Products sticks, which have a similar gimbal mechanism as here.
In the end, does this project offer any decided advantages over an off-the-shelf, inexpensive joystick of similar cost? And we're yet to factor in a handle, buttons and controller. ;)
Now, don't misunderstand me. I'm a DIY guy all the way. I built my own rudder pedals in about '94, and that thing will outlive me! I've also built my own astro binoculars, telescopes, dew heater systems for same, power supplies, camera trackers for astro imaging, and all manner of other geegaws not astronomy related. ;) I value--nay, prize--the satisfaction of doing it yourself.
Definitely fair criticism, these things definitely won't last forever, I tried to focus on having fewer non-printed parts and being compact over extreme robustness, if a printed part starts showing wear after a few years it's not hard to replace it.
The wear is minimalized though by having the spherical bearing take the brunt of the punishment rather than riding on the rotors like the CH joysticks. I'm still using my prototype model a year and a half later and it's holding up, I'm waiting for the pots to die to force me to upgrade to fixed bearings and magnetic sensors.
This is brilliant!
Nintendo designed the n64 joystick with gimbals like that.
So looking at the firmware so I can add a few axes, do I just add them to the rr_bleHID.h file? or is there more to it
You need to change the initial report, and also the data packets you send should change to reflect the changes. I also found that Windows wouldn't accept certain combinations of axes and i couldn't work out why, not an expert on HID.
Note that for bluetooth, if you change the report it won't be reflected unless you forget and repair the device.
Saved my ass! Thank you so much!
Wait why is the "gimbal housing" stl file different in the tutorial to the one is this video? The middle cylindrical "shaft" is extended in the thingiverse file meaning the rotors won't fit.
That's support material, you can just break it off after printing
@@real_robots aaah, yeah in my case it's on there pretty firmly. Think I should mark that as support material in my slicer next time. Thanks for the reply!
@@twandepan ive had to go at it with pliers before
@@real_robots I've tried that but it simply won't budge. I'm pretty sure my slicer treated is as a structural part of the object and connected the support to the gimbal housing accordingly. I will make sure to mark it as support and print it again when I get home (I'm on vacation at the moment).
@@twandepan I just realised I separated them with a 0.2mm gap because that's what I set my layer height to, if your layer height is greater than that it might have no gap at all.
1 joystick in virtual simulation is like 1 potentiomètre or what please ???
The joystick has two potentiometers, one for X axis and one for Y.
Hi how do obtain a copy of of the 3d print file to build this
Thanks
www.thingiverse.com/thing:4732811
can a Hall sensor version be made?
Awesome video - too bad hall sensors are not used.
Which software did you use to create this?
Great design!
I would probably need to modify it before I'll use it because I'd rather have hall sensors instead of potentiometers.
Also, how bad is the deadzone? Taking into consideration that it's two plastic parts rubbing on each other it might only get worse with time.
It's not so bad actually, PLA is pretty hard and smooth so it hasn't worn down too much, about 2 years later and its only got a tiny bit of jiggle.
@@real_robots wow great job😀
What is the name of the bearing that allows it to rotate?
That's a GE12C spherical bearing
Hi
How do I obtain a copy of the 3d print file
Thanks