Hi everybody, this is my first post and also my first attempt to build a diy floor mounted stick.
Building this i made extensive use of http://simhq.com/forum/ubbthreads.php/topics/2958931/Uni_joint_stick_DIY_hand_grip.html#Post2958931 so all credits to JFA2 and to julian265 http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/3231057376/p/1 .
I only followed their instructions.
Stick uses hall sensors and the gimbal is an uni joint (actually only the core of the universal articulation), the rest is welded steel plate/tube; hand grip is carved wood and interface is Leo Bodnar's BU0836X card. Im very aware of my lack of skill...but I did tried my best (so far)
My problem: I'm using the stick "as it is", only calibrating the axes in windows,everything ok, smooth movement but I'm still trying to adapt on my "long stick"..very different feeling comparing to rest of the sticks I've used untill now.
Longer movement means more precision, but gives me some kind of a "lag" feeling, my hand had to actualy TRAVEL from side to side, while using an usual stick,the movement is almost instantaneous...tricky feeling, almost felt like main advantage is it biggest disadvantage.
I'd like to get some advices/feedback from more experimented "long stick" users: should I use some program like Diview to tweak the axes? And how? I don't know, it works perfect, and yet it seems strange.
Also can someone please explain to me what are the meaning of those numbers in joytester screen (min step, precision), is ok or do i have to further tweak the hal sensors?








And joytester screens:

Volats,

Good job.

About JoyTester2:

This minimum value ~60 is common in tests with other joys.

The maximum value - 1074 ~ 1057 indicates that axis reach 10 bits resolution (1.024).

Notice that BU0836-X is 12 bits (4.096), maybe your magnets movement are to small, ie.
HALL sensor center is in the same axis center of magnets, try move up ~2 mm:



Sokol1

The fact that you are using a long stick also can be the cause of the limited maximum value obtained. BTW what is the lenght of the stick and the travel in degree
on each axis?

Thanks for answer. Now I understand a bit more what joytester says.
In my pursuit for perfection , center of hall is carefully aligned in center of magnet (and in the axis of rotation), indeed, hall movement is very small (and it has to be small in order to keep a reasonably movement of the hand grip, I can't swing may arm like 30-40 cm in every direction).
My setup:




What can i do to improve accuracy? Should I offset hall center from magnet axis? (at least this is what i understand)
Also I tested my old stick (thrustmaster afterburner 2) with joytester and was unable to draw a simple spiral, VERY chaotic movement. And yet I can play as usual... Why is that?

afterburner test:


So the improved accuracy (let's say from 10 bit to 12 bit)really maters in game, or is just visible when tested in benchmarks?

Thanks for your help...I'm just trying to get a grip on diy flight controllers (and yes, It's a virus...
)

Stick lenght is 67cm, and axis movement is (roughly) 44-45 degree (so aprox 22 degree on +X and 22 degree on -X; for example).The top of the handgrip travel 26 cm (13 on +X and 13cm on -X, to have the same axis as example).

Volans,

Amount of stick movement ~ 44-45 degrees - is fine, similar a conventional (short) stick.

Look at the picture of HALL assembly that I post above, where center o HALL is 2,5 mm above magnet center (joystick axis).
Play with this and test. Adjusting you reach 90/100% of possible resolution.

Sokol1

If you're happy with the linearity / smoothness of the sensors, you could reduce the length of the stick and mount it on a higher base. This would reduce the amount of movement needed, and reduce the interference between your leg and the stick at full deflection.

Otherwise you can get more sensitivity by using a higher sensitivity sensor and/or higher strength magnet.

You should also take care to match the stick's physical deflection limits with the windows deflection limits, so that you're not needlessly moving the stick further than makes a difference. (judging by the extents reached in JoyTester, you've probably already done this. BTW that's some accurate movement!)

Good job, Congratulations!

You may also concider to reduce the movement of the stick by using a square collar on the top of the box. Try until movement is ok for you and recalibrate. Accuracy is probably going to take a little hit but itīs nothing you can really notice in practice.

Mine is approximately 65 cm and i like it just like it is now. You can also use thick O-rings instead of springs. You get more smooth and "aerodynamic" feel and no spring noise. I found double O-rings to be the best choise.

Cheers

JFA

Thanks for your answers.
Physical movement of the stick is already matched with windows calibration, maximum real deflection is also maximum software movement.
On top of the box I drilled a round hole, acting as a guide to maximum stick deflection, at first it was 12 cm diameter, but then the stick movement was a nightmare, so I had to rebuild the top cover, drilling only a 8 cm hole. I figured that it can't be smaller without messing hall sensors output.

(The old top cover.New one is similar but smaller hole)


Mounting it on a higher base is a no go for me as the whole point of this project was to build a "real" stick.
The sprigs I used were surprisingly hard to find, hard to trim, and not even now I am satisfied with the feeling...I cannot recommend sprigs for centering, as they have a tendency to off-center in time....I will try to replace them with rubber bands, when I'll find some.

Those days I was able to actually play with the stick, and slowly i find it more pleasant to fly with... I remember when first used my rudder pedals (switching from thrustmaster rocker) , I was crying for my money back.... now i can't fly without them.

Great work, Volans

I think once you get it tweaked to your liking, you'll never want to go back to a deskmount. I've been happily using my original for several years, although I'm working on a FFB version now.

As JFA2 mentioned, you might consider a square hole in the top of the box (with rounded corners). The round hole limits response slightly when you want full deflection in both axes.

If you used good quality springs, you shouldn't have to worry too much about the stick getting off-center, and if it does that's why I used threaded upper mounts - adjustment is easy. If spring noise is an issue it's mostly caused when the hook slides on the mount - that's why I used drilled bolts with a little grease.

As your joytest drawings show, the spring system is pretty smooth - can't say as I've seen a high-priced stick do that well. Again, nice work.

9T