Hall sensor joystick progress

[Peter_Pilot]

Hi All,

I have put some more photos in the generic section showing progress of my Hall sensor based joystick. The primary driver at the moment is to play Black Shark without using my CH yoke which is obviously inappropriate. There is plenty more to do (spring feel etc.) but I have at last got it basically working.

I have had the helicopter seat and plinth for some time and was annoyed to find that because I don't strap in when I fly my sims the centre strap and buckle had scuffed the front of my seat plinth. This is interesting!

A number of Gazelle helicopters previously in service with the UK armed forces have found their way into private hands. One such owner was flying his helicopter on a journey when he noticed that the gyro compass was not working properly. He landed at an airport on the way and telephoned his engineer. After making some adjustments it was necessary to test the compass with a few spot turns in the hover. As it was going to be a quick test only a few feet off the ground the pilot neglected to do up his straps. The centre strap became wedged between the stick and the seat and the resulting heavy landing caused significant damage to the aircraft, maybe a write off!

The moral of the story is of course to always strap in whenever the rotors are going to turn, you never know what might happen.

Comments on my stick welcome, particularly if you can see any pitfalls I am likely to face with the spring etc.

Thanks,

Peter_Pilot

[Jordan Farmer]

Wow sir, that looks superb, great work mate :) Looks like youve worked really hard on that, wish i had the skill and tools to do the same kind of thing!

Jordan

[Peter_Pilot]

Hi Jordan,

thanks for your comments. I have had this week off work and it has been a hard slog to get to the stage you see (9am - 10pm the last three days). The only "luxury" tool I have is the lathe otherwise it is all done using hacksaw, files and pillar drill and lots of elbow grease. The lathe was a gift from my employer for 25 years sitting at one of his desks, it is useful for turning small parts like brass bushes etc.

How are your projects going? do you use one 'plane in particular or like me lots?

Peter.

[Jordan Farmer]

Well, at the moment my project is going very slowly, i dont have the time or money, but i have all my plans ready, i have 2 yokes from a C152 and 2 leather seats from a Range Rover, perfect start :D But yeah, my sime will be generic(ish) i will build it around the boeing kinda style, as i love flying the 747, but i will find it much easier to follow my own plans, rather than omeone elses ;) Nice gift from your employer, i think the most ive ever got is a mars bar! Good luck with the rest of it ;)

Jordan

[Peter_Pilot]

Hi All,

In case anyone interested I have modified my cyclic stick this weekend. I changed the fore/aft cyclic Hall sensor to direct drive instead of the 3:1 toothed belt drive I had before (see photos in generic section). I made this change because I was getting too much backlash. I was surprised by this because the lateral cyclic I built first was almost backlash free, I think I had some minor differences in geometry between the two axes and the fore/aft axis somehow ended up with more friction. With my most powerful magnet I can now get 0.5v to 4.5v for +/- 15deg of stick motion which is adequate. It now works really well with Black Shark.

The toothed belts were probably a bad idea in the first place, unnecessary and complicated. However, they can be made to work without perceptible backlash as long as the friction is kept under control, the belt loads are then as light as possible. Maybe belts are easier to use compared to gears in case you need a ratio or you have space constraints.

Comments or questions welcome.

Peter.

[Peter_Pilot]

Hi All,

Update for weekend 19/20 Sep 09.

I have added the spring shown in my latest photo in the generic photo gallery, I was aiming for a grip force of 10N for 15 deg deflection with as near linear relationship between force and deflection as possible. I bought the exact spring I needed, initial tension, rate and preload but the force is a little lower than intended. Maybe I havn't got exactly the preload I was aiming for. I may add dampers later if I can find some suitable ones.

Work to do next is to mount the BU0836X card inside the seat plinth and tidy the wiring. Then I have to wire up some switches and buttons on the grip. After that I will make start on my collective stick.

Best wishes,

Peter.

[Colibri]

Hi All,

In case anyone interested I have modified my cyclic stick this weekend. I changed the fore/aft cyclic Hall sensor to direct drive instead of the 3:1 toothed belt drive I had before
Actually, I was surprised that you used belt drive. I thought that one of the major positive point of using hall effect sensors was to avoid any mechanical need.
I keep an eye on your project as I'm interested in Hall effect for the next evolution of my pit. You do great job !

[Peter_Pilot]

Hello Colibri,

Yes, you are right, direct drive is much better. You just need to make sure you have the right strength magnets for the angle of rotation you plan to use. I got mine from http://www.first4magnets.com/index.asp

Good luck with your project, keep us posted with how you get on.

Peter.

[Colibri]

Thanks for the link. It seems they have everything !
As I don't know anything about hall effect yet, can you explain how to know the magnets needed ?
Do you know a good (... this also means "easy") tutorial about hall effects ?

[Peter_Pilot]

Colibri,

Apologies for the delay I have been away for work.

I went for Hall sensors because I read about them on Mike Powell's web site and I bought his book. I also found out about Leo Bodnar's BU0836X card on Ian Sisson's 737 web site, when I went to Leo's site there were Hall sensors for sale as well so I decided to give them a go. Hall sensors plug into the BU0836X card directly instead of potentiometers. They are very easy, one wire goes to +5v, another wire goes to ground (0v) and the voltage at the third wire varies depending on the magnetic field strength normal (at right angles to) to the surface of the sensor. If there is no magnetic field the voltage is 50% x 5v = 2.5v. As the magnets are rotated the field goes more North or more South. If the field goes more North the voltage goes up and if the field goes more South the voltage goes down (or the other way round!). If the magnets are weak the voltage only goes up and down a little. If the magnets are just right the voltage goes nearly up to 5v and nearly down to 0v for the full travel of your control. If the magnets are too strong the voltage goes up to 5v or down to 0v too quickly, that is before you reach the end of the travel of your control. The voltage cannot go above the supply voltage (usually 5v) or below 0v.

As far as magnets are concerned, I bought Neodymium disk magnets, N42 grade 15mm x 1mm, 15mm x 2mm and 15mm x 5mm. a number of magnets can then be stacked to give the right magnetic field strength. You have to be careful with these magnets, I accidentally let one of the 1mm thick magnets fly towards a stack and it broke in half on impact.

Some links:

http://www.mycockpit.org/tutorials/showentry.php?e=27
http://www.mikesflightdeck.com/ (scroll down the page to get to the Hall sensor info).
http://www.737ng.co.uk/technical.htm
http://www.leobodnar.com/

Best wishes,

Peter

[Colibri]

Thanks for this lesson, Peter ! It seems so easy that I wonder why don't everybody use them. They're must be some inconvenients, do you know them ? I read on Jesse Jellen's website (http://www.computerrepairs.ca/Simulator.htm) (Splah One on Hovercontrol simbuilders forum (http://64.34.169.161/cgi-bin/ib3/ikonboard.cgi?;act=SF;f=43)) that he got spikes with them, does anybody else experienced that ?

[Peter_Pilot]

My system is smooth as silk Colibri, no spikes.

Maybe some of the complication is in the magnets and the type of joystick interface that people use? I have looked at the photos of your cyclic control and was wondering what the electronics are that you have used?

Peter.

[Colibri]

The electronic you can see on the picture comes from the old Logitech joystick I cannibalized to get the grip. I use it only for the grip's buttons.
The main electronic is the USB axes card from Opencockpits (this one) (http://www.opencockpits.com/catalog/axes-card-p-55.html?cPath=21_36). It is located in the center console.

[Peter_Pilot]

Ok, thanks.

I have some iocards but not the USB axes one.