#4108116 - 04/18/15 07:00 AM
Re: DIY hall sensor ...
[Re: f15sim]
|
Joined: Jan 2010
Posts: 464
tirta
Member
|
Member
Joined: Jan 2010
Posts: 464
|
Hi Brandano,
Thanks for the reply. Yes, I am aware that HALL sensor can not reach the full range. But I have used HALL with 2 of my joysticks, and on one of them I can get rather the sensor values rather close to full range:
x axis: 96 - 4068 y axis: 100 - 4060 this is with A1302.
Using this joystick with the current mmjoy is very precise, and when testing it in joy-tester after I calibrate it, I find the values are stable, they do not change.
However, on my other joystick, the one that I change to S495A, I only get around: x axis: 300 - 3800 y axis: 300 - 3800
With this values, mmjoy does not seem can use it correctly. Even after I calibrate it, the maximum and minimum values always get smaller after a while, like I write here: http://SimHQ.com/forum/ubbthreads.php/topics/4092126/Re:_MMJoy_-_Build_your_own_USB#Post4092126 http://SimHQ.com/forum/ubbthreads.php/topics/4093559/Re:_MMJoy_-_Build_your_own_USB#Post4093559
If I understand correctly, using HALL sensor with less than full range should not be a problem with mmjoy, since values get stretched to 0-4096 after calibration.
However with the second joystick, after calibration, the stretched value is 0-4096 for a while, then after I move the stick, the values always decrease, like this:
first, just after calibration: 0 - 4096 (perfect!). then, after a few moves : 100 - 4096 (it started to change) then, move it a few more : 200 - 3900 (more changes) then, move it a few more : 400 - 3700 (more changes) (this is an approximation since the real value always changes a little bit)
so the max and min values always changes, but it is getting smaller all the time.
it might be a bug with mmjoy? or there is something wrong with the sensor or my setup?
|
|
#4108671 - 04/19/15 06:41 PM
Re: DIY hall sensor ...
[Re: f15sim]
|
Joined: Nov 2012
Posts: 161
LocNar
Member
|
Member
Joined: Nov 2012
Posts: 161
|
I've found the 1302 to be the least useful of the Alegro sensors. The sensitivity on them is 1.3mV/Gauss, so take nearly 180 deg of a strong magnet for a full voltage swing, at least with the Bic pen arrangement. The A1301 behaves the same but can use half the magnet, at 2.5mV/Gauss which allowed me to implement some miniaturization. Here's the post I made a while back about miniature DIY sealed Hall pots (I think of them as gen 2 versions of Gene's), which are easy to make and incorporate into controls projects. DIY Sealed Miniature Hall Pots, on the cheap! Since controls don't typically move 180deg, this rules out direct mounting and tends to waste usable range of motion/resolution. Lately I've experimented to good effect with A1324 (5mV/Gauss) and have found them to be most suitable for typical controls, only taking about 90deg for a full swing, allowing direct mounting as well as better utilization of available resolution. That said, I've never had any problems with drift with any of my Alegro's. I only buy them from reputable dealers (Newark Electronics in my case) however, because I hear ebay and other sources are full of sub-par counterfeits.
Last edited by LocNar; 04/19/15 06:43 PM.
|
|
#4108746 - 04/19/15 10:22 PM
Re: DIY hall sensor ...
[Re: tirta]
|
Joined: Nov 2001
Posts: 3,955
Sokol1
Senior Member
|
Senior Member
Joined: Nov 2001
Posts: 3,955
Internet
|
using joystester2 is harder to see, but yes, it does gives me the same changing sensor values.
OK. Using Joytester2 you eliminate problem with JoySetup embed tester. I also try changing the old magnet that I got from dvd player to a stronger one: 10mm x 10mm x 10mm cube neodymium magnet.
I suggest exchange mangetes or axis connections to isolate problem with magnets - a squad mate owner or X-52 notice that one of his magnets became rush and dont work properly. Replace the magnet solve the problem. however, with the new one I get lesser values,
If you still have issues, after exchange axis connection in Arduino board - for confirm that the problem is related with board input and not with HALL/Magnet assembly, rest the board as source of the problem, they can be defective. These DIY HALL assembly involve a lot of (tedious) magnets positions experimentation/test. I have a case in Suncom conversion that the axis became inverted. My next try will be with this TLE.. sensor, despite the PCB building need, looks more "PnP.
|
|
#4287146 - 08/13/16 01:07 AM
Re: DIY hall sensor ...
[Re: f15sim]
|
Joined: Nov 2001
Posts: 3,955
Sokol1
Senior Member
|
Senior Member
Joined: Nov 2001
Posts: 3,955
Internet
|
|
|
|
|
|
|
|
|
|
|
|
|
Exodus
by RedOneAlpha. 04/18/24 05:46 PM
|
|