Diode should be placed with strip pointed for Row's lines:



So, your 4x4 matrix looks OK in this regard.



This button lock down when pushed for ON, like a toggle switch?

If not I think you need an additional circuit.

I don't have experience with LED's, but for keep the LED on is need multiplexer circuit, since an normal push button with just make a momentary contact, making the LED just blink.

This scheme with optocoupler i/c is for BU0836A matrix, but you can use in MMjoy2 as diodes point for Row's.



Notice that in this scheme power source (+5v, Gnd) for LED is external to matrix.

But await an option of people that have practical experience with LED's.

thanks Sokol1 for your detailed reply

Yes the diode is going to the row, I use the same colour code for the wires than the documentation in page 3 or 5 of this thread

And yes the "Push On Push Off (POPO)" is a toggle switch (I was looking for that Word) so the LED is supplied once the button is on ON position
I will try like that, I dont think I would kill the Arduino ?
Any one with advises for Toggle switch and a status LED are welcome to comment

all the best, and thanks again

I don't know if can damage the Arduino...or disturb the matrix...

Just for one LED the ideal and safety solution is a DPST ON-OFF toggle switch, since have two lines of pins (is two independent switches controlled by one lever), you can wire LED with his own power source, can take for RAW pin.

Like this scheme for BBU0836:

S!

Plank here from Rise of Flight. Posting on the RoF forum as KNALP. ( it is a long story...)

Здравствуйте !

I have a question:

I have hooked up a 10k ohm VR to Pro micro ADC input and tested that. Very good. Fine.
Full scale from peg to peg. 000. to 100% . Awesome.
No auto calibration. The thing works very well. хорошо.

I then tried out an Allegro UGN3503UA Hall sensor and it did not go full scale!
It went down to about 20% and up to about 80% .
I tried the calibrate without centre but it was too sensitive and feel bad.

I do not know how to proceed.

To get full 100% of ADC ??? ( I prefer not to windows calibrate this out... fake and not good solution )

Is there a hall effect sensor that works properly with the Pro Micro?
( I mean will it give 000 to 100% signal...not 020% to 080% .)

Thank you for all your work! All involved! Hello from NEW ZEALAND !!!

S!

Plank. AKA Knalp.

Kbalp,

With an hall sensor like this UGN3503uA (Allegro A1324LUAT, Honeywell SS495A1) the placement of magnet relative to sensor is critical, requiring much trial and error until find the ideal solution.
Since this is a DIY thing, require patience and time. Your goal is achieve a more close possible 0 to ~5 v variation, so play with distance from magnet relative to sensor.
The's no an "best" recipe for do this, you need find your own way.

In RoF forum are a topic of guy replacing the Hall sensor in Simped rudder pedal and he describe how find the center position, a laborious process.

UGN350*** is a lower sensitivity sensor - 1.75 m V/G, what make the adjust more difficult, than for example Allegro A1324LUAT.

I have good results using 2 magnets place similar in this GeneWeb assembly "the Bic Hall":
https://www.simpits.org/geneb/?p=299

MMJoySetup has calibration tool, with axis without center - good for throttle axis, and with center, as well "autocalibration".
I don't like this because in when you turn computer the joystick (or throttle...) is not physically centered a new and incorrect center for axis is defined by autocalibration.

An ready to use hall (is like a potentiometer) like Bi-Tech 6127 series make install more easy, but they cost more.
And important, you need get the model with the "efective electric angle" more close to the angular movement of you controller axis (stick, throttle, pedal...).

An example, the 60º model, what is defined by the code after the A (A60). This series has model with 30, 60 90... 300 degrees.
https://br.mouser.com/ProductDetail/BI-Technologies-TT-Electronics/6127V1A60L5?qs=sGAEpiMZZMs0JOhy9PM0UT2KuCf1H3wX0A4NJp8zcxs%3D


A promising model is AS5600, that allow set (once) the "effective electric angle" for match you controller angular movement. See this example of use.
https://forum.il2sturmovik.com/topi...oject/?do=findComment&comment=881857

Magneto resistance sensors are more easy to use, since the "effective electric angle" can be determined by resistors in PCB. Rel4y sell kits of then - in potmter format in DCS forum.

The "best" sensor for use with MMjoy2 is TLE5010/11, but work in digital mode and use 6 wires.

https://aliexpress.ru/item/32634095245.html

http://avimator.net/229wiki/index.php/TLE5010

Привет Sokol.

I have been trying with different magnets etc over the years and always found this problem.
But now I see why there is such a fuss:

UGN350 - 1.7 mV/G =/= A1324 - 5 mV/G

Will try the "BIC HALL" with ugn350 and Promicro until I can source A1324.

For what it is worth, using the cheap VR did have a pleasing feel.
Some kind of turn ratio device to change stick throw to full VR throw...
When the VR gets static you just replace? They are cheap....

I am impressed with MMjoy !

Problem: Win7 seems to think that the MCU is a keyboard and mouse and controller etc all at once.
and the name has strange characters. Will post a picture soon. Maybe I did something wrong?

thank you for your time and effort!

S!

Knalp.

Looks like that you are using an old version of firmware, in what MegaMOZG try implement mouse and keyboard emulation, but remove latter.

Latest versions available (from 2016 - when MegaMOZG goes work for VirPil) is there:

https://sites.google.com/site/mmjoyproject/

Привет.

I made some tests with the Allegro ugn350 sensor:
No good result.
It will not go full 0 to 4.5 volts. No matter how I use magnets etc.
"bic hall" was not much better than a plain magnet !
( it does look cool though...)
I stack magnets together to increase gaus but no better.
The sensor must be saturated. and no more voltage
difference can be achieved...... ??? I try quite hard!

Question:
Will any hall effect sensor actually output full GND to Vcc ???
I do not want to use any "tricks" to fake it. It must work properly.
( RFC demand best use of available means! )

Thank you for point out MMjoy version. I did not see newer one in the list till I took a better look.
( it is hard trying to tell which long string of numbers is larger. )

Today I will order some A1324 and see how well they work.
I hope they can use all of Vcc !
Maybe they will arrive next month .....

Back to the humble VR.... at least they work all the way!
When they are new they are very good!

S!

Knalp.

Hello.
While wait for parts.... trying out VR mech prototype.

S!

Plank.

Oh, the bottom left picture. The large wheel shows the area ( shaded )
that can be removed. Also the cord, belt, wire, chain, string, thread can be
captured on the cams, wheel, round bits so that it does not slide of come lose.

A large main cam will decrease the throw angle. At this point I think for a full size
stick it is too big... S! P.

Check out the PCB's for MMJOY to make life a lot easier
https://SimHQ.com/forum/ubbthreads.php/topics/4538231#Post4538231

[img]https://ibb.co/7jsvdWz[/img]

Any ideas? Works fine on one of my laptops

Does the Pro-micro show up under windows device manager? And do you use a data cable? Many USB cables only work for charging, no data.

I plug in a premade mmjoy2 device and it also doesn't work on my pc. (works fine on the laptop also)

When I plug in the mmjoy2 based device it shows up in device manager as "HID Compliant game controller" and even shows up on the windows usb gamepads but when I try and open it's properties it gives an error




Works fine on a different windows 10 laptop.

used USBDeview to remove all registry elements of the mmjoy2 device and it still gives an error in windows USB gamepad properties.

It shows up perfectly fine in VKB joystick tester though.

This is issue of (obsolete) Windows Game Controllers.

Because you have changed some parameters in the device (Arduino) in another computer, e.g. number of axes and or buttons, VIP/PID and or device name.

In Windows Register remains the old reference, then the error in Game Controllers > Properties. So delete in Windows Register for old controller name references, can search by name or VID/PID before plug.

But for VKB Button Tester (included in MMJoy1 package) or games controls, the controller is fully operational.

Sokol1,

Has Megamozg released any more info about a newer version of mmjoy2? Sometime ago when you helped me through a problem and we determined I had to stick with a much older release you said he mentioned on the il-2 Russian forum about a newer version.

No, last year are a talk about a new version of firmware increasing Encoder support, that actually is just six, and other features.... but nothing more is said about.

Probable with advent of Freejoy and the memory limitations* of ATMEGA m/c compatible with MMjoy2, this is now an "abandonware".

The last version still MMJOY2 [v20160818upd1] (from 2016).

https://sites.google.com/site/mmjoyproject/fajly-dla-skacivania

* Although if removed that LED output that few DIY use (compatible only with War Thunder), more features could be added.

All Mmjoy2, single Arduino micro

Hi Sokol1,

can you elaborate why TLE5010 is the "best " sensor to use?
is it better compared to Allegro A1324?

Because:

- Send information for USB controller in digital mode (binary code) instead analog mode (voltage variation).
- Has 14 Bits ADC
- Is cheap, and you find already assembled in PCB in Aliexpress (PCB quality is... "Chinese").
- Is easy to set relative to magnet, contrary to Allegro A13245 with is difficult find the center of magnetic field of magnet, being more practical for use in gimbals for example.

You can achieve good result with A1324 and similar but these require patience to set.

thanks Sokol1 for the reply.
being in digital mode, does it make the sensor more precise compared to a1324? or about the same?


I find this picture

http://SimHQ.com/forum/ubbthreads.p...fs-throttle-2-0-usberization#Post4348201

is this only for one axis?

if I have three axis, where should connect the additional axis?