Archive for BRIT NERF Home of the British NERF Community

       BRIT NERF Forum Index -> Modifications

Hall-effect cycle control idea

My issues with the cycle control switch placement in the stryfe being the cause for live center cycle control failure (by the time the switch closes to brake the pusher, the now-unloaded gear has spun around and momentum causes it to ever so slightly pull the pusher forward, releasing the switch & thus letting power flow to the pusher again, therefore runaway has caused me to realize that with the xsw in a stryfe, the cycle control needs to be forward, not to the rear. However, real estate in that area is scant and doesn't allow for switch placement without significant alteration. Basically, it's not worth it.

Unless there's another way. After reading minky's triple FET pusher/cycle control setup (Here: for microcontroller use, it occurs to me that there may be a way to set up cycle control with a hall effect sensor which can be used by itself to act as a switch much like the rapidstrike's mechanical switch, or with the addition of a 3rd n-channel FET be used with a board.

When a US5881 switching hall sensor detects the appropriate magnetic field polarity, it's output pin goes logic high (voltage equivalent to the input voltage Vdd) from an open drain. So a pull up resistor is needed for it. But when the field is removed, the output pin is grounded to the ground pin.

Using a p-channel FET to allow current to flow to the pusher when the trigger is released but the pusher is forward (place a magnet on the pusher where the cycle control switch would be pushed down by the RS pusher when it's retracted), since a p-channel allows current flow when the voltage over Vgs drops. The braking would be done by an n-channel also wired to the hall sensor output (rectifier diode would prevent the two FET gates from negatively interacting) and would short the motor when hall sensor output goes high, raising Vgs for un-channel causing current to flow thus shorting the motor, braking it.

To allow this to work for a board, add an n-channel to drive the p-channel and the output of the sensor wired to the board.

Yes, thus is more complex than the spdt cycle switch, but it'd allow for later board integration and also make physical arrangement of cycle control a little easier for future board integration. It also would mean a simple rewire to return full operation should the board die (provided its death doesn't take the FETs and hall sensor with it.

I'm looking at the data sheet for this sensor:

And got the idea from this page:

This is really interesting. I was thinking of something similar in an effort to solve the problem that is Honey Badger overrun with a Nchan and Pchan in conjunction to offer a dead centre braking circuit but with a magnet in the pusher arm and a couple of reed switches in parallel with the mech/pusher/fire control switch on the Nchan gate but I need to draw the circuit diagram to 'see' it. The idea being though that the reed switches stretch out a little over the pusher arm travel and so start braking earlier so the super fast, weak torque badger has an opportunity to stop.

Thanks for the note, though that is actually a diagram drawn by DesignForge at my asking when he showed his select fire RS off to the Modifications forum.

The problem with honeybadgers is that there isn't enough braking available from the motor to stop it. If I'm understanding ON correctly, it's inherent to the physical motor design itself.

A dead center arrangement with hall effect would be even easier. Have a n-channel with source on the negative feed from the pusher and drain on the positive feed to the motor. Use a pull up resistor between sensor vdd & sensor output & a pull down between n-channel gate and source. Wire pulled up output to the pulled down gate.

Theoretically that ought to work for dead center.

From my understanding it's not so much that the HB can't stop itself at all only that it's rubbish torque can't stop the impetus of the armature spinning at the high revs it produces in the timeframe the half bridge circuit/motor braking is active during the cycle. If you a) lengthen the time frame, b) lower the impetus c) increase the motor torque it will brake itself as, as long as it's spinning it will be producing back emf that will brake it until 0. Can't do c) without having a different motor (and for some perverted reason I want to see whether I can make it work), but we can do a) by increasing the window the half bridge is connected during the cycle (in my case via reed switches but a longer switch actuator could work) and we could do a) and b) by reducing the voltage to the motor and therefore the revs which would reduce the impetus of the motor and extend the time the half bridge is connected as the pusher arm won't be travelling as fast (I was thinking one or two 1.7v drop down diodes).

I need a bit of time to get my head around the use of the hall effect sensors but initial thoughts are 'great!' because for some reason I like collecting electrical components. Developing draws of them, just in case.

EDIT: Impetus/Impetuous - let's call the whole thing offff!

EDIT: I just realised each time I wrote impetus I should of written inertia. Well.. Can't be bothered to fix now.

Basically  yes you're right. Simplest solution to the problem is to use a board for brake control. Use a digital or pwm output to the FET which activates the motor short and use sensors to detect when in the pusher cycle the pusher is.

You could have the board using a hall sensor begin the braking (or start a braking sequence) when the pusher is at full extension. Begin the braking at that point so that it will be fully stopped by the time the pusher is at full return. However the timing in the code would require experimentation.

However there's a simpler solution to this issue. Don't use a honeybadger. If you want a 2s pusher motor for 2s or 3s use, get your hands on a fk-180sh-3240.

       BRIT NERF Forum Index -> Modifications
Page 1 of 1
Create your own free forum | Buy a domain to use with your forum