A DIY push-pull spinner

[xefned]

DIY push-pull spinner.

Starting with the "QuickSpin" plans from the Project Arcade CD-Rom, I'm planning to add a spring to make it bounce back and a couple of microswitches for the up and down positions. To keep it simple, I'll let the encoder wheel ride up and down as the knob is pushed and pulled and widen the normal gap between the LED and detector so the wheel doesn't touch anything. I've drawn it out on paper and I don't see why it won't work.

I'll document it here for future generations of mad planets aficionados.

[PL1]

So you're using the encoder wheel to press the microswitches?

My earlier post on this topic still applies.

You might be better off having the entire spinner assembly move up and down instead of just the knob/shaft/encoder wheel.  That way, there are less problems like the variable gap in the optics,  the friction between the encoder wheel and the microswitches when you push/pull and turn at the same time, or the encoder wheel snapping due to a heavy push.

Good luck with your build.   


Scott

[xefned]

So you're using the encoder wheel to press the microswitches?

Good luck with your build.   

LOL!  You crack me up.   

Don't worry, your earlier post was well-heeded. Studied even. Activating the switches with the encoder wheel is a hella no-no. That's just asking for trouble.  A major design goal of this project is to avoid those less-robust construction methods I see in other DIY designs - like rubber bands, super glue, coat hangers, etc. (No offense to the chaps who employed them and shared their designs; I'm just going for low maintenance - build it and forget it.) But still keep it cheap. Man am I high maintenance or what?

My real stumbling block is finding a proper encoder wheel. It would be nice to find a pre-manufactured metal encoder wheel with a bajillion teeth.

The only thing I can find is this 75¢ plastic trackball replacement wheel from groovy. It's low tooth count appears to have worked in the QuickSpin design. But I imagine it requires tweaking the sensitivity in Mame to get good response, and probably still won't cut it for games requiring a high-pulse count per revolution.

[PL1]

My real stumbling block is finding a proper encoder wheel. It would be nice to find a pre-manufactured metal encoder wheel with a bajillion teeth.

I used a high resolution laser gaming mouse to make a pinball shooter (see sig) that could be adapted to a spinner.  You might even be able to use an older 20 IPS (inches per second) optical mouse that can handle ~13.3 revolutions per second on a 3/8" shaft. (~1.5" circumference)

The laser can read either the shaft like the shooter does or an attached flat wheel.


Scott

[xefned]

I used a high resolution laser gaming mouse to make a pinball shooter (see sig) that could be adapted to a spinner.  You might even be able to use an older 20 IPS (inches per second) optical mouse that can handle ~13.3 revolutions per second on a 3/8" shaft. (~1.5" circumference)

Cool, I'll rip up a few old-school mice and see what I find.
I plan to set aside some time to read through your DIY pinball wizard thread in it's entirety. I love all the details.

I photoshoped up my drawing.

DIY push-pull spinner concept drawing
Measurements are rough estimates, based on parts found in McMaster-Carr catalog. Design is untested. (for now anyway.)

The little boxy things with nubs are supposed to represent microswitches–vague, nebulous, dreamlike microswitches with little basis in reality...  Pretend like they are much closer to the large-diameter washer, and can be activated with ~2mm deviation in either direction.

[PL1]

Looks like an interesting design, but my main points of concern are these:

1. Friction against the microswitch nub while pushing/pulling and turning at the same time causing wear. (plastic vs. metal)

2. The encoder wheel 4mm vertical travel range (+2mm to -2mm) may cause problems like hitting the sensor or with the sensor/wheel distance unless the sensor can shift up and down with the encoder wheel and shaft.


Scott

[Mysterioii]

Looks like an interesting design, but my main points of concern are these:

1. Friction against the microswitch nub while pushing/pulling and turning at the same time causing wear. (plastic vs. metal)

Use the microswitches that have the roller on the actuating lever...  example:

http://omvol.en.alibaba.com/product/436002015-213182615/Short_Hinge_Cross_Roller_Lever_Micro_Switch.html

[PL1]

Looks like an interesting design, but my main points of concern are these:

1. Friction against the microswitch nub while pushing/pulling and turning at the same time causing wear. (plastic vs. metal)

Use the microswitches that have the roller on the actuating lever...  example:

http://omvol.en.alibaba.com/product/436002015-213182615/Short_Hinge_Cross_Roller_Lever_Micro_Switch.html

Which increases the actuation distance ==> more of point #2.   

If the roller was over the nub instead of at the end of an arm, it would solve the increased actuation distance issue.  Some arms are removable -- might be able to cut/re-bend it shorter.

Radioshack has these.



Scott

[Mysterioii]

Yeah you just bend the lever arm, put it on pretty much any level you want.

[PL1]

Yeah you just bend the lever arm, put it on pretty much any level you want.

The problem isn't the height, it's the length.  Shorter arm requires less vertical travel (blue) than the longer arm. (red)


Scott

[Mysterioii]

Jeez, pay 50 cents for one and experiment.  For one thing, the pivot point isn't where actuation of the switch actually takes place, there will be a little nub under the lever arm (approximately where you drew your blue line) just like the nub on the regular switches in buttons and joysticks... the lever pivots at the corner but presses down on the nub.  All the ones I've ever encountered are usually adjusted so the lever is already resting on the nub... you barely have to press them at all to get them to actuate.

When you bend it, you can bend it to the left of the nub (left and right being ambiguous, but I'm speaking in reference to the drawing)...  adjust is such that there is already pressure on the nub, just on the threshold of actuating.  Then you'll barely have to depress the roller regardless of how long the arm is.

[Xiaou2]

The original design is far better...   but its also not so easy to produce.

 Try this:  A simple 4" diameter cylinder, connected to the main shaft.   Maybe some toy wheel or roller, which has a perfectly centered hole already on it.

 Use a laser mouse to read both X and Y values, by placing it close to the side of the cylinder (not the top or bottom)  de-cased if possible.  A decent one, which can read a few mm even when lifted off of a flat surface.  Logitec versions can adjust DPI sensitivity pretty easily.  Im pretty sure mame allows mouse input for the vertical axis on DOT. 

 If not, you can just use the X axis for the mouse, and add switches for the Y.  However, Id recommend Leaf switches, as they can be adjusted to react far easier and quicker, are far more durable (wont wear like a plastic micro-switch actuator... and even the smaller roller switches are not that smooth, nor long lasting)  ,and you do not have to overcome that click-pressure threshold before activation.

 In Aiming levels, aiming should be quick and smooth/fluid.  Its hard enough with the real controller, which has almost no frictional changes when pull / pushed + spinning.

 The alternative, could be to use a regular spinner... but use a foot-pedal setup to control height changes.

[pinballjim]

Yeah, Xiaou's got a good point there.  If you use an optical mouse you can get X and Y inputs.  I did it myself with some washers, nuts, and a carriage bolt.  Spinning the bolt got you X by rotating the washers in front of the sensor.  Lifting it up and down got you a Y.

[PL1]

Mysterioii -- I'm not crapping on your idea, just mentioning that arm length can change the travel distance, since they are available in different lengths.

One like this is what I was thinking about.




Scott