Octagonal gate for Sanwa JLW series

[elkameleon]

I have a couple Sanwa JLW's (J-Stik) I got from Ultimarc. I've been getting into Street Fighter as of late, but I find it hard to pull off moves that require quarter circles and such because I cant feel the directions with the standard square gate, so I decided to make one... sort of. I uploaded it to Shapeways, so It can be purchased for 12 bucks, but their minimum order is $25, so you might as well get 2! I haven't ordered it myself to see how it works, I just don't have the cash to do so this week. If someone feels like they wanna check it out, let me know how it goes!

EDIT: I see that Ultimarc actually sells one they made themselves for $14. Sorry Andy, mine looks cooler
http://shpws.me/1MfS

[DNA Dan]

Very interesting. What is this site? Can you send images and have them prototyped for you? I'd like to get the square gate restrictor with gear teeth on the outside so that I can attach a motor to it for 4/8 way automatic switching.

[elkameleon]

I designed the part myself, I just uploaded a 3D file. If you would like me to design something for ya, i'd be willing to give it a shot.

[DNA Dan]

So the idea is to basically keep the same dimensions. The only areas that need to be changed are:

1) The teeth on the outside. I measure these to be about 32P 70-72T. I am thinking an RC car pinion gear will run it as is if the grooves were "deeper" and more shaped like a gear tooth. This may push the teeth count to 70T or less, but 32P seems about right. Here is the pinion gear I had in mind: http://www3.towerhobbies.com/cgi-bin/wti0001p?&I=LXXF68&P=M

2) I think the disk could be all solid. This would simplify it and give it more strength.

3) Around the post holes, I would set these up for bearings, so the action is smooth. I don't have the posts in front of me to measure, but they would have to be small enough so that they ride between the plate and the lip on the post. The problem with the way these restrictors are made is that the posts allow for rotation AND keep the plate in place. I would need to think about the design for this a bit more.

I'd love to see anything you come up with! 

[DNA Dan]

You know there was some discussion about this a while back being done with a servo. Randy tested a setup and concluded that cheap servos just would not hold up to the rigors and times of constant gameplay. Perhaps a more expensive servo is an easier way to go? You just need to be able to screw in the "ball" end of the servo linkage to the restrictor plate. A few issues with this setup are:

1) It's nice that it would "hold" the plate in it's proper orientation, however how long can one feasibly keep a servo powered without shortening it's life expectancy?

2) They make more "heavy duty" servos, but would these even hold up to the duration?

In the previous post I described using a motor for the rotation. The idea here is to use a "braking" function on a motor controller to keep it in place. Although not the cheapest or easiest route, it would probably be more durable. I guess it just depends on what the minimum cost/quality servo you would need for the alternative setup.

[elkameleon]

I guess it really depends on how tightly you screw down the restrictor honestly.

[DNA Dan]

That's the problem with that design. If you make them "loose" they are prone to come out over repeated turns of the plate. Also the whole gate can feel mushy. It's almost as though the posts need a redesign so they hold the plate better but allow for a bearing to be in there. The issue is there are two directions of forces at play here and they are perpendicular to one another. There's the force required to keep the plate flat and solid against the joystick shaft, then there's the force that is needed to twist the plate. I don't know how this would look but I think the posts need to be "tightened" in their own right irrespective of the plate. Then the plate could just "ride" between two nylon washers or something.

[elkameleon]

I'm not sure what you mean by "posts" so here's an exploded view of the Sanwa JLW, the only thing missing is the springs... You'll have to right-click it and open in new tab to get the full resolution.

[DNA Dan]

I am talking about a U360. Which is essentially a modified JLW. The screws holding the restrictor plate are replaced with four posts so that you can screw one end through the restrictor plate and the other end holds a PCB. You can see this about 2/3rds the way down this page http://www.ultimarc.com/ultrastik_inst.html. Also, the switches are removed since the PCB allows for digital and analog mapping.

I think I will try and come at it with an all metal gear servo. I need to finish my cab first.

[elkameleon]

I haven't got a U360 fro Andy to draw up in SolidWorks yet, so I don't know too much about their construction, other than they share the same base, like you said.

[emphatic]

Thanks for showing this site. Looks like a great resource for lots of arcade applications.

[Gray_Area]

Thanks for showing this site. Looks like a great resource for lots of arcade applications.

Yes, I've been thinking along the same lines.....

[SavannahLion]

Thanks for showing this site. Looks like a great resource for lots of arcade applications.

Yes, I've been thinking along the same lines.....

Too bad there's a size limit :p

[Nephasth]

Anyone ever purchase this? At $12 they don't seem like much of a deal compared to Ultimarc's octagonal restrictors considering they come with additional hardware to mount the U360s pcb with the restrictor installed (the hardware isn't needed in any way when using the restrictors with JLWs though). I've got Ultimarc's octagonal restrictors mounted to my 2 JLWs, and in combination with their hard springs, the sticks feel MUCH better than with square restrictors and soft springs. Would like to see an actual pic of these for comparison with Ultimarc's, also curious about how the "white strong & flexible" (~$12 option) material holds up under use.

[Gray_Area]

There's a problem with an octagonal restrictor. The points of actuation are where lengths meet. Not in the middle of a length. Hence, in rotating the octagon 1/8th turn there is an incongruence between the angle between lengths, and the (perpendicular) angle of the directions pushed.