Hacking an Isometric Joystick

[Braeden Hamson]

I dug a little isometric joystick or "pointing stick" out of an old Lenovo. If you're still confused its a pressure sensitive controller. I think its four strain gauges wired up in a wheatstone bridge. As shown in this patent
http://patentimages.storage.googleapis.com/US6509890B1/US06509890-20030121-D00002.png

and full patent

https://www.google.com/patents/US6509890

I can't quite make sense of what they're talking about. I probed mine a little, there are 4 wires, the resistance between two of them was 2.550 k Ω. When i put pressure on the stick it changed to 2.600k Ω. That change seems a little small. That makes me think there has to be some extra trickery to get this thing to work.

So if you guys have any ideas, let me know. In the patent they're using voltage comparators any clue why?

 

[Gryd3]

I dug a little isometric joystick or "pointing stick" out of an old Lenovo. If you're still confused its a pressure sensitive controller. I think its four strain gauges wired up in a wheatstone bridge. As shown in this patent
http://patentimages.storage.googleapis.com/US6509890B1/US06509890-20030121-D00002.png

and full patent

https://www.google.com/patents/US6509890

I can't quite make sense of what they're talking about. I probed mine a little, there are 4 wires, the resistance between two of them was 2.550 k Ω. When i put pressure on the stick it changed to 2.600k Ω. That change seems a little small. That makes me think there has to be some extra trickery to get this thing to work.

So if you guys have any ideas, let me know. In the patent they're using voltage comparators any clue why?

Yup.
Build a sample wheatstone bridge, but don't measure resistance... Measure 'voltage'.
You've been very resourceful finding this information, but seemed to stop a little too soon to understand how the basics of how a wheatstone bridge works.

The whole point of these things is to be able to measure very small changes in resistance. This is done by applying a voltage to the bridge, and then measuring the difference in voltage from one side of the bridge in reference to the other.
It's much easier to watch and measure very small voltages, than very small changes of something much larger.

So... when a strain gauge is at rest, the two opposing sides of the bridge are balances, and the same voltage is present on each side. When a wheatstone bridge deforms and changes by approximately 1-2%, it causes in imbalance in the bridge which causes a very slight difference in voltage from one side to the other.
This slight difference is usually picked up and amplified by a voltage comparator, or op-amp to provide a more readable output.

 

[Braeden Hamson]

Thanks for the info. I just can wrap my head around where I'm supposed to supply the voltage. There's only 4 wires coming out of the device, and a metal plate it's mounted to, perhaps that plate is a ground but I'm not sure.

 

[Gryd3]

Thanks for the info. I just can wrap my head around where I'm supposed to supply the voltage. There's only 4 wires coming out of the device, and a metal plate it's mounted to, perhaps that plate is a ground but I'm not sure.

I didn't look at all the documentation you posted, so I'm going to wildly guess based on experience and current knowledge.

It 'scans' ... ie, it applies voltage top-to-bottom, and measures side-to-side to determine an X-axis reading. It would then apply voltage side-to-side and measure top-to-bottom to determine a Y-axis reading.
That's what I would do if I built it, and it's being controlled by a microcontroller which would make it pretty easy to do.

Keyboards do the same kind of thing to determine key presses... they can't read the whole thing at once, they rapidly read it in segments.
Although, there may be a trick employed to do the reading of both X and Y directions at the same time, but it would seem to be much easier to scan like described earlier.

 

[Braeden Hamson]

Ahh yeah great idea, I'll try hooking it into an arduino to see what it does.

 

[Braeden Hamson]

This is a tough nut to crack. I put ~1v across the gauge. I put it across different combinations of wires. I'll get voltage out but if I apply pressure to the different gauges, I get no change in the voltage across the circuit. I measured voltage across the whole circuit voltage across the gauge which is the same because it's the same node. I'm not using any additional components just an arduino as a convenient power supply the gauge and my meter. Same story with amps.

 

[Gryd3]

This is a tough nut to crack. I put ~1v across the gauge. I put it across different combinations of wires. I'll get voltage out but if I apply pressure to the different gauges, I get no change in the voltage across the circuit. I measured voltage across the whole circuit voltage across the gauge which is the same because it's the same node. I'm not using any additional components just an arduino as a convenient power supply the gauge and my meter. Same story with amps.

How are you measuring the voltage on it?

The change may be very small...

Take a look at the provided image, and try figuring out the 1-dimensional solution first.

Apply your voltage to the top, and a ground (or negative) to the bottom.
Your multi-meter (which may not be sensitive enough) would then be connection with the black lead on one side, and red lead on the other.
At rest, the meter will most likely show 0V. Don't worry if this is not the case. The result however will vary as pressure is applied to the stick. The voltage will go up, and down based on the direction of pressure.
If the meter is not senstive enough to view this... you will need to hook up a form of amplifier. The comparitors you made mention of earlier would be hooked up like your meter, and would amplify the potentially tiny voltage to be large enough to see or measure