SpaceMouse Pro Teardown

[Charles Creations]

Hi Everyone,

I just did a teardown of a SpaceMouse Pro, and I wanted to share what I found and get some help with a question.

The main question for me is how they are doing the 6 axis position sensing. If you look in "BackOfSensingBoard," you will find three unmarked chips that look up at the pads on "InsideKnob." Are these infrared, capacitive, or inductive sensing? My first hunch is they must be optical because the top of the ICs are shiny smooth without markings, but the pads they look at are not perfectly smooth or very reflective. Capactive or inductive wouldn't make sense because the contact is non conductive plastic. These ICs have four pins; two are ground and Vcc, a third is an output, and the fourth in not connected. On the sensing board, there is a analog multiplexer and op amp, so the output must be analog I am thinking.

Thanks,
Charlie

[Rufus]

From their web site

3Dconnexion’s core adaptive micro-precision sensor technology was developed completely in-house with each 3Dconnexion 3D mouse featuring six high performance optical sensors which react to movements in the micrometer range.

I would guess the surface mount things are split IR photodiodes and there is one or maybe two IR LEDs behind the slots in the moving part.

[rx8pilot]

I am ok with just considering it "MAGIC". I love my Space Mouse PRO so much - primarily used with Solid Works. as I think of moving the model, it just starts to move in that direction. It did not take long for my brain to translate thoughts into very subtle hand movements.

[Charles Creations]

3Dconnexion’s core adaptive micro-precision sensor technology was developed completely in-house with each 3Dconnexion 3D mouse featuring six high performance optical sensors which react to movements in the micrometer range.

I would guess the surface mount things are split IR photodiodes and there is one or maybe two IR LEDs behind the slots in the moving part.

Awesome. Thanks for the insight.

There was no electrical connection to the moving part where the slits were, so I don't think there were any other light sources. Perhaps, it functions similar to an optical mouse, and the slit just gives enough contrast for it to track.

I was very impressed with the design as a whole. Built very solid and well laid out. The most difficult part to take apart was the sensor board because the springs in the knob were soldered to the board, and the springs functioned as big heat sinks making them hard to remove. Once I got them out, that whole mechanism in the know was pretty interesting.

[Charles Creations]

I just had the thought that they could have been using the springs as their electrical contact to lights up in the knob. This idea is supported by the connections of the spring on the top of the sensor board. That would be very clever.

[Nagle]

Compare this teardown. That one shows the sensing device, which uses photodetectors and LEDs. Did that part get omitted from this teardown, or what?

I just ordered one of these things. I want to see if I can make a 6 DOF force-sensing wrist for a robot out of it. Force-sensing wrists are useful but incredibly expensive, because they're not used much industrially.

[helius]

These devices are set up to measure travel, not force. The usual way to make a force sensor would be a silicon strain gauge. In theory there's a relation between the force and the displacement (as given by the spring equation), but in practice it can be tricky because of range of travel/dead zone issues.

[benjamin.lundgreen]

I have not opened up the newer version, but the old mouse worked with position sensitive light sensors (Position Sensitive Device, see the markings for PSD on the PCB)

The old version used 6 position sensitive light sensors (https://en.wikipedia.org/wiki/Position_sensitive_device) and 6 slits distributed along the edge of the round pcb.
As you pushed the mouse, the slits would move in relation to the leds and sensors so the light would hit the sensors at a different position.

[lucribas]

Hello,

You can check the US patent 20030103217:
https://patents.google.com/patent/US20030103217A1/en

There are detailed information about arrangements using opto-electronic.

The sensor is a PSD like this:
https://www.hamamatsu.com/us/en/product/optical-sensors/distance-position-sensor/psd.html