While Google specified a handheld controller to their Daydream VR system, Samsung chose a different solution to Google Cardboard's limited interactivity: they added controls to the side of their Gear VR headset. Reading that description, I thought they just added a single circuit board with a few buttons. The reality was far more mechanically complex, resulting in a far higher parts count than I had expected.

This is a Samsung Gear VR SM-R322, compatible with a half dozen Samsung devices including the Galaxy S7 phone. I bought this as a present for a friend who had a Galaxy S7. Years later, he retired that phone and donated it to my electronics tinkering. (After resetting and wiping his personal data, of course.) He also returned my present, still unopened.

Since Samsung has already shut down their Gear VR experiment, there wasn't much I could do with it except to take it apart to see what's inside. I plan to keep the lenses and nothing else.

Gear VR side controls are dominated by what looks like a four-direction pad but is actually a capacitive touchpad. Off to the side are physical tactile up/down buttons (feels like volume control?) and a back button.

An adjustment wheel at the top changes the distance between its lenses and the device screen. (Focal length.) I estimate its range of motion to be roughly one centimeter. Its implementation turned out to be more complex than I had expected.

The face gasket was held with hook-and-loop fasteners. Peeling it off didn't reveal any fasteners or likely hiding places for them.

Pop off the front cover for a look the device holding mechanism.

Headset-to-device communication is done via this micro-USB plug, which can slide between two positions indicated by a green dot over either A or B. Most of device holding force is supplied between this clamp and its opposite number, which didn't have a USB plug. Helping to keep the device in place are small bumpers at each corner.

A thin rubber pad adjacent to those bumpers hid fasteners, one for each corner. That hole next to the Philips-head screw is a part of the focal length adjustment mechanism.

Two additional fasteners were in the middle, hidden under a sticker.

Once undone, the device holder tray can be flipped open though it is still connected to the headset by this cable for USB connection and two of the side buttons. A flexible cable is required to bridge a gap whose size varies based on the focal length adjustment.

Freeing the USB cable required removing a cover to access all electrical connections to the mainboard.

Then the spring-loaded USB connector assembly could be freed and disassembled.

Looking inside the device holding tray, I realized the corner bumpers were more sophisticated than I had expected. I thought each corner was a single piece of squishy rubber, but it's actually three distinct and individually spring-loaded piece of plastic. Multiply this out to all four corners and we end up with twelve sets of bumpers and springs.

Back to the headset main body, removing its mainboard allowed us to open it up to see the focal length adjustment mechanism.

Rotational motion from the top adjustment knob is transmitted via a series of gears and shafts to all four corners.

A close-up look at a corner mechanism translating rotational motion to linear motion.

Under the mainboard is a piece of black plastic that I had expected to host the capacitive touch pad, but it only had the back button. I can see four wires in this cable, which is double the amount I expected. A simple switch should only need two?

The capacitive touchpad is actually glued to the inside of the enclosure. Its control chip has the following markings:

IMAGIS
IST3026C
159V37
1614

A web search on these identifiers led to an iFixit forum post, which pointed to IST3026 touch sensor product page of South Korea-based Imagis.

The glue was tenacious and thus touchpad was damaged during removal. I don't recall ever seeing this pattern in a capacitive touchpad before.

The final bit of electronics is a sensor that sits looking at the user's forehead. Its location implies an optical proximity sensor to see if it is being held up to the user's face.

The lens retention mechanism has three little clips that need a little push to release the lens.

There's nothing fundamentally complex about adding side controls to a Google Cardboard-style headset. But when we add mechanisms to securely hold phones in a range of sizes plus the ability to adjust distance between lens and screen, we end up with a mechanically complex device with a high parts count. It worked really smoothly, though, perhaps all those parts were necessary for proper operation and going with a low part count design hurts functionality.