Avegant Glyph

One wearable, two ways to use it

Glyph combined headphones and a personal display in one wearable. With the band lowered across the eyes, the earcups, nosepiece and optical adjustments had to work together to keep the display in position. Raising the band returned the product to headphone use. Fit and adjustment stability were central mechanical requirements.

My work on first-generation Glyph covered the interfaces between the wearer, headband, earcups and moving internal cable. It included spring and liner development, audio-cavity and earpad prototypes, cable routing and abrasion work, supplier coordination and validation. Button and arm actuation effort, headstrap changes and liner adhesives were also part of my mechanical remit.

Developing the liner, spring and earpad together

Beta feedback described tightness, pressure from the top padding, nosepiece difficulties and repeated adjustment. Those observations gave the fit work a practical starting point: a wearable could hold position and still be uncomfortable, or feel acceptable in one configuration and need adjustment in another.

The liner had to follow the band as it opened and returned without puckering or allowing its flange to escape. PMP’s January 2016 proposal combined a flexible carrier, rubber skin and foam, with edge bonding and clips. Those were proposed construction choices intended to control the liner through movement; the proposal did not establish a final production selection.

The January development plan connected that construction work to experiments and supplier feedback. It assigned me the experimental and test plan, Misha the 3D work, Jill and me assembly and testing, and Chris and me analysis. The planned feedback loop carried the results into revised geometry, Voke’s manufacturing review and engineering changes.

By April, four builds paired 1.2 mm and 1.4 mm springs with 65A and 75A liners, and their force curves had been measured. The 1.2 mm spring was judged insufficient in that comparison; the 1.4 mm spring appeared viable, with further structured testing planned on production-level units. In parallel, the in-house audio-cavity prints were complete, Steph’s testing was underway and the earpad request for quotation had been sent. The prototype work connected physical fit to the earcup and cushion architecture.

Finding where the force comes from

The complete assembly supplied more than the spring’s force alone. Jill’s March measurements compared spring diameters, while April’s deconstruction work compared complete units with configurations that removed liners, springs, optics or the faceplate. Separating those contributions made the assembled structure part of the experiment. It also meant that a measured force belonged to a particular assembly, width and telescope position.

Earpad grip used a different experiment. Pad materials were compared on an acrylic fixture using added load and slip distance. The test record warns that the fixture could produce more suction and contact area than a wearer’s head. Those comparisons could inform material evaluation, but did not establish wearer comfort, retention in use or the production pad choice.

Keeping the cable working as the arm moves

A cable inside a telescoping arm must accommodate repeated movement while avoiding kinks, rubbing and damage near its terminations. Glyph’s cable work considered wrapping, local support, routing and strain relief as interacting design choices. The life-test procedure moved the arm back to its initial position and checked cable wear and kinking at staged intervals, with photographs and reassembly along the way.

The final March reports recorded both passing and failing cohorts under a 3,000-cycle procedure:

March report	Specimens	Detailed recorded result
T1, March 5	1-10	Specimen 9 could not return at 500 cycles.
T2, March 5	11-20	Wear and kink checks recorded OK through 3,000 cycles.
T3, March 16	21-30	Specimen 25 could not return after 1,500 cycles.
T4, March 16	31-40	Wear and kink checks recorded OK through 3,000 cycles.
T5, March 16	41-45	Wear and kink checks recorded OK through 3,000 cycles.
T6, March 16	46-50	Specimens 49 and 50 could not return at 250 cycles.

Tracked Glyph telescope-arm tests by dated specimen cohort. In the March 16 T6 cohort, specimens 49 and 50 of the five tested could not return at 250 cycles. These are dated specimen results, not a production failure rate or a sequence of tooling revisions. The reports contain conflicting summary and comment fields. Later cable concepts reused the T labels and cannot inherit these passes.

Moving changes into hardware

The April update recorded Fujikura’s six-variant testing as complete with no winner and the splint approach rejected. It also recorded released mechanical changes to parts 425-0001 and 425-0006, and work with Amphenol on a tube or sleeve approach. I was the named owner for the cable-kinking, chassis-abrasion and second-source work.

Abrasion work combined CNC chassis modifications, cable rework and a new routing scheme, with longer-term tooling and cable-design changes planned. Validation was underway on 20 headband-lifecycle units and 10 drop-test units. The separate Amphenol effort had 60 samples in production, with the validation protocol still to be defined internally and staged with Intretech. These were active validation tasks, not recorded qualification passes.

The cable record continued into August. Fujikura Electronics (Thailand)‘s revision 13 drawing, 850-0001-13 / FAW-1189F, records changes to branch length and protective wrapping. The drawing defines the revised cable and its requirements; it does not report a qualification result.

My contribution connected the fit experiments, prototypes, supplier alternatives and released mechanical changes within the same product-development effort. The work had to account for the wearer, the movement of the mechanism and the way the hardware would be assembled.