A DIY VR headset can be a simple smartphone viewer or a much more involved PCVR build with dedicated displays, sensors, firmware, and tracking software. Pick the project scope first: the parts and assembly work are very different. The PCVR route can connect to SteamVR, but one builder’s account also shows how sensor setup, fragile displays, tracking drift, and fit can complicate the job.
Two ways to build a DIY VR headset
A phone viewer uses a smartphone, a cardboard enclosure, and two lenses. The method described in Design News’s cardboard VR tutorial includes making plastic lenses and assembling the housing around the phone.
A PCVR headset uses dedicated display hardware and electronics. Tim Stevens’s Relativty-based build combined two LCDs, a display controller, an inertial measurement unit (IMU), an Arduino Pro Micro, lenses, and a printed shell. An IMU senses changes in the headset’s orientation. His setup also used SteamVR.
| Build path | Main components | Assembly work | Described use and software |
| Smartphone viewer | Smartphone, cardboard enclosure, and two lenses | Make the lenses and assemble the housing | Smartphone-based viewing |
| PCVR headset | Dedicated displays and controller, IMU, Arduino Pro Micro, lenses, printed shell, strap, and face pad | Breadboard testing, firmware setup, calibration, and careful display installation | SteamVR with IMU-based head tracking |
The PCVR parts above describe Stevens’s particular build, not a universal parts list. For a more detailed project recipe, the CNCDan DIY VR README lists a separate Version 2.0 design and its assembly requirements.
What the PCVR project requires
Stevens used two 2.9-inch LCDs, each with a 1440 × 1440 resolution, plus 34 mm lenses. He reported a 60 Hz refresh rate for the assembled headset. In a separate design, the CNCDan README lists 2880 × 1440p and 120 Hz as Version 2.0 specifications. Those figures belong to different project descriptions; don’t combine them into one configuration.
The CNCDan README also lists two NRF24L01 modules, a tracking ball, a calibration switch, rods, bushings, an Arduino Pro Micro, lenses, a strap, a face pad, threaded inserts, and screws. Its assembly instructions call for soldering three surface-mount components and wiring the calibration switch, tracking LED, and display-driver power. Wireless controllers are linked to a separate project.
That extra hardware matters when you’re planning the build: the headset, tracking setup, and controllers are not a single interchangeable parts bundle. The README lists 3DoF tracking, which tracks head orientation, and an additional 6DoF setup through PSMoveServiceEX, which also tracks position. It treats 6DoF as a separate setup, not an automatic result of assembling the basic headset.
Test the electronics before fitting the shell
For Stevens’s build, the sensible order was to test the wiring on a breadboard before installing parts in the printed shell. The IMU connects to the Arduino for power and data. Stevens advised powering the display controller separately, using USB power or a USB splitter, rather than relying on the Arduino to supply the displays.
For exact wiring and project files, follow the instructions for the design you choose. The CNCDan README and the Relativty-based build are distinct references, so their component details and specifications should stay with their respective projects.
Match the firmware to the sensor and protect the displays
A sensor board’s label may not tell the whole story. Stevens ordered a GY-91 advertised with an MPU-9250, but reported receiving a board with an MPU-6500 instead. The MPU-6500 lacked a magnetometer. Changing the firmware target to match the sensor resolved his reported compile failure, though he also reported poorer tracking with the sensor he received.
The display installation was another delicate point in his build: clips in the eye boxes broke both exposed LCDs. After modifying the boxes to remove or omit the clips, he held the displays in place with thin painter’s tape. The practical lesson is to test the fit gently before securing the panels; replacing the screens added $62 to his reported expenses.
Once the IMU’s orientation changed during installation, Stevens had to recalibrate the headset. Check the sensor model and firmware target before assembly, then recalibrate after the sensor is installed in its final position.
Tracking and comfort depend on the finished build
In his assembled headset, Stevens reported drift, crossed axes, and frequent recentering while using iRacing. He found the 60 Hz image adequate for casual practice, but also reported visible pixels, cable tug, and pressure on his nose and cheeks. He was ready to stop after about 15 minutes. Those are his observations of one build, not results for every DIY headset.
The reported cost also belongs to that build. Stevens put his initial component spend at just under $180 and his final outlay at $239.93 after replacing the broken displays. Treat those figures as a record of his project, not a current parts quote or a general cost for PCVR.
Before settling on a design, decide whether you want a phone-based viewer or a headset intended to connect to PCVR software. The first route centers on a phone, cardboard, and lenses; the second brings in display electronics, firmware, sensor calibration, and a tracking setup whose capabilities depend on the project.
