AR Thermal Limits - Meta

TLDR

Modified pain tolerance threshold testing – Meta

Background Context

A large tech client was creating a new augmented reality (AR) device, with one very critical new feature. The power required to run this core feature would heat up the device significantly. While the mechanical engineering team could model the prototypes to understand how hot they should get on their surface, they weren’t sure if users were going to be happy with the temperature. This led to the research request, of digging into the temperature tolerance for users with a consumer device.

In addition to the product problems, this was in the middle of 2019 when companies still had strict COVID protocols aimed at preventing the spread of the disease. These safety measures significantly impacted the ability to conduct user research with hardware devices due to the inability to get people into a room with prototypes, and requiring sanitization after each user. For complicated electronic prototypes, it is prohibitive to necessity thorough sanitization after each participant.

When it came down to it, I needed to figure out what temperature would be too high for people to use an AR device.

Project Context

The AR device is planned on being in contact with the user when in use. The glasses would be in contact with the face, and a separate computing unit would either be stored in a pocket, bag, desktop, or in hand while in use.

The stakeholders are worried that the device temperature while in use may approach levels that would cause a user to:

  • Think something is wrong.
  • Be unsafe
  • Be unsustainable
If the device is going to be too hot in it’s current form, then some major changes may need to be made, including changing the design of the product and/or modifying the core features to draw less power.
Research Goals
  1.  Determine use cases most sensitive to temperature.
  2. Determine temperature users would discontinue using device.
Action

Discuss limitations with stakeholders

  • No tools to measure heat flux
  • Safety, endurance, or comfort?
  • Does this need to be done now, during restrictions?

Method: unmoderated remote – pain threshold test

  • Not allowed into the office
  • Potentially dangerous
  • Unable to train and ship thermal prototype device
Adapting

Testing device needed:

  • Hot
  • Safe
  • Cheap
  • No training required
  • Readily accessible

.

.

.

.

.

HANDWARMERS!

First steps need to be characterizing the manufacturing tolerances of the devices before sending them out to participants.

I used a thermocouple DAQ with a few handwarmers at my home to learn the variance expected in heating speed and steady state temperature at each level.

I learned that the heaters would overshoot each level’s steady state temperature and then settle down.

Action

Develop research plan

  • Research goal refinement, recruitment criteria, data collection sheet creation
  • Develop the task instructions for participant

Work with the data custodian to ship handwarmers to participants

  • Allows us to maintain the participant’s anonymity and refrain from sharing the participants address with the Researcher.
  • The anonymity helps the participant feel comfortable sharing their honest feedback and reduces bias.
  • Starts the process of building trust with participants before data collection even begins.
Data Gathering
  • Participants received hand warmers directly from Amazon
  • Task instructions were emailed with a personal data collection sheet.
  • Participants were instructed to complete the tasks by the end of the week. 

Task Instructions

  • Start a timer at the same time you turn on the handwarmer. Hold the handwarmer while continuing to work. If it becomes too hot to continue, please immediately stop and record your time.
  • Please attempt each level for 1 hours, allowing the handwarmer to cool to room temperature between trials.
Analysis
  • Sync up participants quit times with the temperature range that was present on the characterized warmers to understand the probably temperature range when they stopped.

Results

  • The lowest temperature a participant stopped was in the 82-88F range.
  • Four participants stopped during the overshoot period of level 2, which occurred during 105-115F.
    • All of these participants were worried that the device was going to continue getting hotter.
  • The last two drops occurred during the steady state temperature of 100-108F.
    • Here they felt that the heat “got to them” over time
Impact

The high percentage of participants who did not endure the heat scared stakeholders into taking immediate steps to mitigate. The mechanical engineers (MEs) began working to redesign internal components and their placement in order to reduce the temperature felt on the surface area.

Discussions were also had between the designer and MEs to make changes to the design in order to provide more room for mitigating heat. Adding active cooling to the product via fans was also discussed.

Ultimately performing additional testing with the more accurate thermal prototype became critical. This study allowed the more accurate test to get green-lit as soon as COVID restrictions were lessened.

The end (for now)

The AR product shelved until the technology becomes available to operate the core feature without the product becoming unbearably hot for the user.

Learning / Retrospective
  1. Research not only improves the products created, but can also prevent further waste of resources on products that are likely to fail or potentially harm the users.
  2. Confidence in the user research helps persuade stakeholders to take more drastic steps.