Bella & Chloe Lightbox

Product Designer (Apr — Jul 2014)

New Storefront

Shortly after starting at Bella & Chloe as a product photographer, I gained the opportunity to help lead in the design of a new storefront, and re-invent the photography process in a way that prioritized quality over quantity.

A key component of that goal was the need to produce high-quality product photography while maintaining a far more rapid pace than you would find in most professional product photography settings.

In order to accomplish this, I set out to design my own custom photography light box.


A diagram of my new approach to sorting jewelry before taking photos, which helped inform me of the different sizes and types of light boxes necessary.


Research & Testing

While researching ways to improve the process, we first found a company selling 360 degree photography light boxes and bought one.

Ultimately their software was too slow for our needs, and was ultimately built for entirely different needs. Not to mention the price was prohibitively high.

We tested the light box for about a month, and researched other options. We came to the conclusion that what we needed was a not on the market. What we needed was something similar, but with much of the unnecessary technology stripped away and the cost substantially lowered.

This is when I set out to reverse engineer the light box, and completely rebuild my own version of it at one quarter the cost. With the software controls replaced with manual potentiometers, the boxes would be much faster to use.

Before anything else, I wanted to know how the machine worked. So I dismantled it, studied the wiring, PCB, and all the part numbers on all of the surface-mounted chips. I looked at textbooks and researched online.

Going into this, I didn’t know much of anything about electrical circuits, coming out I felt like an amateur electrical engineer.


Planning & Modeling

Once I had a rough idea of how it worked, I made a list of the functions we needed the new box to perform. We needed the same seamless walls of light, manual controls, and three form factors to accommodate vastly different types of jewelry.

While the electrical parts shipped, I got to work modeling the body

I measured every flat edge and panel on the device and made paper models to test if the parts would fold together properly. I then drew vector files of all of these parts that I would later use to cut sheets of plastic panels. I created more complicated parts with CAD software that I could later print with a 3D printer.



Large sheets of plastic were cut down to the panels and placed together. The hinges and other oddly shaped parts were 3D printed from CAD models. Once the entire housing was assembled, I moved on to the electrical components.


Electrical Engineering

Before I could move onto anything else, I tested my if circuit diagrams actually worked using perfboard. Once I knew what I needed, I was able to calculate the amount power needed for the entire unit.

But the amount of power I would need was prohibitively high, and the heat generated would be dangerous. So I had to study a bit more about electrical engineering.


After consulting some electrical engineers, I learned that I needed to use gating to provide more efficient power supply to the diodes. There was the hardware solution using MOSFETS, or there was the cheaper programmable solution using Arduinos. Both of these options were too complex for me, so I found an electrical engineer that I could outsource that task to.

As for the LED circuit boards themselves, I used an etching process that started with a printed mask, spray-paint and a laser cutter. The SMDs were hand-soldered to the board. This process was based on what you might find in industrial scale electronics, but was problematic due to imperfect hand-soldering of tiny surface mounted diodes and the effect of heat on the spray paint.

This is when I came to the conclusion that ordering manufactured LED boards was necessary, and hand-soldering would sacrifice reliability where it saved in cost.



In the end I had a successful working prototype. I was still under my budget, and had calculated that the final product would be much cheaper than initially anticipated.

Unfortunately, there were some changes in priority, and I returned to taking photography with the more expensive light box we purchased and refining workflow.