Development Pipeline Case Study: Helmet Torch Bracket

In our recent feedback survey, lots of people were asking for insights into how our products become a reality. We post a lot about our development on our Facebook page and the new and innovative products we develop keep the company moving forward. So how do products come to the market?

The idea or the product need:

First of all, we need to identify a need. Sometimes a customer will have done that for us and come to us with a problem "I need something like this" and ask us to solve it. Other times, ideas will arise as we're playing or watching others play. Things will be difficult or awkward or simply botched together. We then look for solutions to that problem. The development isn't always a group thing, sometimes our people will just forge ahead with an idea and only discuss it later with refinements. It allows for individual creativity for the good of the company.

In the case of our example, a clip-on helmet torch bracket, the need came from needing light in a camp area that may be demilitarised at a LARP event. Given that LARP usually runs 24 hour (or a close approximation there of) night lights are important. In this case, a head lamp or similar would be fine but given a number of systems are mil-sim that may not always be possible. Keeping the hands free is critical as it allows other things to be done. All of these criteria provide the backbone of our Design Specification that informs further development.

Drawings and Models:

Most of our stuff  gets sketched out on the back of envelopes or in paper development notebooks. There are half a dozen of these scattered around the office covered in the spidery scrawls of our staff. The example in question actually skipped this stage so there's not a lot to show.

Next come the patterns and 3D models. I'll deal with fabric products separately in another post. For 3D printed parts, we use a variety of CAD programs to develop our software but the final stage is always OpenSCAD. If we don't do all of it in that package, it's certainly useful for unifying disparate STL files from various programs and cleaning them up. It produces excellent quality STLs and our GCode interpreter for the printer absolutely loves it so there's no complaint from this quarter!

First prototypes aren't the prettiest things!

First prototypes:

We export through to GCodes for our development printer (a tricolour RepRap Prusa Mendel) to interpret. The printer will then dutifully out put a file according to our 3D model. Essentially, it's magic. Either way, you get a first prototype out at the end of it. Often they're over bulky, have a number of flaws and aren't fit for purpose. 

Testing is sometimes simple, often not

Assess and review:

We then take the prototype and match it to our design specification. Does it do what it was supposed to do? If so, how well? If not, why not? Does it need to be changed? As you can see from the photo the first prototype of the helmet bracket was a little over-engineered and it needed slimming down.

This stage often includes a degree of real world testing. For a lot of products that involves taking them to BUZAN (our local indoor HvZ group) and throwing them at experienced players for them to break. Products that are 90% complete get this testing and everything eventually goes through it.

Implement changes:

Version 0.2 - ready for testing

We slimmed down the piece and figured out a way of printing it in one piece. In doing so we managed to cut the print time by 60%. Further improvements were made to the GCode to tell the printer to print more densely for increased durability and product lifespan. We also noticed that the clip didn't quite settle on the helmet correctly and, over time, would fatigue and fail. We made a few alterations to the model and threw it at the development printer. We also removed one of the torch holders for the production version as normal people don't need two torches. We got a second prototype and it worked - it passed testing, there were no bugs and it was ready to proceed.

The cycle of test, review, implement, test can be repeated dozens of times for a product. The stock extenders and the MkIII holster are examples of products that took (or are taking) many, many cycles to get right. By contrast, the torch bracket in question only require a 2 step process. It very much varies on the product in question.

Final deployment:

After successful combat testing and some regulatory paperwork, the product is handed to the management team who then price and market the product through the website etc. The joy of our business model is that we can produce something once, pop it on the website for sale forever but not need to keep stock floating around. Provided we have the patterns and the people to make it, we can dig out products that haven't been made for months and months and put them together as though the gap didn't exist.

Final product ready for sale!

Keep an eye out on our website for new products as they develop. The LARP season is starting up again so expect to see lots of development as we move forward. We'll be playing and watching players in an environment very different from our home system so it should give us plenty of new ideas and product needs.