The path from the concept to the "mature" prototype device

The publication will be of interest to people working in “hardware” startups, and especially if the founders have a “software” past.
Most people who have worked a significant part of their career in the “software” industry and want to create a “hardware” product may get the impression that the path from idea to prototype is the same as when developing a mobile application or website: designed, painted design, implemented prototype. In principle, these phases are preserved in the development of devices, but the phase of the implementation of the prototype can bring you a lot of surprises and have several iterations. What happened when creating our first product, LaMetric.


The path from concept to a full-featured prototype and a small pilot batch took eight long months, which allowed to improve the initial idea, understand what the final result should be, prepare the product for mass production and before leaving the crowdfunding campaign be confident in the implementation of a quality product to reduce risks in a tight schedule production.

I will focus only on the implementation phase of the prototype, since it took the most time from us.
')

Cardboard layout

After the product idea was embodied in the 3D model, we needed a real prototype to understand how the size of the device should be. Not plunging in the early stages into the intricacies of the various manufacturing techniques for cases, we went an easier and more tested way, having made several versions of the layout manually of cardboard. This method is inexpensive and is implemented very quickly. Also on dummies you can draw with felt-tip pens in the expected places of future connectors or touch buttons. The result is below in the photo.


Cardboard models of different sizes in different environments

The cardboard layout helped to present the device in the right environment, to assess readability from different distances. Since the product will be used in different places (desk, nightstand, wall shelves, shop windows), it was very important for us to understand how the dimensions of the device (displayed text and icons in all these places) will be perceived from different distances and choose the optimal one. Holding the layout in your hands, you will come up with a lot of new ideas and options for using the device. Do not regret the cardboard to try different form factors of the device.

The next step is the implementation of a working prototype “on the knees” with basic functionality.

Working prototype “on the knees”

Having designed the device architecture, it is necessary to create a working prototype in a short time (two weeks, up to a month). Fortunately, there are a huge number of various debugging tools, Arduino, Raspberry Pi, Virt2real and others, which you can easily buy on the Internet. If you talk about the "software", it is better to use the existing technologies, libraries or practices that you have. At this stage, I will not focus, as each case is unique and individual.


Knee Prototype

To turn a set of components into a device, we placed them in a self-made cardboard box, pasted over with a self-adhesive black film.


Prototype “on the knees” turned into a device

This result helped us quickly check the functionality of the device and make some changes in the architecture. Having such a prototype, it is possible to foresee in more detail for yourself the difficulties that will be encountered in the course of further development.

The result was still far from the 3D model and was only suitable for confirming the idea. For real users, his appearance, to put it mildly, did not cause emotions. In the process of creating a concept, we were inspired by Nike + Fuelband , we wanted to achieve the same effect from the display and perception of the product. It was important to see if we could achieve this.
This thought motivated us to improve the prototype.

“Mature” prototype

In most cases, ready-made solutions for rapid prototyping are not suitable. The reasons for this may be many. The first and simplest is that the debug boards do not fit into the case.

Therefore, the following development stages cannot be avoided:

• If it is a part of electronics, then it is necessary to design schemes, boards, prepare documentation for the production of single copies, solder them, make adjustment and start-up. The difficulties that we had to meet, and how they were overcome, we described in the first article on Habré;
• If this is a case, then you need to create models of each part so that they meet all the requirements and capabilities of the production technology. In our case, the device consists of several unique details. I had to try out different technologies and choose more suitable for each individual part. We also shared the experience of choosing the production technology of the buildings in the last article;
• As for “software”, it will be necessary to rewrite and optimize the code to fit your needs. Also try to use the libraries and practices that come with the debugging tools to the maximum;

Depending on the complexity of the product, you can spend a lot of time on the above items, but in return you get a lot of different experiences and nerves satisfaction from the work done.

In our case, most of the time was spent on the screen in order for it to look and function as intended in the concept. This is the clarity of each pixel, high-quality color reproduction and the effect of merging the screen with the body in the off state. At the expense of the hull shape, it was important for us to comply with all dimensions, radii and bends.


4 iterations to the “mature” prototype

The photo shows several iterations performed by us in the course of manufacturing the “mature” prototype, for which the most time was spent:

Iteration 1. Blurred screen
The first surprise we met was the blurring of the image, the “blurry effect”. Initially, we had the assumption that to achieve clear square pixels without additional tricks will not work, the LEDs will illuminate each other anyway. The solution was obvious and simple, to make a square lattice “sieve” that would accumulate the light clearly in the square area.

Iteration 2. Poor case quality, bad white color.
We still did not like two things:
- quality of the casing manufactured using 3D printing technology FDM method;
- not perfectly white color of pixels;

For the manufacture of the body was applied 3D printing technology, which is based on the SLS method. She pleasantly surprised us and coped with the task by 100%. At the expense of white color, we tried to solve this problem using “software” methods (combining different proportions of three colors), as well as constructive methods, trying to choose a front screen with a different shade or to make a grid with white material. The enumerated attempts could bring the result closer to the necessary, but we wanted the ideal one.

Iteration 3. Good quality case, bad white color
Continuing the search for a solution to the problem associated with poor white color, a solution was found and hid in the name of the problem itself. It was decided to use a white LED and redesign the boards so that you can place two LEDs next to a white + RGB. The effect is what you need and no worse than the Nike.

Iteration 4. Good quality case, perfect white color
Having achieved the desired result, we produced a small pilot batch for testing the product on real users. Their reaction was quite different than on the initial prototype. Appeared product perception, not a research sample. Everyone liked the clarity of pixels, the readability of text and icons, the softness of the screen for the eyes, the convenient shape of the device, the quality and usefulness of the product. Many immediately wanted to buy it for home or office.

Good feedback and feedback brought us a more extensive vision of how you can still use the product, which prompted us to make a virtual instrument simulator to be able to generate even more application options for everyone who could not see and test samples of the pilot batch.


"Screenshot" simulator device

Separate remark. At this stage, you will begin to operate the phrase as “mass production”, so you need to delve into such issues as, preparation of a project for production, optimization for production capacity and capabilities, etc. On Habré there are quite good articles dedicated to this topic, post1, post2, post3.

Conclusion

The “mature” prototype gave us an understanding of the fact that you can achieve the desired look and effect from the device, and also added confidence before launching the “kickstarter” campaign that we plan to carry out in the near future. We will definitely share with you the results and lessons learned.