9 tips for prototyping a technical device
Prototyping is the most important stage in the development of a technological product. This path has its own pitfalls, shoals and whirlpools. To avoid them and create a prototype as close as possible to a serial product, you need to have competencies and know the secret fairway.
One of the main success factors for a technology product is its speed to market. Therefore, the faster you go through the design and prototyping stages, the better. Usually this process lasts from six months to a year – it all depends on the degree of elaboration and complexity of the project. The first prototypes can only be a tool for verifying shape, dimensions and use cases. They can be made from scrap materials. Such prototypes are usually called mockups or mockups in foreign terminology. The most important and complex work begins as the detail and depth of the product is developed, when it is necessary to check the technological hypotheses and functionality of the product.
There are many technologies for rapid prototyping – non-serial production, which allow you to make a piece product without large investments. The most common are 3D printing, cutting / bending/welding/machining of metal, cutting and machining of wood, stamping, molding in silicone, thermoforming of plastic, contact molding of fiberglass. Each of these technologies has its own characteristics, which must be taken into account so that the prototype is as similar as possible to a serial device, not only in appearance, but also in its properties. It is possible to cope with this task only with the competence and a staff of experienced specialists. That is why not every industrial design studio is involved in prototyping.
Choosing the right prototyping technology is critical. An error can lead to the fact that the information received at this stage turns out to be irrelevant and goes to the trash. Having made the body of the prototype from wood, it is difficult to hope that its aesthetic and physical properties will be preserved during mass production from plastic. This is especially true of parts and mechanisms directly tied to the functionality of the future product. So you always need to focus on a serial product and its estimated cost and select technologies that will allow you to produce a prototype that is close to the final product in terms of characteristics. So, for example, the same gears and racks, but printed with different technologies from different materials, will behave differently. By the way,
Errors are acceptable
Prototyping technologies often don’t work well together. The production itself can have different precisions and tolerances, as well as different dimensional stability over time. For example, 3D printing problems are shrinkage, roughness, insufficient strength. Sometimes the received parts may not even be assembled into a single product. And then it may seem that mistakes were made at the design stage. But this is not necessarily the case. You should always remember that this is a prototype, not a serial product, and it must perform a specific task, and not meet all the requirements. But it is important to determine all possible assumptions at the initial stage and, at each iteration, check individual characteristics: overall dimensions, internal dimensions, product heating level, functioning of individual elements …
At the very beginning of our journey, we were engaged in a project to create a portable coffee machine for a car that would work from a cigarette lighter. The task did not seem too difficult. But at the stage of development planning and prototyping, a key mistake was made – the customers and developers decided to create a complex product entirely in one go, bypassing the stages of development on mock-ups and prototypes. As a result, the prototype, representing an assembly of many small parts (about 50 parts, not including electronics, fit in a glass), was assembled and refined more difficult than a ship in a glass bottle and constantly turned out to be leaky, which was perceived as a design error.
Working on a similar project now, specialists would confidently decompose it into technological hypotheses and test them in order of importance on prototypes for the main stages of the product’s operation. By the way, in food projects where biocompatible plastic is required that works at high temperatures, as well as where high requirements for accuracy are imposed, printers operating on SLA technology save us.
Innovation requires more iteration
Typically, a prototype goes through three to four iterations until the desired result is achieved. And the more innovative a product is developed, the more iterations will be needed. In general, development after the launch of a product on the market does not stop, but flows into continuous work on improving consumer and other properties.
In the last article, we talked about a personal automatic intramuscular injector. Devices with such functionality did not exist on the market, everything had to be invented from scratch. It was necessary to estimate the dimensions and weight, check the reliability and accuracy of the mechanics, and work out the application scenarios. Through successive iterations, it was possible to bring the product to the target cost price, reduce the number of parts by almost half, resolve a number of physical contradictions with the operation of the motor, discover several “chips” in the user experience, and so on.
Prototypist feet fed
The design of the prototype takes place in an ideal environment where the specialist can access any details and materials at the click of a mouse. However, very often it turns out that to create a product, special screws, springs, gaskets or other components are needed that are not available. By the way, right now we see this in terms of delivery times for various electronic components. To find them, the prototype has to run around the shops and markets. You can try to contact the manufacturing plant directly, but due to the small batch size, the delivery can be very delayed. And then the prototypist has to go around the markets where the necessary parts can be lying around.
To avoid such a situation, “Karfidov Lab” maintains a special section of the company’s knowledge base, which contains typical components that manufacturers do not plan to discontinue production, and also has a track record of contacts of experimental production sites and its own pilot production, so as not to ask large factories for making small orders.
Find the main node
It is necessary to start prototyping from the most problematic and complex structural unit. This may require additional resources, but ultimately this approach will optimize the development budget and reduce the possible amount of future rework and rework derived from this master node. Only after testing the main technological hypothesis can we proceed to creating a common prototype.
Now we are designing a very complex, the key element of which is an innovative system of mechanical gripping of objects using machine vision. First, you need to implement this design based on customer requirements and test it in the field before proceeding to detailed design of the entire project.
Hope for software, but don’t make a mistake yourself
Modern software allows you to make a lot of calculations at the stage of technical design. But, unfortunately, they are not always applicable in real life. For example, in a computer model, a pipe has the same wall thickness everywhere. But in practice, the product may be flawed, and in some places the walls will be thinner than stated.
One of our projects was the development of a toothbrush that would work in two modes – rotational and translational (like a hammer drill). The customer was a network of dental clinics. Her leader came to the conclusion that the simultaneous operation of the brush in two modes will have a positive effect on the health of the teeth. However, modern software does not know how to calculate the interaction of inorganic materials and organic surfaces such as leather. And during the testing of the product, the lip of the tester “wrapped” a little on the brush. The result is a “lip-rolling” machine. The problem, of course, was quickly fixed. But this amusing incident illustrates the thesis that, sitting at a computer, everything is impossible to predict.
Prepare for unusual tests
When prototyping, it is almost always necessary to test the product “for strength” in real conditions.
The most unusual test at the company involved the design of an underwater deep-sea machine for inspecting a gas pipeline laid along the ocean floor for leaks. This device was supposed to operate at depths of up to 600 meters. It is impossible to create such pressure under normal conditions. I had to look for a way out of the situation and negotiate with the Institute of Oceanology, which has its own test pool, where it is possible to recreate the required conditions. The first prototype of the device, by the way, burst due to pipes with walls of different thickness. As a result, it was possible to select other components and create a margin of safety so that the product could withstand heavy loads. It was possible to understand the need for such an improvement only after full-scale tests.
Errors are ok
Prototyping is a long process. And with each iteration, problems and errors are bound to arise. They must be collected and analyzed. But don’t be afraid and don’t get hung up on them! Error immunity is the main quality of a good prototype. And we must also remember that sometimes, while working on one problem, it is easy to create a new one.
We encountered unusual difficulties, for example, when developing an automatic shoe dryer. The idea of the invention is that in bad weather people bring snow and mud on the soles of their shoes, which are carried around the room – an apartment, an office, a medical institution. The development was supposed to provide automatic drying of the soles – I stood on a special platform, waited 30 seconds, the snow melted, the dirt crumbled, you can go further. The necessary calculations were carried out, a prototype was assembled, but it turned out that the hot air leaving the compressor lost heat after passing through the pipes. The outlet temperature was only 60 degrees – too low to dry the soles in 30 seconds. Experimentally, we managed to find the right combination, but in the process more than one pair of soles melted.
When the temperature problem was resolved, the developers had a new one. Placed corrugated pipes began to emit a loud sound, very reminiscent of the sound of the “king of musical instruments” – the organ. This clearly did not fit into the understanding of the correct operation of the device. The developers began to experiment with the configuration of the device, the sound of the pipes also changed. Once the device roared like a MiG-27 on takeoff. But in the end, the iterative approach helped bring the project to the desired result: remove unwanted sound, significantly reduce the total weight of the device, simplify the mechanics of work, and even remove unnecessary functionality. And now the device is on sale.
Finished electronics is an acceptable way to go
When making a prototype, it is quite logical to use ready-made kits, for example, Arduino or Raspberry pi. Of course, they sometimes have redundant functionality, but they have a high debugging speed, a large component base and a set of code libraries, which allows you to quickly resolve the issue. However, there are also disadvantages – cost, already defined geometry, impossibility to provide structural elements. Of course, for a serial product it is more logical and more correct to use printed circuit boards of our own design, but for a prototype, ready-made kits are the correct and reasonable choice.
Alexey Karfidov, co-founder and general designer of Karfidov Lab :
“The process of developing a prototype of an innovative technical product is not always a simple and linear process, sometimes requiring significant going beyond the boundaries of one’s own knowledge and competencies. The main thing is to see the end result, understand its nature and not stop at the slightest difficulty. Often prototyping is R&D, but in science a negative result is the same result. This is neither good nor bad. This is just an excuse to reconsider some points in the product, technology, functionality, etc. This is evidenced by our practice, when at the stage of prototyping a product there is a change in its concept. “