Building great prototype and even greater product- Journey of SHOOL
Folks involved in hardware designing always say “Hardware is hard” but the truth is right foresight will massively improve the journey of a product.
The cycle is very similar to raising a human baby who is very curious, needs a lot of learning, imitates peers and on the journey develops a personality of its own.
Moving from prototype to product is nothing short of magic that involves combining circuitry, housing, software, and materials. Seeing your product idea into fruition can be satisfying, but it takes many steps to get there like we almost used to give up at each stage as hardware is never a small team job, disaster impending every few minutes.
Five Stars Steps of product designing is explained here that NEERX followed to develop our in-house soil sensor: “SHOOL” based on a modular design approach
For our product to shine, what we need is a 'great idea.' This 'great idea' needs to have two things going for it. First, it needs to have a fantastic idea, and secondly, it requires a 'need to be fulfilled' in the market. Most of us have an excellent idea in our minds. Still, the trickier part is finding the real problem to be solved and how our concept fits that 'need.' Having original ideas is incredible. Still, we need to take a step back and understand that we can't always have a new original idea for our product. Most of the time, we can create a dent in the market with not-so-original ideas, improve in the other areas, and provide more value than competitors. Also, we need to note that competition is always great as it proves that there is a market for your product idea. The NEERX team got the opportunity to understand the problem with the likes of ISRO and Ministries in India. We dug deeper and found this Industry in India has been mostly reselling products imported from US, UK and Australia, which translated to high cost of procurement and poor service availability. Such products were limited to research and academia and never saw the light for commercial B2C markets.
2. Discovery and Feasibility
First, as a team we needed to create a detailed assessment of the new product concept's technical, market, and business aspects and determine its core functionality. Questions that we'll want rough answers to include:
What will our product do? Which form will it take?
How many people might want the product? How much might they pay for it?
How much effort will be required to develop it?
How much will it cost to manufacture and ship?
What are some excellent ways to market and sell it?
Will we need external funding for this? What are the potential sources for any budget we need? What will they be looking for in return?
After answering the questionnaire, we outlined the product priorities clearly and listed the product's functionality followed by extended features. You might feel tempted to have a great design even before a minimal functionality, but this is a trap and should be avoided. A quick hack to fasten the feasibility study is to achieve the Minimum Viable Product (MVP). SHOOL in the initial phase was just about sending variable frequency pulses and analyzing them back, pattern recognition in between complex data streams to understand interaction between the transceiver(SHOOL) and the medium(Soil).
Pro Tip - The more you brainstorm, and the more you talk to actual potential users alongside the MVP, the less you suffer in the later stages
3. Detailed Design and Prototype
As a good engineering practice, start with “PoC(Proof of Concept)” - a low-cost alternative to the product idea that can prove the fundamental concept behind that product. However, one should note to differentiate PoC and MVP (minimum viable product) as they are often confused in hardware development. PoC functionality is limited and is NOT identical to the final product. At the same time, hardware MVP is a prototype that can be presented/sold to real customers to gather valuable product feedback. SHOOL PoC circuitry for instance sat in a “watch box” with shabby wires hanging out and an external battery pack being handled by one team member all time. It was daunting to imagine it as a physical product but our focus at that time was to complete field testing and validate the core algorithms. It is intrinsically iterative and on average, 3-50 units are made depending on the complexity and the BOM (bill of materials) cost. Iteration should be forced to think out of the box but be inclusive in design.
The EVT (engineering validation testing) should then succeed the first phase POC prototype. EVT is all about developing work-like and (sometimes) work-like + look-alike prototypes to validate, test, and refine the product's core functionality. At this stage you might want to look at existing Quality Standards such as ISO, IPC, RoHs, CE, etc. which helps in qualifying products for export and even domestic trade. Unfortunately the ISO standard didn’t exist for our category but with immense support from Space Applications Centre, ISRO certain qualifying tests were designed to simulate extreme weather conditions prevalent on earth and corroborate the lifecycle of the product.
In this stage SHOOL spent most of the time sitting in Satellite Payload testing divisions bearing extreme stress in crest-trough conditioning.
A mechanical jig and an electronic-circuit jig was used to characterize the accuracy and repeatability pre and post tests. After completing EVT prototyping, we locked on to deliver the prototypes and enclosures that look like the final product. To obscure design problems a DVT (design validation testing) may be undertaken, SHOOL’s first 3D print was a horrible design where the functional space was only 30%, we then sought help from a thermocouple factory to bring coherence between functional space and design. Eventually we were able to converge to a 3D printed enclosure achieving almost 80% functional space. SHOOL DVT was undertaken once again for field tests as pre - EVT designs could not be used to benchmark performance.
The PVT or Production Validation Testing is the last step before officially commencing mass-production. Usually, 5-10% of the production run is delivered in the PVT, aiming to stabilize the quality of the manufacturable product. Although the PVT is not the most expensive stage, the outcomes may have a crucial impact on quality and volume production cost. Only minor changes are allowed at the PVT. Any significant change in design kicks the project back to DVT. Prototypes released at this phase are also called "Betas," and samples acquired from the mass-producer are referred to as "goldens samples." The SHOOL golden sample is preserved to date (as are multiple prototypes :) ), it may help at a certain stage to restore erroneous changes. The final product was produced in a limited quantity by using the tools for mass production. Electronic layouts and components were revisited using PCB stencils for soldering components. Mechanical DFM is finalized, and the outer case was 3D printed. This stage can be utilized to do customer testing, important to identify their response.
When we introduced an elegant and beautiful looking design, most of the people were awestruck as we took it a notch up by having a superior quality product packaging(something that was ready to lift off to Mars).
5. Mass production
When the stage is set for manufacturing, be ready to prevent problems that may have not existed in the dev/test cycle. Manufacturing Preparation is a frequently overlooked area specially in early-stage startups, yet it's vital to the smooth rollout of a new product. To avoid complications, ensure that you have prepared the following in advance:
If onboarding a Contract Manufacturer(CM) audit the quality systems to ensure that they have the management systems in place to detect a good or a wrong product before you engage them
Create detailed documentation in terms of manufacturing drawings, assembly drawings, assembly sequences, as these will form the basis of your contract with the CM
Carry out a pre-production approval step that checks first-off components and systems (25 to 50 pieces) against the design and quality plan to ensure that they meet the design intent.
“Finally, some advice for those hardware startups considering a Contract Manufacturer (CM), especially a CM located in a foreign country:
Please get the help of an expert who has both experiences in manufacturing and can perform a quality audit of the facilities, intimately familiar with the culture, and can help you navigate it.
Look for a CM that is the most appropriate for your project and can benefit from your business. As a startup, you will most likely want to avoid trying to entice large CM's to manufacture your product - even if they take you on as a customer. You will probably be a low priority because of your low volume of business.
CM will always have a price control which needs clear clauses in contract as it will help in fund-raise due diligence.
Now if you have survived these stages you are left with "simple things": sales, customer support, and service dealing with product returns and defects. And finally, we have to consider the end of the life cycle for the product, finalize the waste disposal procedures, and start thinking of a new version.