The path to bring a new hardware product to market is filled with a lot of challenges that large corporates are usually able to march past using qualities like experience, relationship, money, and people's expertise, which are usually non-existent in the arsenal of startups and individuals. More often than once, we've seen these challenges bury several good ideas/products in the valley of development, and for those who manage to get them over the production finish line, they are usually behind schedule and at an amount the team would wish they could have avoided.
To help individuals, startups, etc. get their new hardware product to market, on-time and with minimal hassles, today's article will share some tips, tricks, and tools to help you navigate the murky waters of hardware product development. The article will take the difference in products (wearables, IoT, Consumer Electronics, etc.) into consideration and present a generalized approach that can be adopted and adapted for any kind of electronics-based product.
Ready? Let's get it.
There are three main parts of Hardware Product Development where I feel startups and individuals looking to bring new hardware products to market can get things wrong. They include:
1. User Research
2. Product Engineering, Design, and Prototyping
3. Manufacturing (Design for Manufacturing, Parts Sourcing, Manufacturing Partner Decisions)
1. User Research
While the importance of user research for the success of a product, business-wise, is general knowledge, it's importance in the design and development process of physical products is often trivialized and overlooked. This is one of the first mistakes usually made by teams/individuals looking to bring a new hardware product to the market for the first time.
Overlooking how users interact with products, components - use environment fit, etc., often lead to dozens of avoidable design/prototype revisions which sometimes cuts deep into project timeline and funds.
The excuse most startups/individuals give is the need to go fast and test quickly. While this is definitely advisable, it calls for caution as hardware is not quite like software, and the agile development process for hardware is a bit more complex as the nature of the hardware, arguably, aligns more with the Waterfall approach. You must take into consideration the fact that your "users" at these points include regulators, sales channels, etc., not just the end-users. So even if you do not have the "killer" features implemented in the first version of the product, it must come with the basic features/properties that makes it usable and acceptable to the "users".
One important approach to ensure this is the use of a technical product requirement document (PRD). Developed during or after the user research, the document usually combines elements from the user research, product goals, and biases, with technical information, to describe how the product is expected to look and work with other seemingly insignificant but important information like how and where it will be sold.
Its goal is to serve as a reference document to ensure designers have a full understanding of what the product entails, and can make strategic and informed decisions during the design process. Lots of templates exist on the internet but I will recommend this cool template by Bolts.
2. Product Engineering Design and Prototyping
More time is wasted and more challenges are experienced in these stages than any other stage of the product development process.
There are different challenges associated with these stages so I will rather share approaches that could help you navigate the stage successfully rather than focus on specific challenges.
1. Product Pre-Design
My recommendation for the first step after figuring out what "users" desire in a product is to run a Pre-Design of the product. The Pre-design process involves the creation of block diagrams and flow charts, and identification of the critical parts of the project like MCU's, actuators, sensors, etc.. At this stage, it is smart to avoid selecting a specific component model or manufacturers and just focus on engineering, setting out selection criteria (e.g cost, Low-Power, IOs, etc) for key components.
I wrote an article about selecting the right microcontroller for your project, a while back. it might provide some insight on what specifications should look like, at least for microcontrollers.
2. Selecting Parts/ Developing BOM
Armed with the specifications/selection criteria developed, it is time to go part's hunting. The goal for this stage is to identify which specific components fit the design parameters you laid out earlier. It is absolutely normal and good that you/your team starts with components that you are familiar with, but this may introduce biases that could be counterproductive. Commit time to research and be willing to broaden your horizons. One of the major contributions to endless prototype design revisions is the lack of proper parts/components research.
To help with this, component manufacturers usually maintain different applications on their website that recommend components; microcontrollers, ICs, sensors, etc., based on specifications provided by users. Use them.
3. Design and Prototyping
With potential components figured out, the next step is designed, putting it all together to see how it works. While CAD Softwares like Proteus offers simulation features that may be useful, the good-old breadboard prototypes and development platforms might be the best way to first test out design assumptions in a hacky-wacky way. It helps remove simulation related doubts and ensure you have something physical to test with. After the breadboard prototypes all checks out, the next step is a gradual march towards your engineering prototype by developing a more refined schematic and a Printed Circuit Board using a CAD/CAM tool.
Design, as expected, takes a lot of time at this stage, but one approach that might help you reduce the time it takes is the use of Reference designs. Component manufacturers provide reference designs including schematics, source code (where applicable), and sample application/use case, etc., to help reduce the time spent on the design stage. Use them.
Design reviews at this stage are a common element but the number of reviews can, sometimes, be reduced with proper planning and foresight. One act of foresight that could help you reduce the number of design reviews is to select a PCB manufacturer and get information about their specification and capabilities before you begin designing the PCB. Capabilities and PCB specifications, from no of layers to min/max specifications for board features like via size, board thickness, etc., differ from one manufacturer to the other. As such, selecting a manufacturer early on and taking into account their requirements and capabilities during the design process ensures your PCB is designed to their capabilities and very little redesigns/adjustments are required.
Selecting the right PCB manufacturer is very important at this stage as it might take at least 2 production runs before you are ready for the engineering prototype. My usual criteria for selecting a PCB manufacturer include:
1. Equipped to manufacture the kind of PCB you desire (For example, not all manufacturers are equipped to produce flexible PCBs)
2. Fast turn around time
3. Support
4. Cost
There are a lot of PCB manufacturers out there but one manufacturer that has consistently checked all the boxes for me is PCBway. Asides their super competitive pricing (as lows a 5$ for 10 PCBs) and fast turn around time (between 24-72hours depending on your location), they have one of the most equipped factories among PCB manufacturers and are as a result able to produce all kinds of complex and advance PCBs. I'm a big fan of their support team which conducts an engineering review of every design you send them, to check for issues that could affect the performance of your PCB after manufacturing. This helps prevent endless PCB manufacturing re-runs, saving money and time.
4. Engineering Prototype and DFM
As you iterate through the prototyping process, it quickly gets to a point where you are ready to make a version of the device that works and/or looks like the final product will when manufactured in volume. This is the stage where you start miniaturizing your components and replacing modules with ICs.
One major challenge startups/individuals face at this stage is usually related to parts scalability. In haste to get the project together during prototyping, Startups/individuals usually create prototypes without evaluating the parts to see if a version of it that is optimized (in terms of cost, large scale availability, form factor, etc.,) for manufacturing, exists. This becomes a problem at this stage and they are often forced to change components A or make changes to component B in a way that could lead to delays in the project timeline and cost more money. Some smart approach to ensure you don't fall into this and other DFM related traps include:
1. Ensure the components whose evaluation kits and breakout boards are being used for prototyping have existed in a manufacturing optimized form that suits your project.
2. Components with a large ecosystem of potential replacements
3. Talk to parts suppliers and potential manufacturing partners to find out the availability of the major components.
4. Technology is moving fast and parts/components can be obsolete within a year. Ensure the major components of your project will still be around and relevant before you are ready for another version of the product.
A PCB manufacturing partner is also a nice to have for this stage as you may need to make a few versions of the engineering prototype before its fully ready. Also due to design miniaturization, most of the components for your project, at this stage, will be SMD types. While you could mount them using your personal reflow oven, the need for improved quality at this stage might mean you work with a PCBA (PCB Assembly) partner. Most PCB manufacturers like PCBway offer this as part of their service and its better to use them. It saves time and reduces logistics-related headaches when the PCB manufacturer also handles the assembly.
3. Manufacturing
Getting your project to the point where you have a final prototype and are thinking of manufacturing is no small feat but the harsh reality is, depending on the kind of future you want for the product, there may be more challenges ahead than you can imagine (sorry).
There are generally two approaches to products at this stage. One is to produce them in small volumes (10s and 100s) while the other is on a larger scale of 1000 and 10,000.
For small volume manufacturing, the approach is usually as simple as finding a PCB/PCBA manufacturer who is willing to take on your products as the volume may not be interesting/profitable enough for a contract manufacturer. One important thing to look for the manufacturer at this stage is parts sourcing relationships and their commitment to your product. One manufacturer I’ve worked for this stage and approach is PCBway. Like I said earlier, they tick all the boxes.
For large volume manufacturing, the process may be a bit more complex and requirements differ from one manufacturer to the other. Irrespective of the manufacturer and approach, two things that you must embrace at this stage is proper planning and knowledge. I will break them down into three points;
1. Start Early
Startups/individuals often wait until designing the product before they start talking to manufacturers and suppliers. Even though you are not ready you should immediately start engaging potential manufacturers. Ask questions about their capabilities and how best products like yours can be manufactured (all with an NDA of course). Their experience and advice will be useful in shaping your final prototype into a form that could be manufactured with fewer reviews.
2. Build Relationships
From parts sourcing contacts to manufacturing partners, the relationship is one of the most valuable resources at this stage. Endeavor to build them.
3. Enlist Experience
Not sure I can really express in words how tough manufacturing is, but a lot of the risks involved can be mitigated with experience. This stage will definitely benefit from having someone on the team who has significant manufacturing/sourcing experience. If you can't afford them on your team full time, get them to work as consultants with clearly stated deliverables. Another smart thing to do is to talk to as many people (other founders, engineers, etc.) as possible in that space. By leveraging on their experience, you will most likely save yourself thousands of dollars in time and mistakes.
In conclusion, while it’s almost impossible to totally avoid challenges that could affect your product timeline, a number of steps could be taken to reduce the risks. The suggestions provided in this article are not final and could vary from product to product but with proper planning, most risks could be mitigated.
Web: https://www.electronics-lab.com/
Twitter: @ emmaodunlade
Note: The content and the pictures in this article are contributed by the author. The opinions expressed by contributors are their own and not those of PCBWay. If there is any infringement of content or pictures, please contact our editor (eric@pcbway.com) for deleting.
Written by
