Is your product design a piece of art that will wow the whole the world? Great, now redesign it with DFMA (Design for Manufacturing and Assembly) in mind.
What is DFMA?
In short – designing a product that is easy and feasible to manufacture and assemble.
This part of the development process is probably the most complex and overlooked side of product design. Many designers are great “shape-givers”, essentially great designers. But what separates the really good industrial designers from the lesser ones is their knowhow about manufacturing and the assembly processes. A great in depth knowledge about injection tooling, injection processes, metal processes, CNC processes, materials, finishes, part assembly, safety – down to packaging and shipping containers.
What is the goal with DFMA?
The overall goal is to produce a high quality product, efficiently in terms of cost and time, with as low as possible failure rates on the production lines – EACH production line. Not only the final assembly but each manufactured part of the product.
When spending money on product development you want to ensure that you can run a cost-down approach while scaling. If your product is designed wrong, with multiple problems and high failure rates on parts and assembly, you are more likely to get a cost-up on the product, due to manufacturers having to produce x% more parts when fulfilling your orders.
There are several steps to consider in order to get a product just right for manufacturing and assembly.
Lets take a look at DFA (Design for Assembly)
- Design your product with the assemble method in mind.
- Reduce part count & types
- Ensure parts cannot be installed incorrectly
- Strive to eliminate adjustments
- Ensure parts self-align & self-locate
- Ensure adequate access & unrestricted vision
- Ensure parts are easily handled from bulk
- Minimize reorientation & secondary operations during assembly
- Parts should be symmetrical or obviously asymmetrical
Let take a look at the DFM (Design for Manufacturing)
- Use existing manufacturing practises. Trying to move the barrier on standard processes can get expensive in setup for machinery and jigs.
- Use standard components and parts.
- Limit secondary treatments on parts where possible.
- Reduce the number of handling for each part during manufacturing.
- Design in order to reduce tolerance restrictions.
- If you are planning to scale quick, make sure the manufacturing processes are scalable.
So all of the above sounds easy enough, but in practice it is not. Even after so many years in the development and manufacturing business our team still spends many overtime hours, to find a suitable fit, that honors the overall design and keeps production costs low.
Here are some of the thoughts and considerations to make before going into tooling and pilot run mode:
From 24 to 2 parts.
- What materials will i want to produce my product in?
- How can my parts be assembled relatively easy?
- Do i need external fixtures?
- Can i combine several parts into one part, making the manufacturing of the part a bit more complex, but save costs on assembly and fixtures/screws?
- Do i have more products in my pipeline where parts can be shared? (Modular design)
- Is there any way to improve cycle times on my parts = a quicker output?
- What compromises am i willing to take on the looks of my product to achieve a higher output and lower cost?
- What surface treatment is needed to achieve the look i am looking for?
- Do i even need surface treatments?
- Can any surface treatments improve handling and reduce rejects?
In praxis for plastic injection it might be cheaper to invest more on your tools, molds and jig, use collapsible cores and perhaps look at 2 component injection or overmolding than trying to fit or fixate a secondary part onto an existing part. This will most likely reduce rejections and improve the overall tolerances on your assembly line.
Electrical components and manufacturing
If you like many others have designed your electronic boards and solutions based on European, US or Japanese components, the chance that the cost for producing such PCBs in China is be more expensive than in Europe will be very high.
It is a good idea to consult a partner in China to source for alternative components with similar specs and performance after having tested and made your proof of concept on normal development boards.
The cost differences can be quite significant if your are just starting out, and don’t have the quantities with you yet.
Sticking to certain branded MCUs / CPUs / IC etc. might still be essential in highly complex products or due to certain skillsets from your engineers. Just make sure the supply chain is available in your chosen country of manufacturing.
If developing all electronics outside of china, you might risk having to reprogram and rewrite your software in order to get a certain cost down solution on your electronic parts.
To improve DFM of your actual PCBA there are many steps involved to make sure you have a flawless board design that is ready for manufacturing.
Electronic engineers usually run a DRC (Design Rules Check), which in itself is good, but not sufficient enough to ensure manufacturing is possible. The actual CAD programs will pick up on certain flaws in a design but not all. Always make sure to let a professional company run a DFM check on your board design.
Usually points to be checked include:
- Drill checks
- Signal and Mixed Layer Checks
- Power/Ground Checks
- Solder Mask Checks
- Silkscreen Checks
All of the above include but are not limited to check of all sizes, drill holes, all soldering mask bridges, all connection layers, all thermal points, all spacings, all routings, all copper vs. soldering masks, clearances, stubs (unfinished connections, layers) etc.
Confirm prototypes vs. files:
After making prototypes of your boards make sure you confirm these vs. the manufacturing files. Some manufacturers adjust received files, based on their capabilities or due to flaws found on the delivered files. It is therefore important to get the latest manufacturing files back from the manufacturer to make sure you always have the most recent files to compare and revise. If the manufacturer has changed something and not informed you, you might have working samples, but if you are thereafter make changes, you are making changes on non-confirmed / non-tested files possible flawed files. This could lead to terrible manufacturing losses if you then go directly to production with a second factory.
Last but not least you need to consider: “How easy should my product be to repair?”
Repair is a natural phenomenon in the electronic product’s world. In a perfect world you would not have to consider this, but nothing is perfect so give it some thought. There could be several parts that sooner or later needs replacement, and you should already when designing the product give this subject some thought. Repairs could include:
- Broken battery
- Cracked or broken screen
- Broken lens
- Broken casing
- Broken speaker
- etc. etc.
Let’s say you choose to ultrasonic weld to plastic parts together or use glue. What are the complications?
- If you during production or the pilot run find a bug/mistake, all of your casings need to be broken to replace the problem part and reproduced.
- During a repair, the casing would need replacement, meaning you need to have such casings in inventory.
Let’s say you use a snap joint without an ejection pin/point, you most likely would need to replace the casing as well. Snap joints improve assembly, but make it difficult to maintain/repair a product.
Some companies however make it a business to repair their own products (not naming names here), this of course is an alternative approach, that can be considered. Making speciality tools to open and close your own products is a possibility if you have the distribution and repair network in place.
There are many manufacturing processes that could be addressed, which nowadays are an integrated part of many electronic products, like wood working, steel, aluminium, fabrics, silicone, rubber, cables, screws, nuts and bolts, LCDs, speakers, logistics and packaging between manufacturing and assembly plants etc, each with their own sets of problems and processes to take into account. Not all can be covered here, but feel free to drop a comment on what should be included in the next article.
Above was just a brief introduction to the DFMA process to consider when designing a new electronic consumer product. In my next articles I will address the following points: CMF (Color Material and Finish) / Locking parts in place / Getting quotations / Getting tooling quotations / The importance of Packaging and Manual making / Traceability / Markings and label / Assembly costs / Pilot run / Cost down solutions / Certifications / Mass productions / Quality Assurance and Quality Control / Logistics / RMA procedures / Long term partnerships / Resource allocations / Specific details when developing connected products / Specific Details when Developing Audio Products