In the past ten days, three clients have presented their new designs (an employee time and attendance clock, a remote activated consumer power strip, and a family of industrial and residential thermostats) in order for EastBridge Engineering to pre-qualify vendors and generate quotations.
Unfortunately, what we received wasn't a useful 'turnover package' or collection of CAD files, assembly drawings, BOM (Bill of Materials) or a well written specification. The documents weren't even a clearly stated set of product attributes of deliverables.
We needed to back-up and determine where our clients really wanted to go... Did they want to provide unique and specialised design data and have the finished units fabricated by contract manufacturer? Did they truly want a (mostly) off-the-shelf product designed by fully integrated manufacturer? Or were they searching for a OEM/ODM hybrid approach, whereby a new product, which incorporated some unique design features, would be teased out of a existing model?
Our discussion pivoted into the differences between an OEM, ODM and OEM/ODM business models. In the automotive industry, 'OEM' (Original Equipment Manufacturer) means a company making 'drop-in-replacements' for the aftermarket. In the electronic industry, 'OEM' refers to one company incorporating a finished component or sub assembly, purchased from an unrelated supplier, into their own finished product. In other segments, this approach is known as 'private labeling.'
An ODM (Original Design Manufacturer) project in the electronics segment involves a supplier building a finished device based on an original design and using the clients' design documentation. In the motion control and industrial segments, it can mean asking the manufacturer to design and build a new product from scratch.
And an OEM/ODM project can include some of the elements listed above and a thousand variations that we can't think of.
These acronyms (and the underlying concepts and definitions) have become so contradictory and confusing that at this point in time, they've become useless. So how do you navigate this potential mess?
As a first step, drafting a clearly written PRD (Product Requirements Document) is the best approach. A PRD can be a single page of bullet points listing the key attributes of the finished product. It can also be a highly detailed and complex document that incorporates market research, competitor profiles, product tear downs and voice of the customer input.
The discussion and effort to produce the PRD often is sufficient to define which branch of the quoting and production path to take. If the new product is closely related to others in the marketplace and only requires minor finishing tweaks, private label is likely the way to go. If you've invented a new product that largely shares the form factor and technologies of existing products, the blended approach could be useful. If it's a breakthrough concept that's going to change the market, the turnkey design approach is merited.
In all of these models, the PRD guides the initial discussion and will help to determine how the product will be built and where the project is placed.
So, before you put much effort into creating your 'design', please write up a clear PRD.
The volume curve
If you are looking for production plastic parts, then injection moulding is a very cost effective solution for quantities down to and even below 100.
However, if your quantities are uncertain or the design not frozen, then CNC machining (computer numerical control) could offer a very economic alternative and comes with the added advantage that quality of finish and accuracy are far superior to alternative prototyping options.
Rapid Prototyping (RP) gets a lot of press for its capabilities as an Additive Manufacture (AM) process, and the 'freedom' it affords to design and manufacture anything. However, when it comes to production the reality is less about the geometry and more about the issues of poor surface quality and finish. If your part needs to be smooth, then currently the only way of achieving this is by vibro-finishing, but this is usually to the detriment of all sharp edges and not just those layers.
CNC offers a good alternative, but the set up and cost per part is usually deemed prohibitively high. The good news is, there is a way round this so that you can benefit from the wide range of materials, colours and the quality of finish that CNC delivers. How? By processing a 'sheet' of parts at a time. Depending on part size and geometry, over 100 parts can be cut from a single sheet - and very quickly.
If we then allow a level of fabrication, even a 'box' can be made by machining, folding and bonding. Using this process as opposed to creating a box from machining a solid block saves both time and money. It's almost like sheet metal fabrication but in plastic! Process time and waste is significantly reduced; structurally the 'box' can be very strong and aesthetics can be maintained up to and including some texture on the outside.
What's more, it offers great design flexibility – with a little lateral thinking even rounded corners can be achieved. Mounting features can then be assembled either as more machined parts, or if they're internal, created as RP parts that dowel into the cosmetic housing.
Design freedom may not be as great as with injection moulding, but on the plus side you don't have to wait weeks to get the first part or have an upfront cost to contend with. Components can be drip fed from early in the process and design changes can be easily incorporated without the need to alter any tooling.
The same principle of manufacturing a 'sheet' of materials can be applied to millturn parts when the technology is used with a bar feeder. Material up to 60mm can be readily used with millturn machines and once programmed the process will run unattended.
We still tend to associate turning with round parts, however modern technology allows for multiple tools such that turning, milling and drilling to be carried out in a single operation. Features can be created off axis, drilled and threaded such that the part is fully finished by the time it's parted off.
All of a sudden we have highly complex parts produced in a single operation with dramatically less labour involved…and yes this applies to metals as well.
With the finish and accuracy of CNC being brought to batch production of plastic parts and pushing quantities higher than RP would economically allow, CNC is a viable contender when quantities are too low or uncertain for moulding.
Post a Comment