When starting or redesigning a new product it has been typical to start by using as many standard parts as possible. This is usually a very good idea as standard, readily available mechanics allow quick proof of concept and a means to getting to the market quickly. The downside to this approach could possibly be per-unit cost being high, high component cost, and high vendor count. This can be due to possible mismatches in technology or sizing that standard product inadvertently pushes your design into.

Let's look at a typical situation. Marketing believes you will be selling 500 units per year and that it will take 12 months to ramp to that level. In the past, based on this information, it was intuitive to start with a commercially available assembly and then move to a lower cost custom version of what you've proofed your concept and started to market with. There still may be validity to this approach, but in today's market there may be a quicker way to market while managing time and other risks available to you.

Some vendors have enough breadth of product and bandwidth of engineering to design and produce a specific part of your product for you. In the case of an actuator with a servo or stepper system, the vendor would need to manufacture stages, motors, and amplifiers, and then have the ability to design and manufacture from startup to very high quantities. Now, if you've found the correct vendor, you can isolate yourself from the risk of designing a new system in a specific time frame. You should be able to decide the depth of the vendor's integration with your design team. Next, work on clearly defining the role of the vendor in R&D, Beta start-ups, production, etc., so there is no mystification of roles. Specific people should be tasked with clearly defined rolls in your and the vendor's company so that communication is clear and concise.

Once the project has been defined you should be able to report up a defined schedule with known costs at different volume levels. Management typically loves the ability to define costs and risks through the process. This is one of the main reasons people go to contract manufacturing, but there is also the ability to work off a clean slate, as well.

Typically, you may have two commercially available motors that make sense, but it would be perfect if you had a motor that at top speed would go exactly X RPMs with exactly X NM of force in an X size frame. Remember, if your motor is rated at 1KW and you are going 72 percent of the motor's rated speed you are simply wasting 28 percent of the motor.

This lack of efficiency can snowball quickly, perhaps from the available motors, and you may need to go up a frame size. This will probably be more money than needed with more weight. If another axis needs to move this motor it has to be bigger, as well. This concept holds true throughout an assembly. For instance, if you've decided that an advantage of having a vendor manufacture an assembly is that the vendor is responsible for the first year of this part of your system's warranty. In this case, the whole assembly would be removed and another swapped in. There really is no need for a coupling since no changes will be made in the field. So now the torsional wind up, the inertia, the weight, and the induced low frequency resonance of a coupling are gone, which further reduces cost and part count of the assembly. A custom system would have the correctly sized motor with the correct actuator, drives, and control interface.

Considerations
This list could prove useful in laying out the case for the new project: