What is DFM, Really?

Estimated reading time: 1 minute

Okay, so what is DFM, really? The term "design for manufacturability" has been used for many years now, but does everyone really understand this concept?

For instance, do you design for 10%? Do you design for a specific manufacturer’s capabilities, therefore making you less likely to seek alternative fabricators? How are your drawings worded?

In this article, I will be discussing the reality of DFM and what benefits you, the end-user, by embracing these practices.

Why Design For Manufacturability at All?

Good question. Even if you only buy your boards from a single source--if you have qualified the company already and feel you can expect certain press parameters and dielectric constants based on what they have provided you--it is STILL a good idea to at least design with some latitude. If your design is .1 mm lines and spaces there is not a whole lot of room to either expand or decrease the traces to achieve certain impedances. Clearly, when you have to ingress and egress out of tight-pitch components and your design takes you down to .003”/.003” there is NO ROOM at all for an etch compensation, so you are typically quoted by manufacturers as quarter-ounce foil start. This foil is so thin that we need not compensate for a loss at the etcher like the other copper weights.

Again, as I have mentioned before in my columns, the general rule of thumb is that for every half-ounce of starting copper, you give all the metal features an etch compensation of half a mil. Asking for 1 oz. starting copper, for instance, with 0.003”/0.003” will normally be a no-bid as fabricators would be hard-pressed to be able to run with .002” spaces at Image prior to etch. (Attempting to compensate the 0.003” traces for 1 oz. copper with 1 mil will result in 0.002” spaces at Image prior to etch.) So, 0.003”/0.003” is usually the limit.

Read the full article here.


Editor's Note: This article originally appeared in the May 2014 issue of The PCB Design Magazine.