Seven DFM Issues Hidden in PCB Design When our product has a large number of Design for Manufacturability (DFM) errors after we have completed the design and sent it to the fab, it can lead to product shelving. This situation is not only frustrating, but also costly.
Considering manufacturing issues early on in a project can help reduce costs, shorten development time, and ensure a smooth transition of designs to production. Conversely, failure to do so will have adverse consequences.
Drawing on years of industry experience, we have identified 7 major DFM issues that hinder PCB manufacturability. While some of the items listed below are design best practices, others are issues raised by builders/fabs. By addressing these issues during the design phase of the project, we will be able to correct any DFM errors that may arise before the product reaches the factory.

So, before sending the design to the manufacturer, we want to be aware of the following DFM issues as they may be hidden in our design.

1. Acute angle

Sharp angles are sharp or odd angles on copper components in a printed circuit board that can cause acid build-up during the PCB creation process. This problem occurs before the washing process, and the sharp angles cause residual acid to get trapped in these areas and not be able to be removed. Eventually, the copper components needed for the Gerber files contained on the board begin to corrode, causing the copper wires to "break" or disappear.

Avoiding sharp corners is especially important for 4 mil or 5 mil traces in today's designs. Because they are thin, they break easily (openings inside the useful copper due to adsorbed acid). Some software has built-in checks for such situations, but if our software didn't have this capability, we would have to manually evaluate any possibilities in the board that could cause this.

How to prevent acute angles: Avoid placing traces into pads at acute or odd angles, keep the angle at 45 or 90 degrees near the pad.

2. Copper bars and islands

Copper bars and islands are free-floating copper on many planar layers, which can cause some serious problems in acid baths. Tiny copper spots are known to float off PCB panels and reach other etched areas on the panel, causing short circuits. Another case is that if they are large enough to not float, it will become an antenna, which can cause noise and other interference within the board (since its copper is not grounded - it will become a signal collector). Some software can search for these problems in the design, however, if our software does not have this capability, they must be found manually and removed from the board design.
Solder mask is very important for PCB design because the scribe lines of the etch traces are very fine and the pins are very closely spaced. The absence of solder mask can result in large slugs of solder (especially between the pins) during assembly, which can lead to short circuits. In addition, it reduces the corrosion protection of other copper on the outer layer. To prevent these problems, it is important to check the alignment between the pad to the etch line and the outline, and the spacing between solder masks (sidebands). Also, make sure that the solder mask does not cover the pins - our board factory can advise the minimum sideband space and alignment allowed.

heat sink

A heat sink absorbs and dissipates heat from an electronic device through contact with a metal substrate or thermal interface material. If the flux opening in the heat sink is too large, it may cause the device to float off the pad once the solder paste melts. To prevent this, reduce the amount of solder paste you put on the heat sink - don't use one large flux opening, instead, try dividing it into several smaller flux openings. This will help ensure that the device does not float and bump into other components during the bake process, avoiding short circuits.

Flux openings are an important part of DFM inspection. One question to answer before sending to the fab: Do the flux layer openings (and dimensions) for all component leads on the PCB fit the board?

NOTE: Our manufacturing engineer should tell us the proper size of the flux opening.

5. Cold solder joint or no solder wire

It is critical to inspect vias within the pads - if the vias are not placed properly, it can cause solder paste to flow into the vias. This will result in cold solder joints or no real solder connection. We need to determine: the percentage of vias that are allowed in the pad before the vias are required to be plugged. Note: It is the holes in the vias that are causing the problem, not the pads in the vias.

Most software should be able to check for these issues, but if the software we're using can't, we have to manually check the design to make sure it meets assembly house standards.

6. Does not include test points

It is important to test the final product as soon as it leaves the assembly line - by including test points in the initial design, we provide a way to scrutinize the success or failure of the board as soon as it is complete. The DFM inspection must include the clearance between the test point and the device, the size of the pads, the backside of the device, and a way to determine these locations immediately after the fixture is fabricated.

Then, use the test point data to create a fixture called a bed of needles tester. A bed of needles tester is a software system that locks the location of test points in a design. With a bed of needles tester, we were able to incorporate design changes back into this test fixture, saving money.

If you wait until the prototype is complete to incorporate test points, you can cause changes to the electronics on the board (which can create crosstalk, noise, and a host of other problems), so you can't really test the true functionality of the board. We will need to fundamentally change the way the design and the board work. By incorporating test points into the board during the design phase, it gives us the ability to lock down existing test points and only modify changes, if any.

Considerations when adding test points to a design: Are they accessible? Does the DM checker ensure that our test points are not hidden? What about the pin spacing (make sure they don't get too close together)?

Note: When we put test points on the board, they become part of the DFM inspection.

7. Between copper and board edge

The PCB fabrication process includes automated transport of the boards into acid baths and baths. Between the copper and the edge of the board refers to the space on the handle on the side of the PCB panel, which is used to transport the board throughout the manufacturing process. If the spacing between the copper and the edge of the board is not properly set, a real manufacturing problem can arise. If the copper is too close to the edge of the board, it can create a short circuit when the board is powered up during the etching process.

NOTE: The equipment used to manufacture the board will control the spacing required to clamp the panel - our board manufacturer should provide design specifications for this.

The consequences of manufacturing failure are not only frustrating, but costly. Forward thinking through design for manufacturability is just one of many ways to avoid encountering any DFM problems. Many of the issues listed above can be automatically identified by software such as Allegro® PCB DesignTrue DFM Technology software. However, if your software doesn't have DFM checks, you'll have to manually identify and resolve them yourself.

The last thing any engineer wants is a "call" from the manufacturer that their board has failed DFM inspection, so it's important to look for these issues during final inspection—whether through automated or manual inspection.