The Flexperts
5 Options for Connector-Less Flex Jumpers
Adding solderable fingers to connect rigid boards.
When designing a flex jumper between two rigid PCBs, where no room exists for connectors, are solderable pins or tabs that extend out of the edge of the flex circuit a possibility? If so, what are the cost and reliability implications?

Quite a few options are available, each with its pros and cons and cost implications. This month, we look at the possibilities.

Sculptured fingers. This construction yields unsupported copper fingers that extend beyond the circuit outline. The fingers are typically ~0.010″ thick. These can be made by starting with very heavy copper (usually half hard) and etching down all areas other than the fingers, or starting with thinner copper and plating up only the finger areas to meet the desired overall finger thickness. The fingers can then be formed to best fit the desired applications (FIGURE 1). The cons to this construction are cost and handling issues. Adding sculptured fingers to a flex circuit will have a modest cost impact, but the biggest downside is these parts are fragile. The fingers can be easily bent out of shape during shipping or handling on the production floor and are difficult to realign once damage occurs.

A close up of an orange jumper is depicted with copper fingers beside a rendered image of another jumper with bent fingers
Figure 1. Sculptured fingers can be formed to fit the application.
Brazed or welded fingers. This construction also yields unsupported fingers that extend beyond the circuit outline. The fingers can be thick copper (I recommend half hard or full hard copper), but more often they are a more durable material such as nickel. While this construction is more robust and tolerant to handle than sculptured construction, the fingers can still be damaged if mishandled. Cost implications vary from vendor to vendor, depending on the level of automation incorporated when the fingers are attached.

IDC contacts. Insulation displacement contacts use special pins with “claws” that pierce the polyimide insulation and contact the base copper (FIGURE 2). These contacts are reasonably durable and come in a range of configurations. The downsides are cost and limited minimum pitch between contacts, and only vendors with IDC equipment support such needs.

Close-up image of a jumper with "clawed" pins that pierce insulation to make contact with the base copper
Figure 2. IDC contacts pierce through insulation to make contact to copper traces.
Unsupported fingers. This construction is typically created by using a laser to ablate the dielectric material covering both sides of a row of parallel copper traces. Unlike sculptured fingers, the finger thickness in this construction will be the same as the rest of the circuit (i.e., very thin). This construction always requires a tie bar on the end of the fingers to help prevent damage. This construction is most often used to connect a flex to a rigid PCB using hot-bar soldering. Unsupported finger circuits are typically only one of two layers, so the additional processing steps associated with the laser ablating and cleaning have minimal cost implications.
Supported fingers. This construction yields the least expensive and most robust final product. The fingers are only exposed on one side, so no pre-punching of the base dielectric or laser skiving operations is necessary (FIGURE 3). The base polyimide strengthens the fingers, making them less prone to damage during handling. If adhesiveless base laminate is used, the resulting construction can be hot-bar-soldered through the polyimide base to a matching set of fingers on a rigid PCB. Alternately, anisotropic adhesive can be used to make the connection the same way. Either of these connection methods provides a very low-profile and reliable connection. The downside is it is difficult to inspect the hot-bar-soldered or anisotropic adhesive connections because you would have to look through the polyimide to see the solder or adhesive joint. Some designers will specify the polyimide material between the fingers be removed. This isn’t free, but it adds only modest cost to the finished flex. The base polyimide remains on the back side of the fingers to help strengthen them, but the material between the fingers is removed to permit inspection of solder joints or anisotropic adhesive connections. If anisotropic adhesive is used, and the polyimide between the fingers removed, be sure the hot bar used to join the flex and rigid PCB has a non-stick finish! If not, there is a good chance the exposed adhesive between fingers could stick to the hot bar.
Keep in mind all these options will result in a more fragile connection than standard connectors, and I highly recommend not using them in dynamic applications. Even when used in static applications, it is advisable to look at ways to ruggedize the connection area to protect it from damage. The most common method is a coat of semi-rigid epoxy over the connection area. A less-expensive option is to add a strip of pressure-sensitive adhesive to the flex adjacent to the connection area. The release liner can be removed just prior to lining up the fingers and can aid in holding the flex in place during hot-bar soldering or laminating processes.
A mid-range shot of the most cost-effective option; a jumper with supported fingers
Figure 3. Supported fingers are the most robust and cost-effective of the discussed options.
Portrait photo of Mark Finstad
Mark Finstad
Mark Finstad is senior application engineer at Flexible Circuit Technologies (flexiblecircuit.com); mark.finstad@flexiblecircuit.com. He and co-“Flexpert
Nick Koop
(nick.koop@ttmtech.com) welcome your suggestions.