# Testing a board without placing a component

Is there a way to build a (functional) board without placing 1 component?

I am designing a PCB and want to have the option to make it work w/ or w/o placing a specific component (X). This component X doesn't impact the main feature of the board.

One solution of this problem is using a jumper - design in a way so that the jumper on/off can make the circuit close/open. But I don't want to use jumper in my board.

Any other approach?

• In the absence of the component, you're saying that a jumper would be required instead? – JYelton Oct 10 '14 at 5:41
• I have seen in some TI reference design doing that. But I do not want those jumpers. – nad Oct 10 '14 at 5:43
• Some PCBs use "solder jumpers". This is just two pads, usually each a half-circle with the flat sides facing, which you close by putting a solder blob across the two pads. No extra wires or components. Would that be acceptable? – DoxyLover Oct 10 '14 at 7:53

There are lots of ways to avoid jumpers. You can use analog switches, digital switches, mechanical switches or other means of customizing a board such as external connections, traces which might be drilled out to open a short for the 'high end' version, solder shorts etc.

It's very difficult to give specific advice without knowledge of what the part and limitations are. Generally a machine-mounted 0R resistor will be almost free, so that would be my first choice if it's a variant decided at manufacturing, never to be changed. Then your option is to place expensive component X or (almost free) component Y. If you need multiple shorts, a machine mounted 4-resistor 0 ohm array is very, very cheap and uses little board space.

You can even buy big expensive-looking 'PQFP' chips that include shorts internally, which you can get house numbered so your design is harder to copy (but they're not cheap).

• Dare I add a socket could be used for said IC. I've got a pcb where one has to drill out two vias to make part of it active. I can't remember why I choose to do it that way, I guess I thought it was a good idea at the time. :^) – George Herold Oct 10 '14 at 13:16
• @GeorgeHerold hey, the 1990s called and they want their techniques back. ;-). I specd sockets so we could build up quantities of boards without tying up  parts in the supply chain, but it's getting hard to do that with the lack of cheap sockets for expensive parts (PC motherboards being a noteable exception). – Spehro Pefhany Oct 10 '14 at 22:32

It's common for commercial PCBs to be made in such a way that the "fully loaded" version of a device and "budget" versions use the same board. Components are simply not populated on the budget version, and the manufacturer doesn't have to create separate boards.

In some cases, the lack of the components simply do not impart whatever functionality they provide, and nothing need be done to the board if they are missing. As a simple example, consider a board that provides additional reservoir capacitors connected in parallel, but some can be left out as conditions warrant; no connections or changes need be made if they aren't there.

However if you have a component that does actually need to be present for the device to function, then you must have some sort of alternative component if it isn't populated. This may be as simple as a resistor that is replaced with a jumper.

If the unit doesn't function correctly without the component, then that contradicts your statement about it not having an impact.

You said that one solution is to use a jumper to close the circuit, but you want to avoid using a jumper. Why? Is it something that you need to be able to toggle more easily? If so, why not use a switch? You could design the board to have placements for both the "X" component and a switch (in parallel), then populate whichever one is applicable.