1
\$\begingroup\$

Imagine a rather noise sensetive application, My question is if there are unoccopied pads for SMD and through-hole are existing on the board and somehow connected to other components through theire trace lines, what would be the side effects?

I have the following PCB which capacitors or resistors are soldered on it (only one of the footprints will be used). What will be the effect of the unsed pads? (This is to measure capcitance/Inulation Resistance):

enter image description here

\$\endgroup\$
4
\$\begingroup\$

The extra pads, like all parts of a net, will have extra capacitance to surrounding nets. Whether that matters is hard to tell. If you have a nearby noise source that will capacitively couple onto your traces and this noise is significant, then it probably will matter.

One way to deal with this is with shielding. That trades off more capacitance to ground in return for greatly reduced capacitance to the noise source. If capacitance to ground isn't a issue, then this is a good tradeoff. One difference to consider between thru hole and SMD pads is that SMD pads are only on one side of the board. If they are on top, then the bottom layer could possibly be used as a shield, eliminating coupling from noise on the bottom side of the board. This doesn't work with thru hole pads since they are on both sides of the board.

If the noise source is external to this board, put it in a metal box. If you are worried about coupling between components on this board, then ground or guard traces may help.

\$\endgroup\$
  • \$\begingroup\$ Thanks Olin. What can be an example of a capacitive noise coupling source? Maybe PC's or other instruements? Is there a simple way to test if on my workbench there is more than normal noise present? Sorry for stoop-id questio! \$\endgroup\$ – Sean87 Jan 2 '12 at 13:48
  • 1
    \$\begingroup\$ Anything electric on your bench could be a noise source, including but certainly not limited to your scope, the power supply in your soldering iron, the flourescent light, and the 50 or 60 Hz from the power line. Then there are more distant but high power intentional radiation sources like radio stations, your WiFi router, etc, etc. \$\endgroup\$ – Olin Lathrop Jan 2 '12 at 14:04
  • \$\begingroup\$ Another side-effect of unpopulated components can be unwanted signal reflections. If the component was populated, the current would continue to flow; unpopulated, once the current "reaches" the end it will reflect, adding large ringing to the signal. Many high-speed chips have datasheets with PCB design tips that include minimizing such stub traces, even going to far as to place the pad directly on top of the trace with zero stub. In the worst case, a long, unpopulated stub could begin acting like an unintentional antenna, receiving and transmitting noise to nearby electronics. \$\endgroup\$ – ajs410 Apr 11 '12 at 17:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.