1
\$\begingroup\$

For a PCB design, the longer the track length the greater the parasitic inductance right?

But why would this be a concern if I'm working at low frequencies (e.g 50-60Hz) because it'll behave similar to a short circuit right?

\$\endgroup\$
1
\$\begingroup\$

A track starts behaving reactively at the point where the wavelength of your signal approaches the order of magnitude of the size of the track. For 60Hz, the wavelength of your signal in copper is about 5000 km, so i wouldn't worry about couple-inch tracks on a board.

For very large current signals (i.e. power transfer) the inductive properties will start to be visible a bit earlier, especially in low voltage designs, because \$v=L\frac{dI}{dt}\$ holds. Then still, you're talking about at least the tens of MHz range for any measurable effect on a typically-sized PCB.

\$\endgroup\$
  • 2
    \$\begingroup\$ Rule of thumb is to start worrying about track length vs. frequency when \$\lambda \text{[m]}= \frac{1}{10} \cdot \frac{c \text{[m/s]}}{f \text{[Hz]}}\$ where \$c \approx 3 \cdot 10^{8} \text{[m/s]}\$ In other words where the track length is one tenth of the wavelength. So in this case, provided that your PCB track is shorter than 500km, you shouldn't worry too much. \$\endgroup\$ – jippie Sep 25 '13 at 6:41
1
\$\begingroup\$

It all depends on what your circuit does. If you are taking signal representations of AC waveforms and putting them into comparators and logic to calculate timing or peak values (etc.) then track lengths do matter because your circuit technology may rely on short leads such as for: -

  1. Decoupling capacitors on chips
  2. Signal 0V integrity
  3. Power rail spike minimization
  4. Reduction of high frequency glitches on logic outputs/inputs

However, if your circuit is just controlling relays and lamps then you're probably OK.

\$\endgroup\$
  • \$\begingroup\$ Ditto microphone amplifiers and other sensitive audio, LF or DC circuits, or in a noisy environment! Here the parasitic inductance won't affect the waveform directly, but it will couple all sorts of external signals you don't want onto your signal. \$\endgroup\$ – Brian Drummond Sep 25 '13 at 7:58
  • \$\begingroup\$ @BrianDrummond there is potentially a monstrous list of reasons to keep track lengths short even when dealing with DC signals!! \$\endgroup\$ – Andy aka Sep 25 '13 at 8:14
  • \$\begingroup\$ Agreed and then some! hence my +1 : but as the question specifically focuses on parasitic inductance, the comment was re: magnetic coupling (and probably should have been more explicit). Short tracks and minimum loop area... \$\endgroup\$ – Brian Drummond Sep 25 '13 at 8:58

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.