really quick 1 liner

Started by Papa_lazerous, September 19, 2007, 05:33:32 PM

Previous topic - Next topic

Papa_lazerous

I've read it before but cant hit a result with search,

at what Frequency would I encounter problems by running pcb traces at 90 degrees to each other??  obvioulsy its going to be quite high but just wondering for purpose of possibility of LFO signal affecting other parts of the circuit.

R.G.

The question as stated is not answerable.

The issue of traces at 90 degrees is usually one of changing the characteristic impedance of the equivalent transmission line of the trace, so its only significant where the frequencies have wavelengths in copper that are four times as long as the traces in question - which you'll never get to in an audio setup. But capacitive coupling works at audio, so other traces running close can cause LFO leakage; so can inductive coupling by inadvertent loops of conductor.

Finally, the LFO waveform matters. If it has sharp corners, that may couple nicely because sharp corners on the waveform require high order harmonics to generate, and that means capacitive coupling of the sharp edges is easier.

Don't sweat the 90 degree corners. That's not your problem. Shared sewer grounds or power traces will nail you, though. Most circuit interference is either high impedance capacitor coupling or a matter of what current goes where.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

Papa_lazerous

Thanks RG, I thought I had read something about 90 degree traces somewhere, it would well be the information was flawed to start.

Paul Perry (Frostwave)

Quote from: Papa_lazerous on September 19, 2007, 05:33:32 PM
at what Frequency would I encounter problems by running pcb traces at 90 degrees to each other?? 

Above 200MHz. What happens, is  a trace on a board with a ground plane on the other side has a characteristic impedance, and as you change the geometry the impedance changes. With every discontinuity of impedance, there is some reflection of signal back to where it came from. So you end up with the sharp edges of pulses being smeared out & echoed. Nothing whatever to do with the frequencies we work at in stompboxes though. If it WAS a problem, none of those rat nest prototypes or handwired FX would work!

focus.ti.com/lit/an/scaa082/scaa082.pdf on what can go wrong

www.montrosecompliance.com/Technical_papers/corners-USA.pdf on why it doesn't normally matter