9v and ground on adjacent tracks

[haryot]

Hi!
I've converted a rather complex schematic to a very tight stripboard layout, and my 9v and ground tracks are running right next to each other. I've seen in some places recommendations to avoid this, but I couldn't find anywhere an explanation of why. Is this a real issue, or can I keep my layout as is?

[Digital Larry]

Main reason I can think of is that it's relatively easy to short out.  Beyond that, it's simply a matter of optimizing your layout and for audio stuff it hardly matters as long as you don't run things which generate sharp pulses of current (like a 555 timer) right next to a high impedance circuit with a lot of gain in the audio path.

[ElectricDruid]

In theory, two parallel conductors with an insulator between them are a capacitor. And in practice...they're a capacitor too!

However, with the voltages we're talking about and the spacings and so on...I wouldn't worry about it. If I was trying to route some high speed MHz comms signal or something, I might start to be more concerned, since the capacitance can load the line and distort the digital pulses, and the signal can get into parallel lines. That said, my Digidelay prototype was done on stripboard and used a highspeed SPI comms link,
and while I did try and keep the connections short, it worked fine.

So "Yes, it's a real issue" and "No, I wouldn't worry about it in this situation".

Larry's "it's likely to short out" is a much more serious concern, frankly. There's a real and immediate risk there.

[R.G.]

It's not a real issue, other than as DL mentioned, if there were not solder mask on the board, it might be easy to short.

Placing power and ground traces close together maximizes the capacitance between them. This is usually done by routing one over the other on opposite sides of the board. This minimizes the area of the (AC) current loop they create, and so makes for less noise and other pickup issues. The capacitance between them is quite small on a single sided board, but it does again minimize the loop area.

The capacitance between them ...can... both couple noise from one to the other as well as acting as an RF filter.

But all this mess is so small as to not matter on a board inside a pedal >>>unless<<< you also run a high impedance trace, like the input to a JFET opamp or a MOSFET gate lead next to any other trace that has noise on it. The ultra-high impedance can pick up noise easily in that kind of condition. But you have to worry about that with a gate trace and every other kind of trace, so it's not different for a ground next to a power trace.

[haryot]

Thanks for all the quick and helpful replies! I'm glad this isn't too much of an issue, but R.G got me worried about my layout there - I do have some of those op-amp inputs running in parallel to ground or 9V, and one of those parallels is going on for quite a few holes... How bad is it? Should I go for a new layout?

[R.G.]

You're down at a level of detail that requires actually looking at it for me to render a suitable answer.

Some of the relevant questions are:
- How long is "quite a few holes" ?
- Which opamp input is it, inverting or non-inverting? Inverting inputs are usually held at virtual ground by feedback.
- What is the opamp doing in the circuit? Does it process audio or only do some other function like envelope detection?
- How much gain is involved? If it's down at unity, you have little chance of issues. If it's running at high gains, it may be quite sensitive. 
- What currents flow through that ground or power trace? That is, how noisy are they?

The simplest thing to do is to build one and see if you have problems. Test the prototype by first trying it with the intended power supply and situations, then by deliberately varying the power supply voltage, injecting noise on it, inserting resistances in the power leads to fake a poor power supply and so on. Mother Nature is quite a stickler for Her rules being followed, and She will remind you where you have forgotten one of the Laws.