General pull-down resistor question

Started by ildar, October 12, 2005, 11:32:00 PM

Previous topic - Next topic

ildar

Are pull-down resistor values critical, or do they need to simply fall within a certain range?
Do they have any bearing on the input or output impedance of the effect?

R.G.

Let's think about it for a minute.

Pull down resistors are there to pull the leakage current of the input and output capacitors to ground, or so close to ground that you don't hear a click when switching. Feel free to use any value you like that makes pops go away. We've found empirically that values between 10milliohms and 10 meghoms are all pretty effective at stopping clicks, but most people use a good, solid run of the mill 1M since the lower values tend to reduce signal levels too much. For stopping switching clicks, smaller is better. One inch of copper wire is a good preventer of clicks. No signal, but no clicks.

To keep clicking from happening, pulldown resistors have to be significantly less than the leakage of the capacitors. How big is that? We don't know without measuring, because makers of caps only say "it's less than xxxx." Therefore, it's not possible to calculate a priori what the theoretically right value is without knowing the specific cap's leakage.

As to having a bearing on the output impedance of the effect, yes, a pulldown resistor appears in parallel with the Norton equivalent output impedance of the effect, so it lowers the output impedance. It lowers the output voltage as well. How much? You gotta know what the output impedance of the effect is before you put the pulldown on. Generally, the output impedance of the effect is down in the sub-50K region, so 1M and above is comfortably big enough not to load things down much.

And on the input, the same is true. The pulldown is in parallel with the input impedance, and acts like a load on whatever is driving the input. So the previous effect, if it has a pulldown as well, has to drive both its own output pulldown and the input pulldown of the next effect, in addition to the actual input impedance of the effect itself. So from the standpoint of preserving signal and not loading things down by changing the input and output impedances too much, bigger is better.

Having thought through this once or twice, most people sigh, and just use 1M. I like 2.2M. But be creative!!
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.