Parallel RC filter questions?

[MikeH]

I've been able to find a lot of info in the net about parallel RC filters- but not a lot in the context of audio.  I'm wondering how to change the frequency of one, and how big of a difference a given component change will make.

I'm trying to fatten up my Hotbox:

http://www.generalguitargadgets.com/diagrams/hotboxsc.gif

I recognize the parallel 220K and 560p combo at the end of the clean channel is a parallel RC filter and therefore would be a high pass and is probably responsible for the bright sound of this thing.   I'd like to bring back some of the low end, I'm just not sure which of the values to change and to what extent.  I'll most likely bump up the 0.022u coupling cap after the first stage, but I don't think that alone will get me there.

Thoughts?

[CynicalMan]

The cutoff of the 0.022 capacitor's filter is only 29Hz on the clean channel so changing that wouldn't help you much. The 560p capacitor forms a high-pass filter, but with the input impedance of whatever comes after it. That means that if the next component's input impedance is low, the sound will be bright, and if it is higher, the sound will be less bright. You could try temporarily putting a capacitor in parallel with it, which would be easily reversible, to see how much of an effect it has. Other than that, removing the 180p cap would reduce the brightness at lower levels on the clean channel, but that capacitor and the 560p capacitor are the only places where bass is being significantly cut on the clean channel.

Edit: I didn't notice at first but treble is being boosted by the the tube stage in the clean channel. Increasing the 0.22u capacitor to 2.2u or higher would give it full frequency response.

[merlinb]

The cathode bypass capacitor on the first tube is currently cutting frequencies below about 300Hz. You could easily increase that capacitance to regain some low end.

[MikeH]

Cool - I'll try that first. Thanks guys.

[MikeH]

That brought some lows back, but it's still brighter than the bypass.  I'd like to get it to the point where it's fairly colorless (in terms of EQ).  So, if I'm thinking about it right, I need to either decrease the 220K resistor, or increase the 560p resistor, to get some more lows, correct?  Although I'm not sure if the problem still doesn't lie with the cathode bypass cap.  How do I figure out the frequency?  I'm using a 2u2 cap- I know this cap can be a lot bigger in most amps, maybe I need to go bigger?

[ayayay!]

So, if I'm thinking about it right, I need to either decrease the 220K resistor, or increase the 560p resistor, to get some more lows, correct?

I'd drop it to 100k and omit the cap completely.  If it's that bright all the time and you've already recovered some of the lows, that should get you there.   

[CynicalMan]

The 2.2u capacitor will roll off bass 1dB or so, but the 560p capacitor probably has a much bigger effect. I agree with ayayay!, you can just remove it.

[PRR]

_I_ don't know what a "parallel R-C filter" is.

> Hotbox: I recognize the parallel 220K and 560p combo at the end of the clean channel is a...

That's NOT a filter until you specify a load.

If the load is infinite, there's no filter action at all.

Let's specify a 220K load (some gitar amp inputs are this low).

At low frequencies the cap is dead-open. So lows pass through 220K:220K divider and are cut in half.

At high frequencies the cap is dead-short. Highs pass through short:220K and come through full strength.

(There's a correction for cable capacitance; indeed all audio "220K" loads have some capacitance. But we'll skip that today.)

So WITH a load it boosts the highs. And some number-smashing suggests it is only the extreme highs.

Bass and middles are just the 220K against the unknown load.

First, FIX the load. Add 100K or 150K from the "clean" switch lug to ground. Does this allow a proper balance between clean and distort, listening only to bass and middles? If too weak, go up to 220K, 330K or so.

You may also, as Merlin says, re-consider the clean cathode bypass. It's pretty gutless, which gives an impression of "too bright". 5uFd-10uFd would be "full bass". Marshall used 0.68uFd with 1.5K for a more-tenor sound. Scaled to 2.2K you might use 0.47uFd. 0.22 is pretty bass-less for "clean guitar". Put in 5uFd-10uFd; if that's too much thunder try 2uFd 1uFd 0.5uFd caps. Let the "clean" channel boom a bit, to contrast the distort channel's tenor-soprano scream.

Now fix the treble. Remove the 560pFd. How's that?

There's also a 180pFd boosting highs around the "clean vol" pot, cut that off.

Guitar usually wants some top-lift. Cutting both caps is maybe too dull. Certainly not a "wrong" tone. If with sufficient bass, it's dull, add back the 180p or 560p. They interact, and the 180p interacts with Vol pot position, it's not intuitive. I can see reasons to do just the pot bypass, just the 220K bypass, or both or neither. Hot-up the soldering iron, collect 100p 270p 500p 1n caps, try different combinations with different music at different Vol settings.

[ayayay!]

There's also a 180pFd boosting highs around the "clean vol" pot, cut that off.

Guitar usually wants some top-lift. Cutting both caps is maybe too dull. Certainly not a "wrong" tone. If with sufficient bass, it's dull, add back the 180p or 560p. They interact, and the 180p interacts with Vol pot position, it's not intuitive. I can see reasons to do just the pot bypass, just the 220K bypass, or both or neither. Hot-up the soldering iron, collect 100p 270p 500p 1n caps, try different combinations with different music at different Vol settings.

I almost suggested that too (rem 180pF), and certainly you could/should try it.  I thought that controlled the volume for both channels and might make the dirty channel muddy, but I see it doesn't now.