Filter Bypass Switching

[John Lyons]

I'm trying to bypass this high pass filter.



If I bypass the cap I still have the resistor to ground which is no good.
If I use a SPDT I can bypass the resistor and the cap but then It's
kind of messy. Anyone think of something simpler?

Thanks

[kurtlives]

Lift the pot's ground connection with the SPST?

[MikeH]

If you use a spdt you can do it.  The pole sits in the same spot, but the pot is connected via the other half of the spdt.  To one side the cap is bypassed and the pot is disconnected.  In the other, the cap isn't bypassed and the pot is connected

[MikeH]

Uber crude doodle of what I mean:

[John Lyons]

Thanks Guys! The SPDT as mikeH drew it will work.

John

[John Lyons]

Now what about a high pass filter?
(I can't think straight today  )

[slacker]

Just have a switch disconnect the cap. Or if you're worried about possible pops when switching put a 1M resistor in series between the bottom of the cap and ground. and then have the switch short out the resistor.

[John Lyons]

The pot will be 25k though, so I'd rather bypass that resistance as the same time
rather than having it in line all the time.

[JKowalski]

The pot will be 25k though, so I'd rather bypass that resistance as the same time
rather than having it in line all the time.

The signal will still be 97% of it's original voltage at the terminal of the op amp. Only 0.2 dB down, and I think the smallest volume change the human ear can hear is generally accepted as 1dB. I don't think it would matter.

[John Lyons]

Ok, thanks for the breakdown. I'll give it a shot.

John

[amptramp]

I would be inclined to use two caps in series, the one that is switched and one that is at least an order of magnitude larger so that regardless of the switch position, the steady-state DC voltages are the same.  Of course, you could also run bipolar power supplies and reference the input to ground.  Then the switch position would not matter.

[John Lyons]

I assume you are talking about the High pass filter?
There is a 1uf cap before the r/c combo of the filter.
I just omitted it for clarity in the schematic here.

John

[earthtonesaudio]

Put the pot on the other side of the 1M, between it and the opamp's input.  Assuming the opamp has extremely high input impedance, there will be no noticeable voltage division except when the cap is in the circuit.

Do use Slacker's suggestion about the large resistor to prevent pops and/or damaging the opamp input... for that matter, you could use the same approach with the highpass, just put a really large value resistor between the bottom lug of the pot and ground/Vref, and short this resistor when you want the filtering.  A really high value resistor together with your capacitor will pass all audio frequencies.

[John Lyons]

With the high pass cap, say .0047. If I put 1m between the pot and ground
and bypassed it for the filter, I still would havea .0047 capacitance in series...
How would that pass all frequencies??

[earthtonesaudio]

Like this:



With the switch closed, it's nearly the same circuit you had before.  With the switch open, you have 1.025M to ground after the cap, giving a low-end rolloff around 33Hz, which should be fine.  If you "need" 20Hz bandwidth, make the resistor larger.

(LINK)

[John Lyons]

Yeah, I got the switching idea but my point is that if the cap is small, which it will be
for the hi pass filter then not switching the cap as well will cut some bass
significantly. This will be used for bass as well as guitar.

[R.G.]

Why not use two caps, each in series with a big value resistor. One cap is smaller, sized for the high pass function, the other is much larger, sized for bass. Your switch then selectively shorts out the big value resistor for the chosen cap. This can be generalized to N caps/resistors, the cap being selected by a 1P-N-throw switch.

Especially where the caps are blocking DC and the whole thing works at a constant DC level, you can use CMOS multiplexers for the actual switches and cold-switch the CMOS from somewhere remote. The CD4051 is an eight channel selector, the CD4052 is a dual four channel selector and the CD4053 is a triple two channel selector.

[earthtonesaudio]

Yeah, I got the switching idea but my point is that if the cap is small, which it will be
for the hi pass filter then not switching the cap as well will cut some bass
significantly. This will be used for bass as well as guitar.

If you use a .0047F cap and 25k pot, then a 1.8M "bypass" resistor will pass everything above 18Hz, without attenuation, when the switch is open.  That's well below anything you can play on a guitar or bass.

[John Lyons]

Hmm, I was seeing it as the cap cutting bass as in a cap in series with the signal
with a large resistance sort of like a virtual open to ground.
So even though the cap is small it will still pass nearly full frequency response?
Since the resistance is large isn't the cap as if  by itself as a coupling cap and cutting bass?

EDIT:
Just simulated it. No freq drop at all that I can see with .0047/1M. Your right!
So how does that work then? It hinges on the resistance to ground I guess
since with no resistance the cap size would dictate the bass rolloff.

Huh...