GlassBlower - switchable high-pass filter?

[tristanc]

I was thinking of making a version of Merlin's GlassBlower booster pedal but with a switchable high-pass filter to cut a bit of bass. I've noticed boosting my amps with a Boss EQ pedal can become a bit wooly if there isn't a bit of bass cut.

There are 3 obvious places to put the cap switching - C1 on the input, C7 on the output, or C3 in the middle of the two opamps.

Are there any pros / cons to each location?

Altering C3 could potentially give an interesting benefit - the higher the Boost setting, the more the bass is cut. But would there be any unwanted issues?

[Mark Hammer]

C3 is your safest bet, and most easily calculated.  You can also use that location to use the "dual-tailed" Rat strategy (used in other pedals as well, like the Ibanez Fat Cat).

Rather than using a single path from the inverting pin to ground, the Rat has two such paths: one setting the basic gain for the entire applicable bandwidth, and the other setting additional gain for mids and highs. 

Why is this potentially better?  A simple bass rolloff is progressive.  So if one were to drop the value of C3 to achieve some degree of bass cut, that cut would continue, at a slope of 6db/oct.  I.E., 12db down two octaves below, 18db down three octaves below, etc.  Achieving a sufficient bass cut at some given corner frequency might mean "starting" the rolloff higher than you actually want.  Of course, there is always the possibility of inserting a 2 or even 3-pole highpass filter on the output.  But then it starts to become a more complicated circuit.

The two-tailed strategy lets you set less gain for this part of the spectrum and more gain for another, without having to add or adapt anything to the output structure of the circuit.  Functionally, this is not really any different than turning the bass down and the treble up on your EQ controls.

[tristanc]

Thanks Mark - interesting idea. Is the below what you meant?

Attached is an example borrowed from the ProCo Rat. On the SPICE schematic attached I added R14 & C8, and modifed R8 & C4 to give poles at 1539 Hz and 72 Hz, respectively.

So with the pot up full there's a ~4dB difference between 250Hz and 2500Hz, and ~3dB from 250Hz to 80Hz (green line).

Turn down the boost and the response flattens on the top end, but there's still a bass cut-off (blue and red).

I increased the value of R8 to 4.7k and made R14 equal to give a similar overall gain to the stock circuit at the top end.

I wonder what poles would be the most appropriate. If the approach is good it could be a bit of trial and error on assembly.

T

[Mark Hammer]

Yup.

[tristanc]

I've just built this as per the schematic below.

The 'top boost' (or bass cut) works very well - just preventing some wooly-ness and keeping things tight. Works very well in to my October (jumpered Plexi) channel (light crunch to louder crunch) and also my higher-gain channel (mid distortion to heavy distortion). The impact of the switch is very noticable. Job done. Kudos to Merlin for another great design.

It's the first time I've used Aisler.net - where you can get the PCBs (minimum 3 of) and parts from DigiKey in a single order, based off your BoM. Really handy. Not as quick as PCBWay / ordering parts separately, but still recommended for those in the EU.