Smart tech people: How complicated is a high pass filter?

**Chris Fitzgerald** · #1 05-04-2006, 08:41 PM

I'm convinced that a HP filter is the single most useful bit of EQ I've ever used for DB, and am thrilled that the new Focus 2R has one on both channels, and that they're adjustable to boot. But for all of the amps that don't have one, how complicated are they to make, and what's the smallest an adjustable one could be made that could be used for just about any amp? If there was a really small one that did nothing BUT high pass (especially if it was adjustable), that would be a great tool to carry in the bass bag wherever you might go. What say?

fdeck · #2 05-04-2006, 09:43 PM

Let me add a parallel question: How much would folks pay for one? I know of one way to do it, and I could be tempted to whip up a design.

Peter McFerrin · #3 05-04-2006, 09:46 PM

The tone control on an EB is probably as simple as a controllable high-pass filter gets: a capacitor, with one leg to ground and the other to the "ground" lug of a potentiometer. The saddle point is, of course, a function of the capacitor value.

We're talking about something really primitive here, though, and you'll lose output levels.

robgrow · #4 05-04-2006, 10:45 PM

Quote:

Originally Posted by Peter McFerrin

The tone control on an EB is probably as simple as a controllable high-pass filter gets: a capacitor, with one leg to ground and the other to the "ground" lug of a potentiometer. The saddle point is, of course, a function of the capacitor value.

We're talking about something really primitive here, though, and you'll lose output levels.

You are actually describing a low pass filter here. Also the point where rolloff begins depends not only on the shunt capacitor value but also on the source impedance.

Peter McFerrin · #5 05-04-2006, 11:16 PM

Oh, wait, I'm stupid.

robgrow · #6 05-04-2006, 11:42 PM

Quote:

Originally Posted by Peter McFerrin

Oh, wait, I'm stupid.

Not stupid at all... just a simple mistake. I get confused all the time.

drurb · #7 05-05-2006, 08:24 AM

Quote:

Originally Posted by fdeck

Let me add a parallel question: How much would folks pay for one? I know of one way to do it, and I could be tempted to whip up a design.

Well, if anyone can do this quickly, efficiently, and properly, it's fdeck. I must say, however, that as long as you're going to carry a separate device, why not just spend the $100 or so on the GigPro?

That way, you get an adjustable high-pass filter with a 12-dB/octave roll-off, adjustable gain, and a favorable input impedance for all kinds of pickups/transducers.

The GigPro circuit seems quite ideal in that it contains active circuitry around a passive equalization section. In English, the frequency-tailoring part of the electronics consists of capacitors and resistors and NOT op-amps and/or transistors that could add noise. This is followed or surrounded (buffered) by active (powered) components.

mje · #8 05-05-2006, 08:35 AM

What DRURB said ;-) ...except: To get a filter with a variable cutoff frequency, you're going to need to use an op-amp to keep it relatively simple. But if you put it in the effects loop of the amp, noise won't be a problem since you've already got a fairly high level signal. Fdeck- would a biquad filter be suitible for this application?

drurb · #9 05-05-2006, 08:40 AM

Quote:

Originally Posted by mje

What DRURB said ;-) ...except: To get a filter with a variable cutoff frequency, you're going to need to use an op-amp to keep it relatively simple. But if you put it in the effects loop of the amp, noise won't be a problem since you've already got a fairly high level signal. Fdeck- would a biquad filter be suitible for this application?

The GigPro does, indeed, have an adjustable frequency cutoff and that circuitry is passive. A an op-amp-based biquad? Yuch!!! You'd need at least three op-amps to implement a high-pass biquad unless you are building a biquad with an op-amp buffer, like here.

mje · #10 05-05-2006, 08:48 AM

Quote:

Originally Posted by DRURB

The GigPro does, indeed, have an adjustable frequency cutoff and that circuitry is passive. A an op-amp-based biquad? Yuch!!! You'd need at least three op-amps to implement a high-pass biquad.

Do you have a copy of the schematic? What's the topology? I'm curious as to how they vary the frequency.

drurb · #11 05-05-2006, 09:07 AM

Quote:

Originally Posted by mje

Do you have a copy of the schematic? What's the topology? I'm curious as to how they vary the frequency.

Well mje, the more I thought about this, the more I realized that this does, indeed, seem to be magic. I could not find the schematic. I did find, however, that the GigPro circuit is based on the L.R. Baggs Para D.I. which, they claim uses "quasi-passive" equalization circuitry. My statement about the GigPro was based on the company's claim. Perhaps they go a bit too far. I'll try to look into it more thoroughly.

Still, my main point stands. Why re-invent the wheel?

mje · #12 05-05-2006, 10:44 AM

I wonder what "quasi passive" means? Maybe they just mean it's a zero-gain circuit. Doing a variable frequency filter with a reasonable range and constant Q isn't easy with just FETs.

**Chris Fitzgerald** · #13 05-05-2006, 12:02 PM

I've heard complaints that the Gigpro color the signal quite a bit. Can anyone verify or shoot this down?

mje · #14 05-05-2006, 12:13 PM

DRURB were talking about how to do this offline, using a pair of FET buffers and a simple RC fiter. I may prototype it, but I may also wait to see what fdeck says ;-)

robgrow · #15 05-05-2006, 03:00 PM

Quote:

Originally Posted by Chris Fitzgerald

I've heard complaints that the Gigpro color the signal quite a bit. Can anyone verify or shoot this down?

The Gigpro has a fixed midrange dip. While it's a very nice product overall, I never could get it to sound right to my ears, so I ended up selling it. The adjustable HP filter, labelled "Trim", is the best feature of the Gigpro IMO. Too bad they don't make the mid filter switchable.

fdeck · #16 05-05-2006, 04:32 PM

You guys are too kind

I started drawing up a biquad, then read DRURB's comment and was embarrassed into looking up a simpler filter type:

More details here: http://en.wikipedia.org/wiki/Sallen_Key_filter You vary the corner frequency by making R1 and R2 the sections of a dual pot. So long as R1 = R2, the Q is constant over the control range -- a convenient result. You solve an equation in two variables, C1 and C2, to set the corner frequency range and Q. Or you model it and adjust by hand until it looks nice. I will model this tomorrow using LTSpice, and post recommended values.

What is this Q? It determines the shape of the response curve near the cutoff. Ideally, it is flat until it starts to drop suddenly, without a "hump." Speaker designers also worry about Q for the same reason. You could design a filter with just two successive RC filters and a buffer in between -- the Q control of the Sallen-Key lets you set a sharper cutoff. The biquad filter gives you independent Q control, but we don't want that here, I think.

I would use an op amp, and add a section at the front for piezo buffering. You could have a dip switch or jumper for 0 dB or 10 dB gain, so the circuit would be usable as a piezo preamp or in the effects loop.

Also, since the op amp is a pure buffer, I will model the circuit with two JFET source followers. Either way, given my favorite op amp type, the thing could run for a couple hundred hours or more on a 9-V battery.

Thinking about pricing something like this, I would multiply my parts cost by anywhere from 3x to 4x. This number is based on my experience selling other small electronic products and having an honest assessment of my hidden business costs. Also, I have had good luck in the past with an "open" design, where the schematic and supporting data are freely available.

To be on the up-and-up, if I begin thinking about going commercial, then I have to inform the moderators of TalkBass that I have become a vendor. So I had better read the forum rules again.

mje · #17 05-05-2006, 07:05 PM

What's the advantage of using the op-amp- parts count? LR Baggs brags that they do a "quasi-passive" eq- and I'm guessing this would be what they're doing- with FETs alone.

fdeck · #18 05-06-2006, 12:06 AM

Quasi passive sounds like maybe passive filters with buffers in between. I think the op amp gives you the chance to have a higher Q without inductors, but the circuit shown above is not the best choice. Oops. The Q of that circuit can't be greater than 1/2. What I think you may need is the Butterworth filter. Maybe I need something like this:

It's late. This problem occurred to me while playing my gig tonight. I will take another look in the morning.

drurb · #19 05-06-2006, 10:11 AM

Quote:

Originally Posted by fdeck

Quasi passive sounds like maybe passive filters with buffers in between. I think the op amp gives you the chance to have a higher Q without inductors, but the circuit shown above is not the best choice. Oops. The Q of that circuit can't be greater than 1/2.

I'm not sure what you mean by a Q of 1/2. Q, for those who are not into geek-speak, is the center-frequency (CF) of a filter divided by its (usually 3-dB down) bandwidth (BW). Thus, holding CF constant, as BW decreases, Q goes up, indicating a sharper filter. The term comes from the old radio days in which the objective was to build very sharp filters into the radio-tuning circuits. The sharper, the better. These filters were referred to as having a higher "quality-factor," thus "Q".

Now, Q is not really a proper metric for a high-pass filter because the bandwidth is (theoretically) infinite. The filter passes all frequencies ABOVE a certain "cutoff." The two most important parameters of a high-pass filter are its cutoff frequency and its rejection slope. The latter is the rate at which frequencies below the cutoff frequency are attenuated. It is specified in terms of dB per octave.

So, a simple high-pass filter with a cutoff of 50 Hz and a slope of 12-dB per octave will pass all frequencies above 50-Hz. The response at 50 Hz will be 3 dB down. Each octave decrease will see another 12-dB drop, i.e., at 25 Hz it will be 15 dB down, at 12.5 Hz it will be 27 dB down.

drurb · #20 05-06-2006, 10:15 AM

Quote:

Originally Posted by robgrow

The Gigpro has a fixed midrange dip. While it's a very nice product overall, I never could get it to sound right to my ears, so I ended up selling it. The adjustable HP filter, labelled "Trim", is the best feature of the Gigpro IMO. Too bad they don't make the mid filter switchable.

I'm curious, have you measured this? It would be interesting to perform a spectral analysis on one of these while it is loaded with typical working impedances.