• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

High pass filter via PEQ?

fatoldgit

Addicted to Fun and Learning
Joined
Feb 29, 2020
Messages
609
Likes
852
I have done some googling on this and while it appears it's possible, the details I saw didn't provide enough info for my addled old brain.

To me I don't see how a PEQ config could create a high pass filter at say 80HZ where everything below 80hz rolls off and all above passes through unmolested.

Thanks,

Peter
 
set f as low as possible (20Hz, 16Hz) and get the high pass effect by setting the attenuation and Q
 
What is the issue you are encountering?

1763275008207.jpeg
 
I have done some googling on this and while it appears it's possible, the details I saw didn't provide enough info for my addled old brain.

To me I don't see how a PEQ config could create a high pass filter at say 80HZ where everything below 80hz rolls off and all above passes through unmolested.

Thanks,

Peter
Some Parametric EQ software has selectable filter types, including high-pass filters so that'd be trivial.

Are you asking how one could create a high-pass filter using Peaking (Bell) filters alone?
 
set f as low as possible (20Hz, 16Hz) and get the high pass effect by setting the attenuation and Q

Can you give me a real world example within some software (say a graphic with this config)?
 
Can you give me a real world example within some software (say a graphic with this config)?
chatgpt: f = 20Hz, gain = -20dB Q = 2
with Q you can now change the width and also the effective point where the level starts to drop. (here at 40Hz or so)
setting the q to 1 will get you closer to 80Hz.. you have to experiment a bit get it right
Far from ideal, but it does 'some' high pass filter...

1763279382460.png
 
Of course it is possible to generate a HPF using PEQ's. The question is, why would you want to?

1763279658701.png


Here is a quick proof of concept. First, we create a 4th order Linkwitz-Riley HPF with a corner freq of 500Hz.

1763279781242.png

1763280158526.png


Then we go to REW's EQ editor and create a flat target. Click "match response to target" and REW starts assigning PEQ's to match the HPF. Then click "Generate measurement from predicted".

1763279810800.png


This is a comparison of the original HPF (red) vs. our HPF created with 10 biquads (green). I did a very quick bodge job for the purpose of this demonstration, I should be able to get a much better match if I spent some time and effort on it.

1763279900701.png
1763279882267.png


And here is a comparison of the phase and impulse response.

Why you don't want to generate a HPF using PEQ's: It wastes biquads. You need 2 biquads to generate a 4th order HPF, but more than 10 PEQ's (1 biquad each) to do an inferior job. And the result? As you can see, it's not quite the same.
 
Some Parametric EQ software has selectable filter types, including high-pass filters so that'd be trivial.

Are you asking how one could create a high-pass filter using Peaking (Bell) filters alone?

Yes, which is the typical PEQ/EQ functionality that DAC's provide (as opposed to the DEQX I use at the moment which has options galore in regard to filtering, EQ etc)

The "issue" is many of the manuals don't provide any detail on the PEQ config options.

For example, the Topping d900 shows this on the product info web page but the manual provides no detail as to what other options are in the drop down list (my general gripe with Chinese products is they don't invest any time on the manuals)

Which started me on the trail where if I assume just a Peaking filter then could you create a high pass filter...and it was stated on various sites that you could but it didn't make sense to me given, as you stated, it's bell curve nature.


1763279703570.png
 
chatgpt: f = 20Hz, gain = -20dB Q = 2
with Q you can now change the width and also the effective point where the level starts to drop. (here at 40Hz or so)
setting the q to 1 will get you closer to 80Hz.. you have to experiment a bit get it right
Far from ideal, but it does 'some' high pass filter...

View attachment 490900
thanks!!!
 
that assumes the PEQ software supports a high pass filter.. that's easy enough when it does.
A highpass/lowpass filter is actually mathematically simpler than a shelf filter.

Below is a shelf filter, low shelf filter to be specific:
1763279899657.png


The idea behind a low- or high- pass filter is simply that at a certain frequency that you want to filter out, the voltage at output drops to zero. Usually means the ratio of impedance A vs impedance B is infinite. Capacitors are able to change between having practically infinite impedance and practically zero impedance depending on frequency.

If we short R2 above, the RC forms a low-pass-filter because at high frequencies, C2 impedance is zero. But what if we want a shelf filter i.e. we want the stop band to have a certain attenuation value or ratio that is not zero? We add R2 such that at frequencies where C2 has zero impedance, the output becomes R2/(R1+R2).

In other words a shelf filter where R2 is infinitesimally small, or basically where the attenuation value (or gain of the stopband or shelf is infinitely negative), is mathematically equivalent to a low or high pass filter.

So basically, if your PEQ has a low shelf filter, just do whatever you would do with a shelf filter, and set gain to be as negatively big as possible.

Can you give me a real world example within some software (say a graphic with this config)?
Actually we could ask the same... can you give us what PEQ you are working with
 
Actually we could ask the same... can you give us what PEQ you are working with
I am using a DEQX HDP-4 so it has all the "good" stuff but some dissatisfaction with it over the last few years has had me gradually remove it's "higher order" functions so I am just left with a high pass filter on two of my four output channels.

I could move to something else that does lots of "stuff" but that just sticks me back in the same boat with functionality that I no longer want/need and as noted above, it's hard with many "normal" DAC's with PEQ to determine just what functionality they have as the manuals are lacking any detail.

Hence the question that if I assume just Peaking PEQ, would that work for a high pass filter.

It's appears it can, so I can play around with Peaking PEQ inside my DEQX HDP-4 (to prototype) and see if I perceive any issues compared to a real high pass filter (can do instantaneous compares with different DEQX HDP-4 profiles). If that works out then I can confidently move on.

I played around with Camila DSP but that means I then need a multi-channel DAC... which defeats my move to a more simplier "lifestyle".

Preversely, having retired early at 58 (now 65) and therefore with any amount of time on my hands I have dropped vinyl (collected since 1970), CD's and music DVD's and am now computer based, with my CD's and music DVD's ripped. I found that after I retired I just didnt want to bugger around with physical media anymore.

This stripping back has lead to me removing lots of hardware from my system (turntable, phono amp, CD transport, DVD transport, pre-amp, DAC's etc) and the removal of the DEQX is the last step on this journey... removing unwanted logical functionality rather the physical functionality.

Thanks to all who replied.

Peter
 
Last edited:
Back
Top Bottom