LF shelving frequency shift

[MagnetoSound]

This is the EQ section of my DDA S-series studio console.

In stock condition it has a LF shelving section with a corner freq of 100Hz, which I would like to drop to a lower corner freq of around 50 or 60 Hz.

Am I right in thinking I just have to increase the value of the 22n cap across the cut/boost pot?

Is there anything I need to think about, in order to maintain a constant slope, or anything else?

Thanks 

 

[JohnRoberts]

This is the EQ section of my DDA S-series studio console.

In stock condition it has a LF shelving section with a corner freq of 100Hz, which I would like to drop to a lower corner freq of around 50 or 60 Hz.

Am I right in thinking I just have to increase the value of the 22n cap across the cut/boost pot?

Is there anything I need to think about, in order to maintain a constant slope, or anything else?

Thanks 

The 22nF sets when the boost/cut stops.

100N sets when it starts.

JR

 

[MagnetoSound]

Thanks JR!  So I need to increase both of them to maintain a similar slope?

Is it as simple as increasing both by the same proportion, say a similar percentage, eg 22n > 33n, 100n > 150n?

The 22nF sets when the boost/cut stops.

100N sets when it starts.

JR

 

[L´Andratté]

The 22nF sets when the boost/cut stops.

100N sets when it starts.

JR

Meaning it actually is bell-shaped?
Low shelf below:

 

[MagnetoSound]

Manual says shelving! It's a 4-band EQ though, so I'm not sure simple Baxandall theory applies ...?

A shelf has two corners, after all.

 

[JohnRoberts]

Thanks JR!  So I need to increase both of them to maintain a similar slope?

Is it as simple as increasing both by the same proportion, say a similar percentage, eg 22n > 33n, 100n > 150n?

I don't know, what do you mean by corner frequency?    ;D

The 22N shunts across the pot stopping all boost/cut at rising frequency ... making it bigger will stop boost/cut sooner.

The 100N in series with the wiper disconnects the pot below a frequency, so making it bigger shifts that starting point lower.

Seems easy enough to experiment  with values.

Back in the day I would add an extra cap in series with the input leg 6.8K sized to add a HPF pole that would shift the pole higher when more bass boost was commanded  (shelving bass boost can easily become too much of a good thing).

I'm sure there are equations floating around ... but I'm old and lazy.

JR

 

[JohnRoberts]

Meaning it actually is bell-shaped?
Low shelf below:

Mid band EQ is bell shaped but shelves by definition are not...

Having huge gain at DC is never desirable.

JR

 

[JohnRoberts]

Manual says shelving! It's a 4-band EQ though, so I'm not sure simple Baxandall theory applies ...?

A shelf has two corners, after all.

OK disregard my comments then....  8)

JR

 

[L´Andratté]

Having huge gain at DC is never desirable.
JR

I see.

OK disregard my comments then....  8)
JR

Nope.

 

[L´Andratté]

Back in the day I would add an extra cap in series with the input leg 6.8K sized to add a HPF pole that would shift the pole higher when more bass boost was commanded  (shelving bass boost can easily become too much of a good thing).
JR

Very interesting.

 

[MagnetoSound]

OK disregard my comments then....  8)

JR

Now, now  ... thanks again, I have learned a lot today.

 

[MagnetoSound]

Back in the day I would add an extra cap in series with the input leg 6.8K sized to add a HPF pole that would shift the pole higher when more bass boost was commanded  (shelving bass boost can easily become too much of a good thing).
JR   

Interesting indeed. I feel sure I have experienced this effect more than once. Is it a commonly applied 'trick of the trade'?

 

[Khron]

One word: LTspice

Yes, there's a bit of a learning-curve, but once you get a hang of the basics, you can easily whip up simulations of simple stuff like this, and experiment with component values

 

[MagnetoSound]

Yeah, I guess it's overdue.

One word: LTspice

Yes, there's a bit of a learning-curve, but once you get a hang of the basics, you can easily whip up simulations of simple stuff like this, and experiment with component values

 

[JohnRoberts]

Interesting indeed. I feel sure I have experienced this effect more than once. Is it a commonly applied 'trick of the trade'?

It was one of my "tricks". I don't think I've seen other examples, but every day the ancients are stealing our best ideas and I used this decades ago at Peavey where schematics were published (at least shared with repair centers).

When designing for the fixed instal market where everything feeds through output transformers to step up for constant voltage distribution then down again at speakers, excessive bass that saturates magnetics is very undesirable. I've already shared this before but one of my patents involves a simple diode clipper incorporated into a Baxandall bass boost/cut circuit to clip just the bass fraction of the signal (US05509080 Roberts). This allowed users to boost the bass as much as they wanted while keeping output bass levels in check. It sounded surprisingly good for a diode clipper, because it only clipped the bass when too large, allowing the HF content to pass through cleanly. 

JR

PS: An actual trick of the trade that I did not use, was undersizing the capacitor in series with the gain leg in a mic preamp. This would deliver full bandwidth at modest gain for the data sheet 20-20k spec, but at high gain the LF content would fall off, attenuating the 1/f noise and other LF sonic markers for a noisy preamp. Coincident with using this cheat, that same company spent millions advertising how quiet their preamps were.    Since there was only one company doing that much advertising, you should be able to guess who they were. 

 

[abbey road d enfer]

Meaning it actually is bell-shaped?

Yes. The additional 100nF capacitor adds a pole that's actually very close (ca. 65Hz) to that created by the variable branch (ca.165Hz); nominal 100Hz sits at the geometric mean.
Actually this topology was commonly used in HiFi systems for midrange EQ. The slope is very mild, which makes it unsuitable for surgical EQ, but adequate for tone massaging.

EDIT: tried to post a pic of simulated graph, but server is full! I'll post later.

 

[ruffrecords]

EDIT: tried to post a pic of simulated graph, but server is full! I'll post later.

I just tried to post a pic and goat a "server upload folder is full" message. Is that what you got?

Cheers

Ian

 

[abbey road d enfer]

I just tried to post a pic and goat a "server upload folder is full" message. Is that what you got?

Cheers

Ian

Yes, that's what I got; I figured out it would be solved soon, which it is now.

 

[MagnetoSound]

Yes. The additional 100nF capacitor adds a pole that's actually very close (ca. 65Hz) to that created by the variable branch (ca.165Hz); nominal 100Hz sits at the geometric mean.
Actually this topology was commonly used in HiFi systems for midrange EQ. The slope is very mild, which makes it unsuitable for surgical EQ, but adequate for tone massaging.

Thanks for the graph, Abbey. This explains why the LF isn't doing what I want it to at all!

I need an LF control that will be active in the lower octaves, at least down to 60Hz, with the same amount of boost or cut. Should I bypass the 100nF? Or should I just increase the value somewhat so as to extend the range but still cut off before DC? I think I should increase the bridging cap as well so that it avoids boosting the upper bass range too much.

Truthfully, I haven't had time to do anything with this yet. It might still be a couple of weeks before I can try anything out.

 

[JohnRoberts]

Thanks for the graph, Abbey. This explains why the LF isn't doing what I want it to at all!

I need an LF control that will be active in the lower octaves, at least down to 60Hz, with the same amount of boost or cut. Should I bypass the 100nF? Or should I just increase the value somewhat so as to extend the range but still cut off before DC?

Truthfully, I haven't had time to do anything with this yet. It might still be a couple of weeks before I can try anything out.

I remain apprehensive about boost at DC so do not short the cap... scaling it larger seems appropriate but I would  still like to see boost taper off at 20Hz and below.

JR