Hi Frequency Limiters?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

SWAN808

Well-known member
Joined
Sep 13, 2009
Messages
91
Are there any DIY projects/schematics out there for high frequency limiters perhaps like the ones used in the Empirical Labs products (FATSO/DERRESSER)...or even the mastering type HF limiters...
 
The main difference between a DBX902 type deesser and the good high frequency limiters for disk cutting are that the mastering HF limiters look at the rate of change  between  two signals and don’t have a fixed threshold.

When I say good HF limiters I am referring to the Neumann BSB74 and the Maselec HF Limiters.
 
gyraf said:
My favorite for cutting-type is the Ortofon STL631

In a roundabout way, it works in a similar way to my G21..

Jakob E.

The STL631 is interesting for the passive signal path with active side chain. It doesn’t work particularly well as an HFL.

The later STL 732 is better. It has conventional compressor controls. I like the one knob approach of Neumann and Maselec better.
 
The Dangerous  Music compressor has a mode that changes the threshold from a hard level to rate of change threshold. There is also a high frequency boost shelf in the side chain. I haven’t tried it but I imagine  using the compressor with both of those modes activated would make a good HF limiter.
 
Thanks I was looking to build something stereo for processing harsh ITB signals/tracks, so maybe a deesser would not be as good as a HF limiter...I have found a schematic for the Ortofon 732 but I will have to study it to see if it's something that could be replicated...any thoughts?
 

Attachments

  • Screenshot 2019-09-23 at 01.15.17.png
    Screenshot 2019-09-23 at 01.15.17.png
    947.1 KB · Views: 117
SWAN808 said:
Thanks I was looking to build something stereo for processing harsh ITB signals/tracks, so maybe a deesser would not be as good as a HF limiter...I have found a schematic for the Ortofon 732 but I will have to study it to see if it's something that could be replicated...any thoughts?

Deesser and HF Limiter are kind of used interchangeably depending on application.

The  732 needs an RIAA Equalized signal at the sidechain input. That would have to be added to the schematic.
 
SWAN808 said:
Thanks I was looking to build something stereo for processing harsh ITB signals/tracks, so maybe a deesser would not be as good as a HF limiter...I have found a schematic for the Ortofon 732 but I will have to study it to see if it's something that could be replicated...any thoughts?
I would think the most difficult part is to get hold of the triple FET (H8  MM555); I would think it's unobtainium. You may use discrete FET's but you'll need to match them. Unless you can find their characteristics, you'll have to experiment with different types. It's not an easy task.

An HF limiter typically limits the HF content of a wide-band signal by introducing a variable LPF.
A de-esser can operate in one of the following modes:
  • A limiter that limits the whole signal but has pre-emphasis in the side-chain. Usually described as wide-band mode.
  • A filter divides the signal in its LF and HF parts; limiting occurs only on the HF part, then the signals are recombined. Usually described as split-mode.
  • A variable LPF is applied to the signal; similar to HF limiter
  • A variable depth tunable "eliminator" filter (cut-only bell filter).
There may be other types I'm not aware of (or forgotten).


 
abbey road d enfer said:
I would think the most difficult part is to get hold of the triple FET (H8  MM555); I would think it's unobtainium. You may use discrete FET's but you'll need to match them. Unless you can find their characteristics, you'll have to experiment with different types. It's not an easy task.

An HF limiter typically limits the HF content of a wide-band signal by introducing a variable LPF.
A de-esser can operate in one of the following modes:
  • A limiter that limits the whole signal but has pre-emphasis in the side-chain. Usually described as wide-band mode.
  • A filter divides the signal in its LF and HF parts; limiting occurs only on the HF part, then the signals are recombined. Usually described as split-mode.
  • A variable LPF is applied to the signal; similar to HF limiter
  • A variable depth tunable "eliminator" filter (cut-only bell filter).
There may be other types I'm not aware of (or forgotten).

thanks...the Ortofon circuit may be a bit over ambitious...perhaps I should try some de-esser options you mention...perhaps I could set up some simple tests using a THAT chip...
 
SWAN808 said:
thanks...the Ortofon circuit may be a bit over ambitious...perhaps I should try some de-esser options you mention...perhaps I could set up some simple tests using a THAT chip...
I think you need to define your needs. Is it for disc-cutting or general sibilance reduction?
 
abbey road d enfer said:
I think you need to define your needs. Is it for disc-cutting or general sibilance reduction?

I want to experiment with a circuit to soften high freq, similar to the FATSO warmth circuit...or in plugin format the SOOTHE plugin...as a stereo processor to work in a hardware processing chain alongside the DAW...

 
gyraf said:
Like this sort of behavior?

/Jakob E.

While this behaviour might be appropriate in some cases, surely you can bypass it yes?

Edit: or is it a mistaken name label and not really the Fearn VT7?
 
SWAN808 said:
I want to experiment with a circuit to soften high freq, similar to the FATSO warmth circuit...or in plugin format the SOOTHE plugin...as a stereo processor to work in a hardware processing chain alongside the DAW...
I don't know how Warmth in FATSO works, maybe some slew-rate limiting. Never used it but the various videos have not convinced me of its value.
SOOTHE is a very nice tool, unfortunately there's no way you can DIY such a thing. DSP and FFT are the words.
 
abbey road d enfer said:
An HF limiter typically limits the HF content of a wide-band signal by introducing a variable LPF.
This (sliding LPF) was also a popular single ended noise reduction approach. Side chain generally clamped down on HF response when valid HF signal was lacking, opening up when HF signal is present.  Even more effective for noise reduction when combined with downward expander.
A de-esser can operate in one of the following modes:
  • A limiter that limits the whole signal but has pre-emphasis in the side-chain. Usually described as wide-band mode.
This is the only de-esser approach I used because I generally included de-essing as an additional feature inside a full range compressor/limiter. This way the de-esser function is only affecting the side chain circuitry.
  • A filter divides the signal in its LF and HF parts; limiting occurs only on the HF part, then the signals are recombined. Usually described as split-mode.
Generally difficult to put Humpty back together perfectly if you only limit one bandpass of the divided up signals.
  • A variable LPF is applied to the signal; similar to HF limiter
  • A variable depth tunable "eliminator" filter (cut-only bell filter).
There may be other types I'm not aware of (or forgotten).
Many ways to skin any cat...

JR
 
JohnRoberts said:
Generally difficult to put Humpty back together perfectly if you only limit one bandpass of the divided up signals. Many ways to skin any cat...

The Neumann BSB74 splits the signal after it is delayed with a bank of inductors. Then the bands are recombined after HF limiting. I think Jakob once referred to the BSB74 as S&M engineering. An apt description.
 
JohnRoberts said:
"A limiter that limits the whole signal but has pre-emphasis in the side-chain. Usually described as wide-band mode."
This is the only de-esser approach I used because I generally included de-essing as an additional feature inside a full range compressor/limiter. This way the de-esser function is only affecting the side chain circuitry.
Understandable in view of integration in an already crowded console channel, but unfortunately the most obtrusive sound-wise.
I don't know any mixers that have in-channel gates with split-mode.

"A filter divides the signal in its LF and HF parts; limiting occurs only on the HF part, then the signals are recombined. Usually described as split-mode."
Generally difficult to put Humpty back together perfectly if you only limit one bandpass of the divided up signals. Many ways to skin any cat...
The three most common approaches are:
  • Using 1st-order filters
  • Using complementary (substractive) filters
  • Using higher order filters combined with all-pass filters and accept that the resulting output is not minimum-phase
The two first options are not discriminating enough and push the user to make compromises.
The last one is generally accepted as satisfactory.

I'm trying to figure out what the Neumann BSB74 does. I suspect the cross-over is 1st-order and the inductor-based filters are delays that are not related to the recombination, but instead are there to delay the signal to align it with the time-response of the detector. Reading the schemo is a PITN; I think I'll print it on three separate pages and assemble them with adhesive tape, early 20th century style...
 
abbey road d enfer said:
  I think I'll print it on three separate pages and assemble them with adhesive tape, early 20th century style...

The print is a four page fold out IIRC. The inductors are a separate PCB that forms the other side of the module. It may be a separate print.
 

Latest posts

Back
Top