Passive Mastering EQ

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
16,152
Location
Norfolk - UK
I decided to start a new thread following on from the old  6dB EQ thread in the Drawing Board. This is mainly because the thinking has moved on a lot and I am now well into designing PCBs for it.

As it stands, the design will feature:

1. Up to 10dB boost/cut in 2dB steps
2. Up to 6 bands
3. Up to 10 frequencies per band
4. Mid bands will be RLC bell shaped response
5. Highest and lowest band can be shelving instead of bell shaped
6. Optional variable sharpness
7. 600 ohm operating impedance

The EQ will be mono. If you want stereo you can build a pair. Because all parameters are switched they should track quite well.

I plan to use Don Audio RM8 inductors and may have some specials designed for this project.

I have chosen a 600 ohm operating impedance because it allows the use of lower values of inductance which should avoid any out of band self resonance effects they may be a problem for people using higher sampling rates.

As I mentioned in the last Drawing Board post, there will be three types of PCB

1. A common switch board. This holds two Grayhill  1 pole 12 way switches. One is the boost/cut level control and the other is the frequency select switch. Both are connected to 12 way Molex connectors.

2. EQ board which holds  the LCR required for a particular band. Not sure how many variations of this there will be. We may need one dedicated to bass because it will need a larger inductor than the other (but standardising on Don Audio RM8 cores might avoid this) . A second would use a smaller inductor for low mid and a third might use axial leaded inductors for high mid or high. Any of the boards could be used for shelving high just by shorting out the inductor position. Similarly, the larger inductor types could be used as a shelving low just by shorting out the capacitor positions. This board will connected to the 12 way connector from the frequency select switch. I am toying with the idea of this board piggy backing onto the switch board if only to keep frequency  selective connections as short as possible.

3. Interconnect board.  This has the EQ in and out connectors, the single resistor ladder used by all bands and a bunch of 12 way Molex connectors that connect to each band.

An initial layout of the two switches board is attached. The three blue circles are 3mm clearance hole for piggy backing the EQ board. RQ and the connector Q are options for varying the Q. I don't really think these should be necessary as the Q should be determined principally by the characteristic resistance of the circuit and the inductor values. The Q control simply alters Q by adding series resistance but it also reduces the amount of boost/cut

Cheers

Ian
 

Attachments

  • switchboard.png
    switchboard.png
    11.4 KB · Views: 365
And here is a rough layout of an EQ board. Each of the 10 frequencies has a pair of series capacitors. This makes it much simpler to create odd capacitor values. Each capacitor footprint will accept a 5mm spacing or a 7.5mm spacing capacitor. I just used a VTB9042 inductor to make sure it would fit. In practice this will be replaced by an RM8 core type.

Cheers

Ian
 

Attachments

  • EQboard.png
    EQboard.png
    16.5 KB · Views: 280
nice job Ian,
looks very interesting. i suggest you make more room for bigger caps though. it would be nice if one could use axial types as well
 
fragletrollet said:
Cool! Would it be possible to add a switch for 1db steps as an option to 2 db? I think 2db steps for a mastering eq might be a tad too coarse?

The current design uses 12 way switches which is enough for +-10dB in 2dB steps. Using the same switches you could alter the resistor ladder for 1dB steps but you would only get a range of +-5dB.

Cheers

Ian
 
salomonander said:
nice job Ian,
looks very interesting. i suggest you make more room for bigger caps though. it would be nice if one could use axial types as well

I was pretty certain somebody would ask that one!  I do not want to make the board more expensive for the small number of people who want to use axial  caps. As the design stands there should be enough room to fit a single reasonably sized axial cap in place of the two regular radial types. That is about all I feel justified in providing at this stage.

Cheers

ian
 
Recording Engineer said:
I'm interested in this, but for tracking rather than mastering.

It will work just as well for tracking. I have several EQ designs you can use for tracking but most use regular pots for level setting so repeatability is not a feature. This started of as another tracking EQ but because of the repeatability and the tracking between channels, several people suggested it would be good as a mastering EQ and I have not designed one of those before.

At present, for tracking,  I have in mind each band will consist of  a 3 inch high switch PCB that will fit vertically in a 2U rack.  Piggy backed on it will be the EQ board for that band. Five bands should fit easily in a 19 inch rack.

For mastering I am thinking of a 3U enclosure with left channel on top and right channel below. The 3 inch switch boards would be mounted horizontally. You should again be able to fit five bands across a 19 inch 3U enclosure.

Cheers

Ian

 
ruffrecords said:
Are you suggesting a +-3dB range is sufficient for mastering?

Cheers

Ian

Yes, absolutely.  I master records for a living.

The main low Q EQ I use every day is two bands,

+/- 0.5, 1, 2, 3, 4 dB at 30/50/75/125/250/500/800

and

+/- 0.5, 1, 2, 3, 4 dB at 250/2k/4k/8k/12k/16k/18k/40k

I rarely touch +/- 4dB but often wish I a stop between +/- 1dB.  If I need more than that range we are in real trouble and I have plenty of other options - a Porter and my digital EQ, DMG Equilibrium.

 
ruairioflaherty said:
Yes, absolutely.  I master records for a living.

The main low Q EQ I use every day is two bands,

+/- 0.5, 1, 2, 3, 4 dB at 30/50/75/125/250/500/800

and

+/- 0.5, 1, 2, 3, 4 dB at 250/2k/4k/8k/12k/16k/18k/40k

I rarely touch +/- 4dB but often wish I a stop between +/- 1dB.  If I need more than that range we are in real trouble and I have plenty of other options - a Porter and my digital EQ, DMG Equilibrium.

Excellent. Good to have input from a professional end user.  A +-3dB version would only need a different set of values in the resistor ladder and the insertion loss would also be smaller so you can probably get away without any dedicated gain make up and make it 100% passive. I'll work out the resistor values for this

Cheers

Ian
 
Yes, I agree! For mastering I hardly boost/cut more the 3db. Most of the times it is in the 0 to 1.5/2db range, so for mastering this would be the sweetspot I think.
Looks pretty interesting though, keep going!
 
Looks like the +_3dB version would be better suited for mastering. Ruairi mentioned 'low Q'. Anyone care to put a number to that?

Cheers

Ian
 
ruffrecords said:
I was pretty certain somebody would ask that one!  I do not want to make the board more expensive for the small number of people who want to use axial  caps. As the design stands there should be enough room to fit a single reasonably sized axial cap in place of the two regular radial types. That is about all I feel justified in providing at this stage.

Cheers

ian

hey Ian
i understand. actually i was not thinking of allowing them to sit flat as that would certainly make the pcb huge. but if you only gave a little more space between the cap lines one can use even big axials in vertical position. just an idea. would be nice to use high quality stuff in a mastering one.
 
ruffrecords said:
Looks like the +_3dB version would be better suited for mastering. Ruairi mentioned 'low Q'. Anyone care to put a number to that?

Cheers

Ian

Well, as John and Abbey have pointed out many times this is a tricky topic and coming up with a number that means the same thing to everyone is not easy.

I believe the Q on the unit I use is somewhere around 0.6, and when I use digital EQ I usually start with 0.7. 
 
maybe one could design the switches so that an extra toggle selects between say 3db or 10db boost. dont know if thats possible in terms of circuit... but it would be super handy.
 
salomonander said:
maybe one could design the switches so that an extra toggle selects between say 3db or 10db boost. dont know if thats possible in terms of circuit... but it would be super handy.

Unfortunately that is not practical. There is a master ladder of eleven resistors that define the steps. You would need to switch all eleven to change from 3dB to 10dB boost.  The characteristic resistance that determines the Q of the bell curves would also change so to keep the same Q you would really need to switch all the inductors and capacitors.

Cheers

Ian
 
Hi  Ian,

can you implement that air thing at 40k don't know if it is doable in a passive eq ,just my 2 pence worth , i third that 0.5 increments

Regards

Denis
 
Back
Top