The People's Equalizer

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

OneRoomStudios

Well-known member
GDIY Supporter
Joined
Jul 8, 2004
Messages
559
Location
Minneapolis, MN, USA
**If this should be in Drawing Board, feel free to move**

I've been thinking about EQ's a lot lately. There aren't a lot of DIY projects available for EQs, and the ones that are out there tend to fall into one of three categories: passive LC with tube/DOA make up (Pultec-style), active inductive (Neve/API style), or gyrator. There are a few other interesting projects here and there too, but they often suffer from the same issues as the others I mentioned - they're usually hard to get a hold of, discontinued, expensive to build, or not something I'd want to bother with.

I've been wishing for an EQ project that was easy to tune to my tastes (not dependent on inductors), could be built with either cheap parts or fancy parts as budget permits, and most of all, is both great-sounding and highly useful. I've been tossing around ideas for a while and have come up with what I think is a good concept, but I'd like to open the project up to others here to solicit feedback, and also to get some help with the PCB layout so we can have a board that accommodates lots of options.

The idea is this - an EQ that could be built with either That Corp line receivers /drivers, and IC opamps, for a cheap/quick build, or with transformers, discrete buffers, and DOAs for more 'color.' You can think of it as an EQ that could grow with you and your budget. The topology I'm thinking consists of active HP/LP filters, Baxandall shelving filters, and a Wien Bridge mid-band. This gives you a pretty flexible EQ, with a minimum of components and hard-to-get parts.

Here are some details as I'm thinking of them:

I'm partial to the HP/LP filters used in the Urei 539 (http://www.jblproservice.com/pdf/Vintage%20JBL-UREI%20Electronics/UREI-539.pdf). Both of them are wrapped around a single opamp (not including the buffer driving them), and I like the 1 or 2 pole switch on the LP - very useful. With 150nF caps and a dual 100K pot, the HPF would range from around 10Hz to around 320Hz, which feels pretty good to me. With 1nF caps and a dual 20K pot, the LPF would range from around 25KHz (closer to 30KHz in single-pole mode) down to around 6KHz - also pretty nice.)

The Baxandall shelves can also both be wrapped around a single opamp, like Rod Elliot does here: https://sound-au.com/dwopa2.htm#s7. It's easy enough to calculate a handful of different cap values for each band to provide 4-6 frequency selections (for example, the high shelf could use a 10K pot, 1K5 resistors, and 1n2, 1n8, 2n7, and 3n9 caps to provide 16KHz, 10KHz, 7KHz, and 5KHz shelves).

The Wien Bridge mid-band is a simple pseudo-parametric  choice for the mid frequencies that only needs a single opamp. You could even cascade more than one if you wanted more bands. Rod Elliot has another nice example here: https://sound-au.com/articles/eq.htm#s7. With 3n3 caps and 100K and 10K pots, you'd get +/- 12dB sweepable between 430Hz and 4.8KHz.

Building something like this with integrated line receivers/drivers and IC opamps would be very inexpensive, and the ability to build it with transformers, DOA's (3 needed), and discrete buffers would make it a relatively simple-to-build discrete EQ.

I mocked up an example front panel for those who need a visual.
b7bbCmT


What would make this especially enticing is if a PCB could be made to have footprints for both the monolithic ICs, and DOAs/various transformers - a universal/upgradable board.

Thoughts?
 

Attachments

  • EQ-Layout.png
    EQ-Layout.png
    13.6 KB · Views: 230
shabtek said:
You may be describing “perf board “

Ha!

There wouldn't actually be that much to making it adaptable. There are already lots of kits that allow for either discrete or monolithic opamp (JLM Baby Animal is a good example), and there are kits out there that allow for either discrete/transformer output or balanced line driver (see DIYRE EQP-5). The rest of the circuit would be worth having on a PCB for easy assembly and modification. I just don't have the footprints/layouts for this kind of thing handy, so was hoping someone could help out with that part.
 
OneRoomStudios said:
Ha!

There wouldn't actually be that much to making it adaptable. There are already lots of kits that allow for either discrete or monolithic opamp (JLM Baby Animal is a good example), and there are kits out there that allow for either discrete/transformer output or balanced line driver (see DIYRE EQP-5). The rest of the circuit would be worth having on a PCB for easy assembly and modification. I just don't have the footprints/layouts for this kind of thing handy, so was hoping someone could help out with that part.
I don't understand what it is you're stumbling on. Do you have a PCB software? They all offer the possibility to create your own devices and footprints.
Now you can't provide footprints for all types of xfmrs, but you can choose a couple of candidates.
 
abbey road d enfer said:
I don't understand what it is you're stumbling on. Do you have a PCB software? They all offer the possibility to create your own devices and footprints.
Now you can't provide footprints for all types of xfmrs, but you can choose a couple of candidates.

I have and use Eagle CAD, but I figured that some other folks might have the footprints handy (preferably in Eagle format) for DIP 8/DOA and for multiple transformers (as JLM does). Obviously not all transformers would be able to fit, but it'd be nice if it could fit the usual suspects from Cinemag/Jenson/Lundahl, etc. I also thought some folks might have input on the topology or concept.

Is this really that strange of an idea? It seems like this place used to be more open to this kind of collaborative idea in the past. Maybe I'm barking up the wrong tree.
 
OneRoomStudios said:
I have and use Eagle CAD, but I figured that some other folks might have the footprints handy (preferably in Eagle format) for DIP 8/DOA and for multiple transformers (as JLM does). Obviously not all transformers would be able to fit, but it'd be nice if it could fit the usual suspects from Cinemag/Jenson/Lundahl, etc. I also thought some folks might have input on the topology or concept.

Is this really that strange of an idea? It seems like this place used to be more open to this kind of collaborative idea in the past. Maybe I'm barking up the wrong tree.
I also use Eagle, and have grown weary of using third-party models and footprints. I always make my own. Good for the brains and good for the hair (not pulling it).
 
+1 on making your own footprints.  You do it once and then have it for all future designs.

I get the concept you are going for,  the issue that often comes up with the one size fits all approach is that everyone inevitably wants something different.  So you need to figure out how to overcome the too many chefs in the kitchen situation.
 
OneRoomStudios said:
Is this really that strange of an idea? It seems like this place used to be more open to this kind of collaborative idea in the past. Maybe I'm barking up the wrong tree.
Possible; you may get answers you'd prefer addressing your request to the Eagle forum.
 
So I have a friend who is a talented engineer/tech/designer. Some of you might have even heard of him. I'm always telling him about the projects I'm working on, and how I have to order unobtanium parts from surplus houses all over the place (and in other countries), and he appreciates that it keeps my happy and occupied, but his reply one day was something like, "That's not my thing anymore. My goal now is to design great sounding stuff with easily obtainable  components that will likely still be available 20 or 30 years from now."

So I like the idea of this People's EQ. The Volks EQ, if you will.

When I first joined this forum I was surprised there wasn't one of every kind of device, designed through collaboration and crowd sourcing. A GroupDIY EQ, a GroupDIY Preamp, a GroupDIY Compressor, etc....

There's so much talent up here it's amazing. Would love to see something like this.

 
Some time ago I developed a very simple equalizer: shelving low with variable turnover frequency, four overlapping full parametric filter sections and shelving high with variable turnover frequency. All + or - 15 dB. Also a gain control +/- 12 dB in case you gain or lose level due to heavy equalization... All pretty cheap parts, no 'exotic' integrated circuits but 'good old' TL074 etc.
Works perfectly. It can make right what is wrong. (And also wrong what is right... :D)
Baxandall circuits and schematics for parametric sections you can find online.
This thing has proven to be a problem solver!
 

Attachments

  • Equalizers.jpg
    Equalizers.jpg
    75.3 KB · Views: 252
This is what I used for the parametric sections. The only 'difficult' part is the dual reverse-log 100 K potentiometer.
(You could also use a rotary switch with resistors of course, if you prefer repeatability.)
 

Attachments

  • ParamEQ.PNG
    ParamEQ.PNG
    34.8 KB · Views: 235
OneRoomStudios said:
**If this should be in Drawing Board, feel free to move**

I've been thinking about EQ's a lot lately. There aren't a lot of DIY projects available for EQs, and the ones that are out there tend to fall into one of three categories: passive LC with tube/DOA make up (Pultec-style), active inductive (Neve/API style), or gyrator. There are a few other interesting projects here and there too, but they often suffer from the same issues as the others I mentioned - they're usually hard to get a hold of, discontinued, expensive to build, or not something I'd want to bother with.

...SNIP...

What would make this especially enticing is if a PCB could be made to have footprints for both the monolithic ICs, and DOAs/various transformers - a universal/upgradable board.

Thoughts?

Well if you're asking for ideas, for me an ideal channel EQ would be 4 band. Low, low-mid, high-mid, high, and maybe  a high pass.

Top and bottom bands can be switchable between shelving and peaking.
 
My opinion is ....

- there is a need for a stereo unit, one set of controls for both channels

For each channel of the stereo [only] pair, a number of buffered 'filtered sections' in-line (series)  with each other  ...  quality chips are  great for this purpose.

Done with some 'alps' pots for gain, some suitablely modest 'decked rotary' switches for freqs,  and a simple toggle for each band 'in/out' 

..  with an overall relay bypass (for the stereo channels)  and of course, balanced inputs and outputs. One might add a traffo O/P option if one so desires .... 

I like a DRV chip bal output :)  as well as a properly driven traffo output [ both types available at same time]

Further, I like the 'Bax type of eq'  circuits done in the style of 3-band  but  with seperate 'hp and lp' filter sections.  :)

What's not to like? for the People  8)
 
OneRoomStudios said:
.
Is this really that strange of an idea? It seems like this place used to be more open to this kind of collaborative idea in the past. Maybe I'm barking up the wrong tree.
I think it’s a great idea. Don’t give up. Forums are weird; sometimes they behave the exact opposite of what you think they should.

Also the culture here is generally fierce independence, even in terms of sharing, so I think less collaboration occurs because of that.

I’m hopeful someone will step up and want to offer ideas and help for you.

Best,

Mike
 
Thanks to everyone for the ideas and encouraging words! I've been thinking about the design some more, and it seems like the easiest and most universally-appealing design might be:

-Stereo with one set of controls (you can always use just one channel if you need a mono EQ)
-Four bands plus filters set up as:
Hi-Pass, Low Shelf (Baxandall), Low-Mid (Wien Bridge), High-Mid (Wien Bridge), High Shelf (Baxandall), Low Pass
-While it would be nice for the high and low bands to be switchable between peaking and shelving, the combination of shelving and hp/lp filters can result in peaking, and it keeps the circuit simpler (plus, I like the sound of Baxandall shelves)
-This would require 4 opamps per channel (not including line drivers/receivers). 8 opamps for a stereo unit might be a few too many to use DOAs, but the option could be there.
I still think the option to use transformers or IC receivers/drivers would be nice

I'd be happy to start piecing together a schematic if someone were willing to take on the pcb layout (I don't have the time to do both).
 
ruffrecords said:
I created something similar back is 2012. the universal passive EQ. There seemed to be lots of interest, I had some boards made but hardly anyone bought them. I eventually more or less gave away the ones I had left and open source the PCB layout.

https://groupdiy.com/index.php?topic=50484.msg639985#msg639985

Cheers

Ian

Thanks Ian,  that's definitely a cautionary tale.  I didn't know about that project (must have missed it somehow).  That's exactly the reason why I'm not too keen on doing all the PCB layout/ordering myself. I've had to eat the costs of PCBs and front panels before, not to mention the time spent in design. I was hoping there would be enough other interested people to help hoist this project into reality, but there may not be.
 

Latest posts

Back
Top