The People's Equalizer

[ruffrecords]

When guys order PCBs up here, like for the Bo Hansen DI or these other projects, how many boards do they usually order?

My own experience is mostly a couple.

Cheers

Ian

 

[Ricardus]

My own experience is mostly a couple.

Cheers

Ian

Nonono.

I mean the people who stock them.

 

[Ricardus]

My thoughts on this EQ and other projects like this, is I would definitely buy some if it fit my needs. A stereo EQ with one set of controls is the kind of thing I would probably only build one of, TWO MAX.

But individual "channel" EQ units I would build many many. Like 8 or more (in time).

When I discovered the Bo Hansen DI I initially ordered two boards. I loved the sound so much my goal was then to build 2 with every supported (on the board) transformer flavor, which would have been a total of 6. Which I did. Then I became so obsessed with making another pair with Carnhill traffos, I did that too. That made 8.

I'm looking for a nice channel EQ option, and I am hoping to find it up here one of these days.

 

[ruffrecords]

Nonono.

I mean the people who stock them.

Do you mean how many boards to sellers usually order from the PCB manufacturers?

Cheers

Ian

 

[Ricardus]

Do you mean how many boards to sellers usually order from the PCB manufacturers?

Cheers

Ian

Yes. To get the best prices and such.

 

[ruffrecords]

Yes. To get the best prices and such.

Depends on the size of the board and the expected demand. Most boards I get in batches of just 10 (passive EQ and tube mic pres). The only board I consistently sell more of is the little DB25. I now order a batch of 50 per year.

Cheers

Ian

 

[cpsmusic]

I'm interested in this too although more from a user perspective as opposed to a designer's.

The more I though about the "Volk EQ" the more I thought that the best place to look for inspiration would be a well-thought-out channel strip EQ.

Cheers!

 

[Ricardus]

I'm interested in this too although more from a user perspective as opposed to a designer's.

The more I though about the "Volk EQ" the more I thought that the best place to look for inspiration would be a well-thought-out channel strip EQ.

Cheers!

Yes, exactly.

 

[fragletrollet]

I love the idea. Personally I'd love more diy stereo eq's available, preferable with switched frequencies with resistors on the main or aux pcb. I can sacrifice customizability for less wiring/ease of build, as (imo) there's a lot of more open ended pcb's that allow more customizable setups. This way you could make a standardized frontpanel layout that the pcb slips into.

Hp filters a must, low pass is also nice so I'd vote for both. Baxandall shelves are also great, and two midrange (low mid and high mid, but even better with the midrange bells can overlap frequencies, but not a must) would make it a powerhouse.

I've been eyeing the jlm freq500, that's the closest thing I've found to your idea (but mono). I work with a lot of stereo sources so a stereo solution would be great.

I also love high density modules, as space is scarce, and I want to have as many processors as close as possible to my mixing position (reaching for processors in separate racks furter in the back/side of the studio makes it hard to trust your settings, as you have to go back and forth from the eq and the monitoring sweets pot.

Ssl had a stereo version of one oftheir xrack eq modules I believe; I might be unrealistic but to fit something like this into the api500 module width would be spectacular. But a double wide module would maybe more feasible (especially if using transformer based i/o).

Can't really help with anything more than a thumbs up and hoping to see some progress on the project! Will be following with great interest.

 

[OneRoomStudio]

Ok, it seems like there might actually be some interest here. I mocked up a spice model, and it looks promising (see attached). This is assuming it was fully discrete,  built with 2520s, and had transformer i/o (10k/10K in, 600/600 out). It would be easy to add options for THAT IC i/o and IC opamps though, which would make this a very cheap build.

The first section is the Low-Mid semi-parametric band (260Hz - 2k8Hz variable, +/- 12dB), the second is the High-Mid semi-parametric band (718Hz - 7k8Hz variable, +/- 12dB)  . The third section is the Bax shelves (I plan to add switches to select different caps for different roll-over points, high and low), and the fourth section is the HP/LP filters, with a switch on the LP filter to select 6dB/oct or 12dB/oct. The HP is fixed at 12dB/oct. The final stage is for makeup and output drive (would be altered for THAT corp line driver output).

I didn't include all the switching (each section should be able to be bypassed and frequencies should be able to be selected for the Bax shelves), but it's pretty solid. With everything in and set to "0," it's within +/- 0.5dB from 60Hz to 12K. If you bypassed the filters it would be basically flat.

A few things:

For maximum flexibility, I think it makes the most sense to make any PCBs mono. If you want stereo, buy two and find the appropriate 2-gang/4-gang pots or switches. It might also make sense to leave pots off-board so that people can use switches if they want, or whatever footprint pots they want. That would make the wiring a little more annoying though. As always, there will have to be some kind of balance between simplicity and flexibility.

I doubt 500 series is in the cards - too many controls for the front panel.

Thoughts?

 

[Ricardus]

For maximum flexibility, I think it makes the most sense to make any PCBs mono. If you want stereo, buy two and find the appropriate 2-gang/4-gang pots or switches. It might also make sense to leave pots off-board so that people can use switches if they want, or whatever footprint pots they want. That would make the wiring a little more annoying though. As always, there will have to be some kind of balance between simplicity and flexibility.

I doubt 500 series is in the cards - too many controls for the front panel.

Thoughts?

Mono PCBs sounds great to me.

As for the rest, I wouldn't mind either way. Pots/switches mounted on the PCB or not.

The only suggestion I would make would be to make it as future-proof as possible. Choosing components that hopefully won't become obsolete and unobtanium in a short time. (so if we could make it entirely out of 555 timers that would be awesome)

I know that's a difficult guessing game, but just a thought.

 

[OneRoomStudio]

The only suggestion I would make would be to make it as future-proof as possible. Choosing components that hopefully won't become obsolete and unobtanium in a short time. (so if we could make it entirely out of 555 timers that would be awesome)

I know that's a difficult guessing game, but just a thought.

The goal is to design it so you can use almost any standard transistors/opamps. The real issue is that by making it through-hole, it's already using obsolete parts, and making it surface-mount wouldn't be very DIY friendly. It's getting harder and harder to find through-hole transistors, but there are still enough out there that folks should be able to find them. I'm a fan of 2N5457 FETs, but J201's or whatever would be fine. The output pair (I used ZTX450/ZTX550, which are still available) is just a normal complimentary pair and you can use any variant you want, you just have to be careful to put the right pins in the right holes. The plan is to include footprints for standard DIP-8 opamps too, so that should be as future proof as it can be.

 

[fragletrollet]

A few things:

For maximum flexibility, I think it makes the most sense to make any PCBs mono. If you want stereo, buy two and find the appropriate 2-gang/4-gang pots or switches. It might also make sense to leave pots off-board so that people can use switches if they want, or whatever footprint pots they want. That would make the wiring a little more annoying though. As always, there will have to be some kind of balance between simplicity and flexibility.

I doubt 500 series is in the cards - too many controls for the front panel.

Thoughts?

Nothing is impossible...  8)

https://www.thomann.de/gb/ssl_xrack_stereo_eq_modul.htm

Sure, no DOA's or TX's, but stereo 4-band in something comparable to to 500 series (width). 

 

[OneRoomStudio]

Nothing is impossible...  8)

https://www.thomann.de/gb/ssl_xrack_stereo_eq_modul.htm

Sure, no DOA's or TX's, but stereo 4-band in something comparable to to 500 series (width).

Haha, sure. If you want this to be surface-mount and cost 1,000, then yes (those multi-deck pots are pricey).  The X-rack format is also 9 cm^2 larger than the 500 series, so all those pots and switches would be pretty squished together.

 

[fragletrollet]

Haha yeah I`m probably being unresonable...  but there sure is a lack of pure stereo eq's in the diy market. But ofcourse mono boards are more customizable than a strictly stereo board. I dont really mind smt if it can drastically reduce the size of the module, because space is more of an issue where I live than the availability of gear  8)

Looking forward to what this manifests in, good luck and all the best! Will follow the progress with great interest 

 

[abbey road d enfer]

You can't have "stereo parametric EQ" and "future-proof" in the same sentence. 4-stack frequency potentiometers are always custom parts.
Wien-bridge EQ's request RevLog frequency pots, that are also hard to get.
For this reason I favour the SVF EQ/filters that uses law-steered linear taper pots. More than one opamp per stage though.

 

[OneRoomStudio]

You can't have "stereo parametric EQ" and "future-proof" in the same sentence. 4-stack frequency potentiometers are always custom parts.
Wien-bridge EQ's request RevLog frequency pots, that are also hard to get.
For this reason I favour the SVF EQ/filters that uses law-steered linear taper pots. More than one opamp per stage though.

This project is best described as "semi-parametric" at best, since the Q is not adjustable.

As for the pots, I'll quote Rod Elliot:
"Resistance Vs. Frequency: Ideally, you'd use an antilog pot but unless the taper is really a reverse logarithmic type (very rare and generally considered unobtainable), you are better off with the linear taper.  Although it gets a little touchy as you get close to minimum resistance (maximum frequency), linear pots generally have better tracking than log or reverse log, and are far more predictable."

He goes on to present a table of linear pot rotation vs frequency in a Wien Bridge circuit. You can see it here: https://sound-au.com/project150.htm

In practice, linear does the job just fine, and the pots (even multi-deck) are widely available.

 

[abbey road d enfer]

This project is best described as "semi-parametric" at best, since the Q is not adjustable.

Fully parametric or semi parametric, you still need dual pots for frequency.

As for the pots, I'll quote Rod Elliot:
"Resistance Vs. Frequency: Ideally, you'd use an antilog pot but unless the taper is really a reverse logarithmic type (very rare and generally considered unobtainable), you are better off with the linear taper.  Although it gets a little touchy as you get close to minimum resistance (maximum frequency), linear pots generally have better tracking than log or reverse log, and are far more predictable."

He goes on to present a table of linear pot rotation vs frequency in a Wien Bridge circuit. You can see it here: https://sound-au.com/project150.htm

Do you really have hands on experience of that? I have and can assure you it's not pleasant, unless you restrict the sweep range to 1:3 or 1:4, which requires switching capacitors for acceptable range.

In practice, linear does the job just fine,

This is very debatable. The same site shows that for a given rotation of about 15° the variation is +10% on the CCW side and +100% on the CW side: that's for a 1:10 sweep range. Most non-switchable midrange EQ's tend to favour a range of 1:20 or more.

 

[john12ax7]

Linear pots will cause large jumps with small movement.  The referenced article shows 2:1 frequency jump going 90% to 100%. Very suboptimal imo.

Dual reverse log are somewhat available,  and you can get semi custom at around 250 pcs,  not too bad.

 

[OneRoomStudio]

I'm looking into some custom pots now. If I end up going down that road, it might be worth getting custom dual concentric pots, which would make this viable in 500 series format.

If that were the case though, I'd need to make a choice - either limit it to one variable mid band (so it would be HP, Low Shelf, Mid band, High Shelf, LP), and have room for an output volume control, or give it two mid bands, but make the gain fixed at unity while flat. There's only so much room on a 500 front panel.