1-band parametric EQ

[RobA]

... I'm really excited to see how much space all this will take.

Well, I'm going to do a bit more playing with the layout, but right now, it's looking like 35mm by 45mm as a lower limit. I might actually be able to save a bit more space by deleting all the unneeded pot connections, but it isn't likely to get much smaller than that using PTH parts. I could probably get smaller by going to SMD resistors and op-amp and using a mix of SMD and PTH caps. But even there, the number of off board connections is a limiting factor. One good thing with going to some SMD parts is that it opens up the use of the TS924 which is only available easily in SMD now.

What do you think? How open are you to using SMD parts? Is the size OK if I can get it to around 40mm by 35mm using PTH parts?

[kothoma]

I think this size is quite acceptable. Compare it to the number of pots that could be attached to it.
And at least two of them (frequency, cut/boost) would always be present or one would choose a simpler circuit.

I'm not a big fan of SMDs. As yet. SMD JFETs are somewhat unavoidable but soldering them makes me sweat. Maybe I could work my nerve up to ICs. But those little resistors are really intimidating.

How about SO14 pads in parallel within DIP14 pads?

[RobA]

I think this size is quite acceptable. Compare it to the number of pots that could be attached to it.
And at least two of them (frequency, cut/boost) would always be present or one would choose a simpler circuit.

I'm not a big fan of SMDs. As yet. SMD JFETs are somewhat unavoidable but soldering them makes me sweat. Maybe I could work my nerve up to ICs. But those little resistors are really intimidating.

How about SO14 pads in parallel within DIP14 pads?

OK, I'll try to stick with PTH and maybe add the embedded SO-14 into the DIP-14 if I can make it work and not cause the size to increase.

The SO sizes really aren't too bad once you get used to them. I've been avoiding the really tiny resistors and caps. I use the 1206 size which then also works with the 0204 mini-MELF resistors which I prefer in the audio paths. The size of those isn't too bad at all. But you do have to develop a technique with the MELF's or they tend to roll away and disappear.

[kothoma]

Hm, looking at this



it's probably not possible to get it inside. At least it would be a tight fit.

[RobA]

Hm, looking at this



it's probably not possible to get it inside. At least it would be a tight fit.

No, you can't get the SOIC-14 inside the DIP-14. In the end, I did two different layouts. The first is all PTH and the second uses SMD IC, resistors and some caps. I left the electrolytic power cap and the actual filter caps as PTH though.

But, before I place the order for the test boards at OSH, it would be nice if someone checked over my schematic. There are some weird things that can happen with the high shelf/PEQ circuit. As was talked about above, you can make it as either a high shelf or a PEQ. There are two jumpers that select which one is going to be used. Then the rest of it is set by which parts are filled with a jumper and which with parts. There's blue text on the schematic that describes what needs to be done for each.

Here's the schematic,
http://rock.it-frog.com/Downloads/Docs/PEQ_Schematic.pdf

[kothoma]

Here's the schematic,
http://rock.it-frog.com/Downloads/Docs/PEQ_Schematic.pdf

I looked over it and compared it to the sources discussed and to the Spice sim. Looks good to me. I see you've omitted the pulldown resistors.

Hm, I wonder if the second band could be done with less jumpers (and space).
There's this sequence  ---C11-+-R10---RV6--- for the PEG with the midpoint (-+-) going to the non-inverting input (pin 12).
For the shelf this sequence is simply reversed with the same midpoint (and probably the pot has to be wired the other way).
So if the geometry of the layout permits you only have to jumper (swap) the start and end points.
Just an idea.

Edit. Or you could replace this sequence with the following: ---Rx---Cx-+-Ry---Cy--- with the C pads doubling as pot pads and polulate them like this:
PEG: Rx = jumper, Cx = C11, Ry = R10, Cy = RV6.
Shelf: Rx = R10, Cx = RV6, Ry = jumper, Cy = C12.  (plus C10 = jumper, of course)
Just a crazy idea.

[RobA]

Thanks for the input on those. I'll check it out and see about altering the layout and how much it would get in return.

[RobA]

I took a look at the changes you suggested. The first one would be good but it doesn't get any reduction in size and I kinda prefer being able to label the jumpers to easily identify which configuration they go with.

Your second idea is ingenious, but it doesn't mesh well with my OCDish need to have the pots lined across the top of the PCB. In other words, the two together (pots acting as caps, and sitting on the top row of the layout) would actually end up making it bigger.

I've got the two layouts in at OSHPark. They end up being this big:
PTH 43mm x 36mm
SMD 43mm x 24mm

The width is set by putting the pots and toggles across the top row. At this point, I'll probably order the PTH version this week and do some testing on it to see how well it works. I am going to do a little looking into other parametric EQ topologies first and see if it might be better to go another direction.

[kothoma]

Thanks for the update. I guess you're right, with all the different options clear labels and easy access to the pads are preferable.

The layout sizes seem very acceptable to me. I'm curious to hear what your testing shows.

Other topologies: MFB bandpass, inverting (Wien bridge) bandpass, and SVF come to mind. Any others?

[RobA]

Thanks for the update. I guess you're right, with all the different options clear labels and easy access to the pads are preferable.

The layout sizes seem very acceptable to me. I'm curious to hear what your testing shows.

Other topologies: MFB bandpass, inverting (Wien bridge) bandpass, and SVF come to mind. Any others?

I was going to start looking at the MFB type first. I don't know of others yet, but I need to do some research. What I'd like to find is something that would play well with either an active tilt EQ or a Baxandall.

[kothoma]

You're certainly aware of the following fine pages:

http://www.electronics-tutorials.ws/filter/filter_7.html
http://sound.westhost.com/articles/active-filters.htm
http://www.rane.com/note122.html

http://www.sound.westhost.com/project150.htm (Wien Bridge Based Parametric Equaliser)
http://sound.westhost.com/articles/state-variable.htm

[RobA]

You're certainly aware of the following fine pages:

http://www.electronics-tutorials.ws/filter/filter_7.html
http://sound.westhost.com/articles/active-filters.htm
http://www.rane.com/note122.html

http://www.sound.westhost.com/project150.htm (Wien Bridge Based Parametric Equaliser)
http://sound.westhost.com/articles/state-variable.htm

Thanks! I have seen most of them, but you saved me tons of time by making it so I won't have to search through my disastrous mess of bookmarks to try and find them again .

[RobA]

I keep forgetting to mention that I've verified the PTH version of the two band parametric EQ. I've shared the project on OSHPark now. I'll do a build doc for it soon, but the link is here,

https://oshpark.com/shared_projects/3W6nzzYI

This project corresponds to the schematic that I linked to above. I was playing around with it a couple of days ago and it's actually pretty usable as a pedal by itself if you populate all the knobs and switches.

[kothoma]

Many thanks for sharing your 2-band parametric EQ!

I think it turned out great and very much appreciate your hard work that made this happen.

[dropanchor812]

I missed the development, but I've been looking for something like this.  Thanks for sharing!