The Passinwind Open Source Preamp

[BQFS]

HPF is a frequency cutoff control for a high-pass filter, Freq is the frequency cutoff control for a low-pass filter, Res is the resonance for the low-pass filter, the switch is (I believe) a midrange frequency selector

Thank you very much!

 

[BQFS]

The 3 way switch gives you three midrange center frequencies and won’t do anything unless you have boost or cut dialed in via the Mid control.

The very first post in this thread shows the basic resonance and frequency functions, although your unit has a wider sweep range than the original design. I’d recommend leaving the Mid control at noon (flat) while you’re coming to grips with the the Res and Freq controls. If you just want to cut highs the Res control needs to be at nine o’clock or lower, then just sweep Freq until you get what you want. As soon as Res is higher than 9 o’clock it’s in boost mode, and at higher settings it can boost quite a lot. You can think of those two controls as a highly interactive variable frequency treble control, although it can go deep into the midrange area as well.

Thank you very very much!

 

[dcoyle]

I start out with the Res at 10 or 11 o'clock (didn't realize 9 o'clock was flat) and turn the Freq up all the way. Freq goes out up to 7 KHz, IIRC, where most MI 12" or larger speakers are falling off pretty steeply, although not PA or bass rigs with horns. For me, my speakers, this is flatish. To boost some treble I turn the Res up to 1 o'c or far enough to notice, and sweep down from full CCW to find the frequency that works. Then I play with how much boost on the Res. I'm seldom above 3 o'c. Treble boost feeling can be enhanced by some mid cut, sometimes.

 

[Passinwind]

I start out with the Res at 10 or 11 o'clock (didn't realize 9 o'clock was flat) and turn the Freq up all the way. Freq goes out up to 7 KHz, IIRC, where most MI 12" or larger speakers are falling off pretty steeply, although not PA or bass rigs with horns. For me, my speakers, this is flatish. To boost some treble I turn the Res up to 1 o'c or far enough to notice, and sweep down from full CCW to find the frequency that works. Then I play with how much boost on the Res. I'm seldom above 3 o'c. Treble boost feeling can be enhanced by some mid cut, sometimes.

You may actually be at closer to flat than what I described, depending on other interactions. But in any case I tried to make the most useful settings generally fall between ten and two o'clock on all the controls, because that seemingly tends to be about as far as many people are comfortable with moving them.

 

[BQFS]

Very helpful as well. I really appreciate these responses and have sent them to my friend. Thanks ever so much to you both!

 

[dcoyle]

One of the really powerful things about the PW3B is that as well as having a variable frequency and amount of resonance for the treble section, it has a variable amount of resonance at the bass end. The Hpf changes the frequency of the -F3 of the high pass crossover, but also the Q, or resonance, or bloom versus taught-ness the note has.

I start with the Bass flat at 11 o'c, and twiddle the Hpf to get the amount of "bounce" I want on the E string. A lot of bump is nice for pulse, but tight enough to not go flabby. Thinner is nice for articulation. After getting the right feel on the Hpf, I dial up the Bass control to get the proper amount of bass energy for the room.

I expect DB players have a very different strategy to use the Hpf to protect speakers, but I play 4 string EB through large, low Q speakers, indoors, and I'm not really worried about over excursion.

 

[aquataur]

I was waiting for some components, so I played a little with my SPICE implementation.
I was interested to see how well such a "filter type" circuit can do a pickup revoicing. I know that SVF's are used for that, but nothing speaks against an implementation as Charlie has done it. Since this is not going to be built into a bass, but rather resides in an external box, it was of particular interest how it would do in the vicinity of an already voiced bass, as most off-the-shelf instruments will be (hopefully).

If you are interested on what I found, you are invited to read the document on my web pages. There is no way this would fit into a structure like this here.

http://me.aquataur.guru/musicstuff/passinwind.html

have fun.

 

[Passinwind]

I've been refurbing some older v2 builds, still tweaking away on different EQ curves and input impedances.

This is my oldest existing PW3B-LPF build, which went to NAMM in 2017:

It's battery power only, and by now I've lost track of how many different boards have been in there. Mid freqs are 240/420/700, LPF sweep is 280Hz-9kHz.

And then this one is now optimized for my Breedlove acoustic guitar, which already has an onboard Fishman preamp:

Bass peak frequency is bumped up about an octave, mids are 240/432/800Hz nominal. Sweep is 500Hz-8kHz, all pots are custom Omeg conductive plastic jobs. This build is non-inverting now, unlike the earlier v2 ones.

And finally, a very rare v3 build (aka WTHPF), @Jim C has the only other one like this:

HPF sweep is 20Hz/120Hz, Mids are 240.423/720, LPF sweep is 660Hz-8kHz. There's just something about these v3 builds for bass guitar.

 

[Passinwind]

Exercising my new computer a little, this sim of 336 setting combinations took about a minute to render:

That represents turning the Bass, Res, and Freq pots over the min to max range in various combinations. If I add in three midrange frequency switch positions and the Mid boost/cut sweep it will be unreadable, I reckon. This is for my newest v2 acoustic guitar box BTW. Graph X axis limits are 25Hz and 20kHz.

 

[Sanj]

Hi Charlie, I'm curious about your naming convention for the various iterations of pedals -- WTHPF vs. PW3B-LPF.

 

[Passinwind]

Hi Charlie, I'm curious about your naming convention for the various iterations of pedals -- WTHPF vs. PW3B-LPF.

The WTHP fork builds have all been portrait mode rather than landscape, this is the more fully featured 1590BBS version:

So basically just a cosmetic thing.

 

[aquataur]

And here are some customizations explained.

 

[Passinwind]

And here are some customizations explained.

Nice writeup, certainly way more than I'm willing to do by now, and I'm glad you're enjoying playing with one of the old unsupported versions. I'd appreciate you emphasizing that aspect though, especially since v4 is actually meant only for pedals and works very well for my primary design case of using active basses, which have always been my strong personal preference.

Also, even v1 was built assuming at least one normal volume pot (i.e. 250k-500k) being in the instrument in front of it. I think you may still be missing the point of lessening the pickup's native resonant peak, but that has been a long running theme in this thread and the only way to know what input impedance you prefer in any specific build is to try as many different ones as you can stand.

In any case, nice to see another alternative design fork popping up, and I hope you and @Tassieviking may take a newer and better v2 fork somewhere tangible.

 

[Sanj]

Charlie, I was wondering why the LPF board used separate single op-amps instead of one dual. Thanks!

 

[Passinwind]

Charlie, I was wondering why the LPF board used separate single op-amps instead of one dual. Thanks!

It’s mostly a legacy biproduct of there only being a single opamp version of OPA145/OPA2145 back when I did that design. But it also makes for a nicer board layout, IMO. And I sometimes like to use different opamp flavors in those two slots, which may border on cork sniffing.

 

[Sanj]

which may border on cork sniffing.

 

[Passinwind]

Speaking of cork sniffing, I decided to revisit the two unloved one board through-hole v2 builds I stashed away some time ago. I've never built to the spec @aquataur used and ended up revamping a lot, but these use similarly conservative specs since I was mostly trying to work out something that would work with the opamps @Tassieviking wanted to use, which didn't really pan out anyway. TL072 works OK with this spec stability wise though , and it's less expensive than OPA2134 or 2132 and also available at Tayda anyway. Personally, I'd just buy them from Mouser since they're less than a buck apiece there. Anyhow, in my last Mouser order I used a little slush fund money to buy a couple of LT1113s, which I have found to be very forgiving and good sounding in many other applications over the years. Not inexpensive though, or so readily available overseas from what I gather. Current pricing for the DIP-8 ones is $10.68 a pop at Mouser. I can always use them in something else if this swerve proves unworthy.

So I have these two test pedals, with close but not identical specs:

The one on the left got the LT chips and the one on the right got cheap and cheerful TL072s. Despite the substantially different noise specs I found no appreciable difference in noise floor. Both are very quiet though, at least. The pedals did sound a bit different however, and the knobs tended to point in somewhat different places when I tried to match the two as closely as possible. And in all honesty, both were a bit meh.

I'm not one to give up without a fight though, so I finally came to grips my new to me MOTU interface and did some RTA testing. These older designs have always correlated a little less fully with my Spice models than newer ones, for whatever reason. I found a few clues and revamped the silver one with a different LPF tuning, which sounds really good now for both bass and acoustic guitar. It'll be a couple days before I can do proper A/B testing with the different opamps, but if the initial trend holds I think I might be able to revive the compact v2 after all. I already took a stab at yet another board layout, which wouldn't use the cheap stock Tayda/PPCB 4 hole enclosure, but Tayda's drilling service is inexpensive and moving the bottom controls a bit lower would make for an easier build and definitely a better looking PCB layout. So equal spacing horizontally and vertically, like so:

There's plenty of room inside for a 1/4" longer PCB, but my gut feeling says the the Bass and Mid controls should really be in the bottom row. Not sure I have it in me to do all that by now, but maybe.

 

[Tassieviking]

Those pre drilled enclosures made it really tight between the pots and drilling 6 holes on the front panel should be easy to do for anyone who can build their own pedal.
I actually wonder if there is really a need for a battery or not, I have a feeling most people who would go for this style of pedal would use a proper power supply.
But since it can be made to fit a battery maybe it should be made that way.
I started to play around with the footprints of the op-amps, I made one up with a DIL 8 pin with a SMD DIL footprint inside it so either TH of SMD version could be used.

 

[Passinwind]

Those pre drilled enclosures made it really tight between the pots and drilling 6 holes on the front panel should be easy to do for anyone who can build their own pedal.
I actually wonder if there is really a need for a battery or not, I have a feeling most people who would go for this style of pedal would use a proper power supply.
But since it can be made to fit a battery maybe it should be made that way.
I started to play around with the footprints of the op-amps, I made one up with a DIL 8 pin with a SMD DIL footprint inside it so either TH of SMD version could be used.

Agreed that that the pot spacing was uncomfortably tight, that was my primary reason for wanting symmetrical spacing. I don’t think a battery option is a high priority at all, but I also don’t think a footswitch or top jacks are needed in most cases, so what do I know.

I’d like to see how you did the nested opamp footprints, my tries have never been workable at all. But I consider having bypass caps right on the power pins mandatory, which you may not. Frankly, I’d rather just use SOIC-8 opamps and 1206 resistors and call it a day, but many people still seem to be uncomfortable with that, which I respect.

 

[Tassieviking]

Here is a quick pic from KiCad of the footprint.
I place the TH and the SOIC on the board with 2 VIAS and then I add the traces after.
I can make the footprint joined into one but I haven't figured out how to add the traces to the footprint so they get placed later.
I made the footprints but they might still be in the KiCad7 version on my old computer, I have to go search for them.
I also have one for the resistors and one for diodes that can be TH or SMD