A vote for the DOD Envelope Follower/Filter

[Mark Hammer]

I had an etched board for the DOD 440 (GGG layout), and decided to stuff it and wire it up this weekend.  First build in a while, actually.

I grabbed a pair of photocells from a pack I had bought recently, installed them at a 120-degree angle to each other on the board, and filed down the plastic on a 5mm white LED so that is nestled in nicely between the LDRs, and put a piece of duct tape over the assembly to keep ambient light out.  And, to my great satisfaction, it worked out really well.  Nothing particularly special about the LDRs.  These were cheap ones I got somewhere for what I seem to recall was 10 cents each, spec'd at approximately 10k-50M.  Some of the drawings I've seen show the LDRs as one single photocell bridging the pair of 220k resistors.  This is incorrect.

True to form, I did a couple of mods.  I increased the sensitivity of the envelope follower by changing the 470k feedback resistor for 560k.  I also tried changing the Q of the filter by increasing the value of the 430k feedback resistor.  Initially, I used a 560k, with a 1M8 resistor switched in parallel to yield 427k (i.e., I installed a toggle for high/low resonance).  

560k turned out to be way too much, yielding a very unpleasant intense "chirp/shriek" at the top of sweep (i.e., what felt like about 15-20db louder).  I remembered that (former DOD engineer) David Difrancesco had a short piece about modding the 440 in a 30 year-old issue of Polyphony, so I dug it up and saw that he recommended use of the stock 430k and a 100k pot in series.  Clearly, 560k was beyond the appropriate range, so I rigged up an arrangement that would add 82k to a 430k resistor.  Even that proved to be a bit much, and still yielded the chirp, albeit not as deadly.  I think I'll use a 68k instead.  The chirp was not quite as bad when the filter offset was turned down to the lower bass-ey range.  I suspect that the 68k "high-resonance" mod will allow for a usable sweep for both higher and lower offsets.

EDIT:  Since that resistance change also alters the gain of that stage, it might be useful to include an output volume control; especially since the schematic shows an unterminated output cap...POP CITY!
The envelope-extractor/rectifier section uses a 1uf averaging cap, and nothing else besides that and the LDR specs to adjust the feel of the sweep.  I may switch the 1uf for 4u7 or even 10uf this evening, with a toggle to insert a bleed resistor for fast/slow decay.  We'll see how that works, and I'll report back.

The unit has a very pleasant sound for a swept bandpass.  Much more vocal-sounding than many of the other simple bandpass units I've made.
In general, a pleasing build.  Glad I took a chance on it.  

[Blitz Krieg]

helpful, thanks

[Beo]

Thanks Mark, this is one I never looked at.

Your story makes me wonder though. I'm not one to throw together a circuit or breadboard if I'm not intending to box it. I kind of have an inertia or (pun) resistance to it. As a result, messing with new circuits is always big endeavor for me... I read everything I can about a circuit, pick and choose between different variants, and then start thinking about footprint, onboard pot placement, and etch graphics. I'm probably missing out a lot on learning and experience experimenting with different circuits... but my solder time is pretty limited, so when I hobby, it's got a be towards an outcome that goes in a box.

A little off topic I suppose, but my main point should be, these kinds of posts are really helpful for me, cause I don't find the time to experiment with lesser known pedals, and your experimentation makes up for time I can never find.

[Mark Hammer]

Glad to help out Beo.  I have several bins of stuff that is either wired up with pots and jacks hanging off it, or mostly stuffed and wired up but not yet up and running.

A lot of it I build just to hear how it sounds, and to understand how something works.  Thankfully, a lot of these circuits are simple enough to throw together on perfboard, or have layouts provided such that a board can be etched, drilled and stuffed on a Sunday afternoon.

Boxing up IS often the most tedious and costly part.  It's not just the control-layout planning, machining, painting, and legending, but the cost of the boxes and the knobs.  A stuffed board with a bunch of stuff hanging off of it requires none of that additional labour and cost.  So I can understand the reticence to become seduced by building unless one is going to commit to a finished product.

If I have to be the scout that comes back with a rash and says not to go in those bushes because there is poison ivy, so be it.   

Incidentally, do any of you know what sort of filter this unit employs?  It's a configuration I'm not familiar with; although maybe if it was drawn differently, I'd slap my forehead and utters "Ohhhh, one of those."

[Kipper4]

can wait to see how you get on Mark.
I'm not sure what kind of filter that is. initially i thought it could be a multiple feedback filter but now I'm uncertain.
good luck.

[Mark Hammer]

I switched out the 1uf averaging cap for 4u7, and it works fine.  I think I'll leave out the different decay settings.

The change to 430k+68k improved the "chirp".  I still get a bit when the sweep range is moved higher up, but nothing that earns a "Turn it down!" from the kitchen.  It results in a nice audible difference in filter emphasis that offers more options when the sweep range is moved to the mids and lows.  At the moment, I don't have a bypass switch wired up, so I may find an volume control necessary or unnecessary when I am finally able to compare filter/bypass levels.

AFAIC, you can never have enough envelope-controlled filters.   

[Kipper4]

AFAIC, you can never have enough envelope-controlled filters.   

Too true I love my filters.

[aron]

I bought a modded one on eBay, I like it and use it a lot. Maybe someone on here modded it.

[Mark Hammer]

Do you have any idea in what ways it was modded, or was it simply sold to you as "modded"?

[aron]

Sorry just took a look at it - FX25B GREEN with a range switch. I love envelope filters. Always had one. Mutron Micro V, Oberheim VCF etc....
I really regret selling my Mutron Micro V. At the time I thought I could build ANY pedal. oh well.....

[PRR]

> what sort of filter this unit employs?

Wien.

[Mark Hammer]

Sorry just took a look at it - FX25B GREEN with a range switch. I love envelope filters. Always had one. Mutron Micro V, Oberheim VCF etc....
I really regret selling my Mutron Micro V. At the time I thought I could build ANY pedal. oh well.....

It may not be required, since the FX25B includes a clean-blend control, but the standard FX25 and 25B taps the filter at the bandpass output.  I've been successful at tapping the lowpass output from it for a bass-playing friend.  Just note that the lowpass output is louder than the bandpass.

> what sort of filter this unit employs?

Wien.

Thanks Paul.

[nocentelli]

Any suggestions for a simple but effective way to rig up the two LDRs with the single LED into a neat package for the breadboard? For single opto's, I'd usually just shove them face to face into a suitably narrow section of shrink-tube, but it's near impossible to keep two LDRs properly aligned this way.

[davent]

First thought would be use a clear resin to cast them into a unit then light proof the assemblage.

[Mark Hammer]

Although I've experimented with building them, I have not put any to work.

With that caveat, what I've tried, with some physical success, is to file down the sides of a regular 5mm dome-shaped LED, so that an LDR could be laid flat against each side of it.  A piece of heat shrink tubing (big enough) can then be slipped over the entire thing.  You end up with a 6-legged structure that can be mounted vertically (i.e., all 6 legs pointing down).

In my particular instance, I used the GGG layout, drilled an extra hole for the additional lead (though in hindsight, a slightly bigger middle hole would have likely accepted a lead from each LDR), and angled the LDRs so they were pointing inwards.  The LED was bent over at right angle to the board, and the dome filed down to make a fat V, that fit in the little V-shaped corner made by the two LDRs.  Viewed from overhead, it looks sort of like this.

L    /  ___
D  /  /     ]  LED
R  \  \___]
    \

[nocentelli]

How about using two discrete vactrols? Would the LEDs be better put in series or parallel?

[duck_arse]

in the past, I've used an 8 pin IC socket, with the led mounted in the middle and an ldr at either end, craning over the top of the led. it needs a custom hat/cover to block the light, obviously. another method I used was a pen tube, with a hole drilled in the side for the led, and the ldr's poked either end, and blacked up some.

both methods were for playing with, and not actually built into anything.

[Kipper4]

What about super glueing the ldrs to the LED. As for a light tight enclosure you got me.

[Mark Hammer]

In use (i.e., both input and output jacks occupied), most pedals will be light-tight, with the exception of exceedingly bright status LEDs that shine into the chassis through their bottom (easily fixed with a dab of paint on the bottom of the LED).  The chief concern over DIY optoisolators is setup and adjustment before they get boxed up.  Well, that and assuring the LDRs stay in the optimal position, relative to the light source.

[Cozybuilder]

"What about super glueing the ldrs to the LED. As for a light tight enclosure you got me."

Black RTV