"Funky MF (Multimode Filter)" - LM13700 Multimode Envelope Filter

[StephenGiles]

I would be interested to see a combination of your filter with my sweep generator from the Space Filter, which is dying a death out there!

http://www.4shared.com/file/89784300/b7d68e64/Space_Filter_S__H.html

I think there might just be a little more scope. What do you think Cliff Schecht?

[liquids]

I've thought about it.  I don't have sound samples though, so it's hard to get excited unless I know.  I had a filter that I believe was based on the MXR.  It didn't have that aggressive resonance I'm actually most interested in, which even the mutron/q-tron circuit I have only begins to approach.  I'm really most interested in the filter section of the EH Microsynth (StephenGiles Space Filter is a big interest, but I lack some parts and the debugging skills to do that, for the time being)  or an MS-20/Funk-a-duck thing!  From the sound samples, this is aggressive enough and looks doable, so it will scratch part of the itch.

Maybe the MXR is more aggressive than I'm thinking, though, anyone got samples?  I'll search.

[StephenGiles]

I've thought about it.  I don't have sound samples though, so it's hard to get excited unless I know.  I had a filter that I believe was based on the MXR.  It didn't have that aggressive resonance I'm actually most interested in, which even the mutron/q-tron circuit I have only begins to approach.  I'm really most interested in the filter section of the EH Microsynth (StephenGiles Space Filter is a big interest, but I lack some parts and the debugging skills to do that, for the time being)  or an MS-20/Funk-a-duck thing!  From the sound samples, this is aggressive enough and looks doable, so it will scratch part of the itch.

Maybe the MXR is more aggressive than I'm thinking, though, anyone got samples?  I'll search.

No samples for Space Filter as I have failed Gordon Brownly  to get a filter section to work with the sweep generator   - not that I've really spent that much time on it. The Multimode filter clearly works. I would prefer to have 2 filter sections in parallel to give a more throaty vocal sound, as in the Bifilter follower. The envelope follower is fine on "funk a duck" sounds which I personally can't stand, but on longer sweeps suffers from bad ripple. This is why I think the LM13600ified Space Drum sweep generator will provide a smooth long sweep - because it does to the VCO in the Space Drum.

I'm busy for the next few days so someone might like to have a go. 

[orangetones]

The Multimode filter clearly works.

Well...  I tried my hardest to get this thing to work on the breadboard and still, it didn't work like the sound samples posted by freq. central.

Has anyone else breadboarded this and gotten it to work properly?

[frequencycentral]

I would be interested to see a combination of your filter with my sweep generator from the Space Filter, which is dying a death out there!

The Multimode filter clearly works.......The envelope follower is fine on "funk a duck" sounds which I personally can't stand, but on longer sweeps suffers from bad ripple. This is why I think the LM13600ified Space Drum sweep generator will provide a smooth long sweep - because it does to the VCO in the Space Drum.

I'd be interested it that too Stephen, and I for one follow every post on the Space Filter thread. My Multimode sounds great (though I can't claim to have designed it), though I'd have to agree that it ripples badly on long sweeps, I'd love a follower that was more forgiving.

I've got an idea that a leading edge envelope detector would be the way to go: http://www.diystompboxes.com/smfforum/index.php?topic=73059.msg592144#msg592144

[liquids]

Well...  I tried my hardest to get this thing to work on the breadboard and still, it didn't work like the sound samples posted by freq. central.

Has anyone else breadboarded this and gotten it to work properly?

I finally breadboarded it - it's was the one thing I wanted to do before I go away for a few days tomorrow.  Sad to say, nearly no success.  Man, is that an eyesore on the breadboard.  Something like this, it seems breadboarding it once and then starting over is a good thing, so you can visually know what kind of space you need and where when you do, to keep it clean and neat. 

Anyhow, I've gotten it to pass a signal, one that is oddly distorted. My voltages on the LM13700 match yours Rick, dropped to 9.48v, other than your voltage on pin 15, which I thought was supposed to be not connected to anything--thought you list 5.99V on that pin  Oops?

I double checked all my connections and audio probed, etc, preliminarily, but no luck. I Can't check my voltages on the op-amps because I used a different op amps and also the most sensible section of the quad op amp to use, rather than following the layout posted.  I also need to solder up some more pots for breadboarding for this one. I'm lacking the LM324 that this, the original circuit, and the Space Filter seem to request...time to do an order, methinks....

The LED test proved simply to light up constantly.  DMM on pins 1/16 showed no voltage change when playing, so there is something happening there. I don't understand the envelope section...maybe I will replace it with a simpler one? The only thing that I have to show for it, is that when I disconnected the 10K from the buffer going toward pin 16, I could hear the filter set manually and the resonance pot affected the sound.   I'll mess around for fun the rest of the night and then maybe start over when I get back next weekend. 

[frequencycentral]

This does seem to be a problematic circuit. Strange. The way I work with new designs is to breadboard until I'm ready to build, then draw up a schematic from the breadboard, and build from that schematic, so the schematic I built from is the same one you built from. If you still want to experiment with this, can I suggest that you refer back to the circuit I based (stole?) my design on, build it on your breadboard, test it works, then systematically modify it to my design. Thats how I did mine, it started off as the Synbal circuit, then I modified it. I will be relieved when at least one person has built it without issues, as I'm feeling big guilt at the moment for all the wasted man hours. Boogdish at least got the filter section working, though he hooked it up to a different follower.

[StephenGiles]

This does seem to be a problematic circuit. Strange. The way I work with new designs is to breadboard until I'm ready to build, then draw up a schematic from the breadboard, and build from that schematic, so the schematic I built from is the same one you built from. If you still want to experiment with this, can I suggest that you refer back to the circuit I based (stole?) my design on, build it on your breadboard, test it works, then systematically modify it to my design. Thats how I did mine, it started off as the Synbal circuit, then I modified it. I will be relieved when at least one person has built it without issues, as I'm feeling big guilt at the moment for all the wasted man hours. Boogdish at least got the filter section working, though he hooked it up to a different follower.

I tend to draw a circuit as I go when breadboading, as it is a real pain to unravel a birds nest afterwards!!

[liquids]

This does seem to be a problematic circuit. Strange. The way I work with new designs is to breadboard until I'm ready to build, then draw up a schematic from the breadboard, and build from that schematic, so the schematic I built from is the same one you built from. If you still want to experiment with this, can I suggest that you refer back to the circuit I based (stole?) my design on, build it on your breadboard, test it works, then systematically modify it to my design. Thats how I did mine, it started off as the Synbal circuit, then I modified it. I will be relieved when at least one person has built it without issues, as I'm feeling big guilt at the moment for all the wasted man hours. Boogdish at least got the filter section working, though he hooked it up to a different follower.

I've refered back the original circuit as a built and actually did start from scratch after I tried it first, for neatness, to the same result. I'm frustrated that I can't seem to get the envelope section...when I have the two parts connected, there is no filtering, but disconnecting that 10K resistor, I hear the filter, with the resonance pot sounding as if to finction more like a manual filter control...when set high, it's bright, set low it's muffled. Odd.

With the original circuit, it's hard for me to break it down into it's parts, though refering back to your schematic helps as many of the values of course are the same.   I don't mind the man hours, it's worth it to me, not a waste!  I'm going to give Stephen's space filter a shot next, but I'll come back to this maybe...I do have it sitting on my breadboard still, so maybe it depends on what itch I get.   

I have to confess as I hinted at earlier, that I didn't have all the pots breadboarded, and had some as preset resistors.  If you could post some good 'starting' points for the pots, since there are a few 'widely' variable values there that I can't even guess at.  Twisting them didn't do muchfor me though.  I did, for example have a decay set 'halfway' as a 470k resistor to ground the whole time....maybe that was throwing it off a bit.  I also ignored the HP/HP taps for simplicity and personal preference. 

I understand filter settings theoretically, but from your experience in particular, what are good values to start at here, say? Likewise, have you noticed preference for pickups type/pickup output/running a buffer?  Also, looking at the circuit, you have the signal running to the top part, the op amps (this is the voltage control section, correct?) bled off the (potentially unbuffered) input signal splitting also into the filter section (that is, the parts that hits one op-amp stage and then the LM13700 stages)?   

[Cliff Schecht]

Something to remember with the LM13700 is that transconductance is a function of temperature and varies all over the board from one LM13700 to the next (although the two OTA's within one IC are matched well). This essentially means that the position of a pot for Rick's build might not work so well for you, especially if it's the envelope follower input offset potentiometer. The attack and decay pots can be just resistors, but the offset pot sets the initial position of the filter (a DC level that determines what the output current and hence the filter cutoff frequency). Since each LM13700 varies so much, this pot position won't be the same from one board to the next (plus you have to factor in 5% resistor tolerances and the 20% pot tolerances).

[StephenGiles]

Something to remember with the LM13700 is that transconductance is a function of temperature and varies all over the board from one LM13700 to the next (although the two OTA's within one IC are matched well). This essentially means that the position of a pot for Rick's build might not work so well for you, especially if it's the envelope follower input offset potentiometer. The attack and decay pots can be just resistors, but the offset pot sets the initial position of the filter (a DC level that determines what the output current and hence the filter cutoff frequency). Since each LM13700 varies so much, this pot position won't be the same from one board to the next (plus you have to factor in 5% resistor tolerances and the 20% pot tolerances).

They won't notice that

[orangetones]

They won't notice that

What do you mean by this?  Who are they?  The chips? builders won't notice this information?

I had a devil of a time with this circuit.  If you are having trouble too, read back and see what I did and make sure that isn't the issue for you too.

[Slade]

They won't notice that

What do you mean by this?  Who are they?  The chips? builders won't notice this information?

I had a devil of a time with this circuit.  If you are having trouble too, read back and see what I did and make sure that isn't the issue for you too.

Sorry, I didn't read all the pages before this, but I wanted to ask you if you have tried with a TL072 instead of the NE5532? I had a trouble with a NE5532 in a delay that later worked fine with a TL072...
Greetings

[StephenGiles]

They won't notice that

What do you mean by this?  Who are they?  The chips? builders won't notice this information?

Los fuzzboxerios!!

[chicago_mike]

Hey all,

Just a couple quick Q's :

1. How would I add led's like in Freq Centrals unit> For the HP, BP and LP modes?

2. I am almost done building 2 of these and wonder where I could put the blend control? Can it be off the pcb and would it just blend the dr input and the output?

thanks!

[frequencycentral]

Hey all,

Just a couple quick Q's :

1. How would I add led's like in Freq Centrals unit> For the HP, BP and LP modes?

2. I am almost done building 2 of these and wonder where I could put the blend control? Can it be off the pcb and would it just blend the dr input and the output?

thanks!

I used a double pole 3 position (DP3T?) rotary switch to select the filter mode/LEDs. The cathodes of all three LEDs are grounded, the anodes are connected to each of the three positions of the switch, the common of the switch is connected to +ve via a 1.5K resistor.

Not sure of the best way to implement a blend pot, but would guess just feeding the clean and effected signals into lugs 1 and 3 of a 22k Lin pot would do it, maybe with a single opamp buffr after it?

I'm using mine a lot, wedged in between a Big Muff (before the MF) and a Clari(not). The Clari is set to a fast slapback, and the fuzz distorts the filter ina really wild way. I'll post up some soundclips, it does sound awesome!

Good luck with your builds, I hope they work out, I'm beginning to feel this circuit is jinxed!

[sean k]

Thanks for the links to the original FC!, I went and got all the rest of the synth drum pdf's as well and have been especially interested in the synblo which uses an air tube onto a crystal mic element to control envelopes.

Looking for crystal mics was interesting too and eventually I found a Utube vid of this guy who built one with a home grown crystal in a toilet roll with paper cellotaped over one end and a stick, or somesuch, between the paper and the piezo to affect the surface of the piezo.

I've only googled the synblo but I found a much better scan than the electronics anonymous scans.

I also quite like the 40106 and 4070 on the synbal and what I can see is, with my entirely limited knowledge (and I can't find my electro resource book) is that the 4070 is used as a mixer of sorts that as each gate goes high, or low?, it takes that signal. Cool! could be interesting as an LFO mixer 

Seems the earlier post where the LFO was applied to pin 1,16 of the OTA through an 800k resistor combined with running the envelope into an LED might be the best way to meld the operating point of the two functions.

[chicago_mike]

Well I have faith that mine will work.

[frequencycentral]

...faith...

That's what I use, far better than solder or book learning. 

[frequencycentral]

I think I might have found the bug in this design. I just breadboarded the envelope follower for another circuit I'm working on. I used a 4558 for the two opamps that make up the envelope generator - and it didn't work. As soon as I changed to an LM324 for those two opamps it worked. Is that it? I hope so. Strange, I would have thought it would work the same with 4558s.