Ludwig phase II filters

[Gus]

R.G. or ?
I can't find the filter design in the Phase II in books I have or on the web.  I read most of the threads/posts on this circuit and R.G.'s tech overview some interesting stuff.

[R.G.]

I'm pretty sure that if one did some analysis that you could make this match up to one of the active filter formats by replacing transistor transconductances and so on with "resistors" in an active filter. I never wanted to spend the time. There may be magic buried in there, but I never spent the time to dig for it.

I've done a couple of OTA based state variable filters that do the same job with one LM13700 per filter, and they work fine. That ended my interest in digging out how the Ludwig filters work. I stuffed them into a conceptual block labeled "complicated, limited range voltage/current controlled filter" and left them there.

The problem with an OTA state variable filter in this application is that you have to do serious work limiting the frequency swing range -down- to what's needed.

[puretube]

R.G. or ?
I can't find the filter design in the Phase II in books I have or on the web.  I read most of the threads/posts on this circuit and R.G.'s tech overview some interesting stuff.

It`s not in the boox, but here,
and in the office...



btw.:       ...       ...       ...   While you`re at it: youze guyz seen this young-one yet???





[edit] and: this one...

[Gus]

puretube

thank you for the links.
What has not been done with before with 7787634?

[R.G.]

[edit] and: this one...

How 'bout that! He invented the MOS Doubler 11 years before I did.    Think how much he'd have made from it if he'd used it for guitar effects instead of the silly world of pro audio and broadcast electronics. 

And I thought it was just "obvious to one skilled in the art". 'Course, that's also what I thought about the dying battery simulator that was disclosed and patented some months after it appeared at Geofex. I think I just expect too much from 'one skilled in the art'. 

[Cliff Schecht]

RG you've said too much already..

[Cliff Schecht]

This filter is *really* interesting, I've never seen one done quite this way before. I also have no hands on experience with this circuit so I can't say much about it without some study. I suspect it has a relatively limited range because the diff pair they use will crap out if the current gets too low or too high (or at least this is my experience). This tends to limit the usable range quite a bit unless you come up with some sort of current boosting circuit to drive more current into the cap.

Ok, after I guessed I went and read the patent a bit. Didn't realize these were formant filters with some sort of gyrator circuit in them. Even cooler! I also saw in the patent that the range covers about 2 octaves which is enough for most modulation type guitar effects (this would be pretty lame in a real synth). I'm not sure what the general consensus on this effect is but I think the heart of this circuit would have all of the non-linear goodness that one can expect from completely discrete-based voltage controlled filters. Even if the Phase II effect isn't all that great (I saw someone say it's "not the bees knees"), I think it's a cool basis for some sort of VCF.

[R.G.]

I think the limited range may be a side effect of the gyrator operation as well. An L-C filter's center frequency is a function of the square root of L*C, so changing only one of them makes the filter frequency very limited. And that's before the addition of the current edge effects in diffamps.

Having messed with this a bit, I still have my first impression of the filters intact - boy, that's a lot of work for a tunable filter! It's probably fun to mess with, but even after doing the PCB for the Ludwig, I went off and did the state variables version. I know by experience that I can dirty up and oddity-up an SV. 

The time this thing was current was kind of the last hurrah of complex tens-of-discrete-transistor stuff in mass production. The IC swept all that away in the next few years.