Subcommander LPF question

[edenexter]

Hey everyone, been experimenting for a while with some of the projects on MFOS and enjoying every bit of it! I'm not new to electronics but I wouldn't call myself experienced.

The time has come that I could use a little bit of help. I've breadboarded the VC-LPF from the subcommander guitar project and I think it sounds great! So great that I'd like to use it out of the context of the subcommander. I understand that just taking this bit out of the whole project might lead to problems but I'm curious to see if I can get this filter working correctly without the rest (input stage & output stage).

So far the filter works absolutely fine except for when I sweep the cutoff all the way down. There seems to be a range where the filter switches off or something. When I sweep back up there is a bit of a pop just before the sound comes back. When using an external LFO for the control voltage this popping gets pretty irritating, it's still fairly subtle though (but not enough obviously). I've played around by changing the resistor values going from -12v and +12v to the COF pot so the range is a bit better and it won't sweep in to this 'dead' zone, but my (10v p-p) external-LFO will still pull the COF down further so that I get pops...

If I'm too vague in my description I'll happily go into more detail of what I'm doing and what I'm trying to achieve. Any help would be much appreciated!

e: Link to the subcommander http://www.musicfromouterspace.com/analogsynth_new/GUITARSYNTHAUG2007/GUITARSYNTHAUG2007.html

[jordroid]

Hello and welcome,

the sub-commander VCF looks to be a state variable filter, a very common type of filter to use. I don't have a direct answer to your question, but there is a lot of information about state variable filters out there, and you may be able to get improved performance by reading up on them and breadboarding some stuff, and seeing how the subtle changes alter performance.

There is a standalone SVF project from MFOS, it includes voltage controlled resonance, which is kind of a luxury, although i don't find it all that useful most of the time, YMMV

there is the "VCF-1" from Thomas Henry here http://www.birthofasynth.com/Thomas_Henry/Pages/VCF-1.html you can replace the two CA3080s with a single LM13700, they are easier to find and cheaper, and it sounds like you already have some. No VC resonance in this design.

and there is much written on SVFs in the electronotes series, available in paper form from here http://electronotes.netfirms.com/ this is probably the place to find the most comprehensive information.

just googling "state variable filter" will probably get you a bunch of info as well.

One of the big benefits of a SVF is that there are simultaneous low pass, band pass and high pass outputs available, i'm not sure why Ray didn't include them in the sub-commander project, maybe just for simplicity, but there are some really cool sounds to be had from the band pass and high pass outputs.

I have not run into your exact problem, but have built several SVFs, including the MFOS standalone SVF and the Thomas Henry VCF-1, maybe the sub-commander SVF is not quite as optimized, or maybe there is some error in your breadboard that is undetected.

I see that the sub-commander SVF uses the built in darlington buffers in the LM13700 package (pins 7, 8, 9, 10 & 11), most of the SVF designs i see don't use them, and instead use external buffers. I'm not sure if the built in buffers are prone to problems or if they are ignored for other reasons (most of the SVF designs descend from designs using CA3080s, which did not have internal darlington buffers), maybe a place to check anyways.

Since you are the breadboarding kind i would recommend breadboarding a few different types and seeing which you like best, they tend to be very similar with minor differences, usually in how the different stages are buffered and some resistor/capacitor value changes.

regards,

jordan

[inlifeindeath]

you could try a pot to attenuate the amount of CV going into the filter. This would limit the range of the sweeps somewhat.

[edenexter]

Thanks for the replies!

I've tried attenuating the CV, but to remove the problem the sweep got very short. So that didn't work well plus its only a symptomatic fix..

Good news though, cause I replaced the darlington buffers with simple voltage followers using the tl072. The sound cleared up nicely!

Too bad the buffers are so problematic, but looking at all the examples Jordroid posted many designs don't use them.

Thanks again!