Electronic switching system to change multiple components simultaneously?

[HD Evans]

I'm looking for a simple, reliable, and low power solution to instantly change values in a system of state variable filters.  One filter is attached to each pickup coil, two filters are used on humbuckers, and for the time being I am designing for a maximum of 5 coils/filters.  To make things more user friendly "tone modules", which will consist of matched and encapsulated resistors and/or capacitors,  will allow users to install desired preset voicings.  The tone modules will be responsible for setting the gain, resonant frequency, and Q of the circuit.  In case you're wondering, the coils are wound using traditional geometry but make use of 36AWG wire to bring impedance to 1k ohm or less:  resulting in a flat response through the entire audible spectrum.

A fancier system is already further through development, and uses digital pots in conjunction with a low power microcontroller to allow flexible presets and switching via midi or bluetooth. 

This brings me to the question at hand:  how best can 20 component values be switched electronically?  Would it make more sense to build two filters onto each coil so that only 5 switches need to be flipped each time?  Or, am I simply missing a more obvious and simple solution?

If it would help I can post some schematics as well, I just need to move my sketches from paper to the computer.

Thanks for your input.  This will be an open design, although it will also be a commercial product, so your assistance will go into something that will be hacker/diy friendly. 

[R.G.]

I'm looking for a simple, reliable, and low power solution to instantly change values in a system of state variable filters.
Electronic switching; something like CMOS analog switches as long as the 100-500 ohms of the switch itself can be made negligible or compensated for.

In case you're wondering, the coils are wound using traditional geometry but make use of 36AWG wire to bring impedance to 1k ohm or less:  resulting in a flat response through the entire audible spectrum.

That doesn't necessarily follow. Big wire and low resistance has less effect than number of turns and hence turn-squared inductance in a pickup. It's the inductance and self-capacitance of the coil that makes the response vary. Neglecting magnetic leakage, output voltage scales linearly with turns, inductance scales with square of turns, and capacitance is more geometry related than number of turns related. Cutting number of turns is the way to get wide response, but you really have to work hard on getting the level up without introducing noise. This the same conundrum faced by phono cartridge designers with moving-coil cartridges.

A fancier system is already further through development, and uses digital pots in conjunction with a low power microcontroller to allow flexible presets and switching via midi or bluetooth. 

This brings me to the question at hand:  how best can 20 component values be switched electronically?  Would it make more sense to build two filters onto each coil so that only 5 switches need to be flipped each time?  Or, am I simply missing a more obvious and simple solution?

Ditch the manual design and go uC controlled.

[Gus]

Wow

free design request for a commercial product?
Will it pass UL and CE?

[HD Evans]

Yeah, finding a sweet spot for wire gauge that's fills a traditional bobbin to avoid making things even more complicated has been a challenge.  Noise is looking alright thus far using better op-amps.  Otherwise we would be looking into matching transformers and still might go that way.  The design team consists of three engineers, including myself, who are happy to tune everything until it's perfect.

Thanks for chiming in RG, I was hoping you would. 

As for UL listings, they're basically opt-in and have become a huge pain in the ass over the past decade.  Just because I want to share a design with the community while making it readily available doesn't mean there's some urgent need to deal with regulatory groups if they aren't required by law.

[amptramp]

I thought one of the reasons wall warts are so popular is that they can be obtained as UL, CSA, VDE or whatever is required and the rest of the design does not have to change.  At one time, you could get wall wart cases with the plug-in pins installed and put your own circuitry in it.  I built a DC power supply for battery radios that way.  Sayal stopped carrying these several years ago.  Maybe there was a concern that people could build unsafe circuitry into a wall wart and make it look like safe circuitry.

For the other question of filter changes, you could use switched-capacitor filters which can implement a change in pole frequency just by changing oscillator frequency.  You may not be able to change Q as easily, but you may not need to.  A use for switched resistor filters is the Paul Nelson phaser that uses CD4066 sections to control filtering, eliminating the need to match FET's.  This uses duty cycle modulation of a bank of CD4066 sections to simulate variations in a resistor.  Control with a processor is the most generalized answer, but maybe for what you are trying to do, you don't need it.

[R.G.]

As for UL listings, they're basically opt-in and have become a huge pain in the ass over the past decade.  Just because I want to share a design with the community while making it readily available doesn't mean there's some urgent need to deal with regulatory groups if they aren't required by law.

Actually, they were a huge PITA since at least the 60s; they are considerably less so since about the 90s, when many countries let you self certify and repealed the provisions that made you a criminal if you didn't use the state testing labs.

Safety standards are opt-in in the USA. But if you plug it into a wall socket, and don't have the first thing it hits be a certified wall wart, the damage awards will be much bigger if anyone gets hurt because the lawyers will claim lack of due dilligence.

CE is NOT opt-in in the EU. If you sell in the EU, you're allowed to self-certify if you want, and are barred from selling in the whole EU if anyone catches you not meeting it. In other countries, it may be criminal to not certify (and get caught).

If your unit plugs into the wall socket and does not use a certified wall wart, think long and hard about not certifying.

The only thing more of a PITA than certifying would be paying damages and fines, or going to jail.

"Sharing a design with a community" doesn't necessarily free you of the legal onus. If your design doesn't include reasonable (i.e., a jury thinks so) cautions against self damage, the parents of some kid who hurt himself or burned down the house following the design, even wrong, will haul you into court. This can be expensive, if very educational.

[HD Evans]

EU standards never cease to amaze me.  In the adhesives industry I have mostly dealt with chemical regulations, where UN involvement and "global harmonization" will result is MSDS in excess of 50 pages for non-toxic water-based wood glues.  Thanks for the heads up...  As always I will protect what I can behind my LLC and get expert advice for the rest.

Everything thus far is designed to be on-board and battery powered.  I don't know if the super flexible stomp box with 30 knobs will ever go beyond a project for myself and anyone sufficiently warped to build one for themselves.  To go along with that thought:  I've only been using a SVF design because of my desire to fine tune every pickup in my guitar...  But now that I'm playing with the more user friendly tone modules as described above...  Is there any reason not to scale things back in the interest of simplicity and noise reduction?  Hopefully none of the tones would push the limits of circuit stability in the course of copying the 2nd order roll off of a passive pickup?