MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

AS3320 (CEM3320 clone) input voltages
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author AS3320 (CEM3320 clone) input voltages
twigathy
Hello! I'm starting to design PCBs based around some of the reference designs on datasheets as a way to learn a bit more about electronics, and hopefully make some interesting noises too. So far I've got a triple passive attenuator (Decent enough way to learn Kicad... think at-at-at but on a PCB) and a very simple four-channel drone thing mapped out as a PCB, based on this: https://www.lookmumnocomputer.com/simplest-oscillator. Eventually, I hope to get PCBs made, the modules tested, and maybe throw the designs up on github for anyone to download and pass to their favourite PCB fabbing place.

Anyway. now it's come to something a bit more complicated, the AS3320. Using the data sheet over here (https://electricdruid.net/wp-content/uploads/2017/06/AS3320.pdf) I mapped out a circuit but I have some silly questions for you lovely DIY people that maybe someone can help me out with thumbs up

1. AS3320 seems to want +/- 6v on the frequency control, and +2/-18v on resonance control at the very most. How do I ensure that CV signals coming in meet these standards? Is this something to build in to a design, or do I just have to remember to not plug super-hot-CV-signal into this? The resonance control especially worries me smile
2. I've noticed quite a few modules include separate control knobs as well as CV, how is that usually implemented? Can I just use a really dumb voltage divider (Pair of resistors?) and a potentiometer to go from 12v to +/-2.5v for example?
3. Do people have a preference for "Plugging in CV overrides the manual pot control" vs. "CV and pot are summed"? I figure both might be useful...? How does this usually get implemented with an mono audio jack with an earth and signal pad? (Or is the magical "TN" pad on the PJ398SM/THONKICONN good for something after all?)
4. The AS3320 datasheet reference design seems to only be a low pass filter. It seems like these might work for other flavours of filter: https://electricdruid.net/cem3320-filter-designs/ ? (I'll probably make three different boards with low/high/band pass, or go crazy and try to fit https://electricdruid.net/wp-content/uploads/2015/06/SynthexMultimodeF ilter.gif onto a single PCB smile)
5. Swapped power connectors - is there any way to really easily combat that? I assume it's better to cook some poor cheap component than a whole module if power cables are hooked up the wrong way around ... but how do people usually do this?

Anyway. here's what I've got so far. It's pretty rough and still needs some more love before I'm ready to print it out and test fit parts, which for various reasons I can't do for a couple of weeks anyway!


I haven't even thought about panels yet. That's a whole other problem for future me d'oh!
DozenCrows
1. If you want to have other people build or use this then build limitations into the CV circuitry to ensure the voltage ranges are right. If it is just for your own personal use, you could not do this - but you might forget in the future and harm the IC. Also sounds like you'll learn some useful stuff from building that in.

2. You can add a manual control to a CV input as you describe (pot + voltage divider). The recommendation is to use voltage references (e.g. LM4040) as the source voltages rather than the power rails, for stability and accuracy. Depending on where the CV goes, you may just be able to connect it directly (e.g. if it goes into a virtual ground as on an inverting summing amplifier) or you may need to buffer it first.

3. This depends on what CV control behaviour is required, IMO. It's useful to sum them as you can then set a base level via the pot, and modulate it through the CV jack. That base level can be set to zero if you just want the CV jack input to drive the control. BTW that extra pin on the Thonkiconn connects to the input when no jack is in the socket - a "switched jack". This is what is used for normalisation.

4. Yes, there are other modes that work as per the article you link. I suggest you start with a single mode and get that working first - easier to debug.

5. Have a search here on MW - there are some excellent threads that go into reverse power protection. Common solution are in-series diodes or MOSFETs.
DozenCrows
The full 3320 datasheet goes into some detail on the resonance control, which is actually a current - see http://www.bustedgear.com/images/datasheets/CEM3320.pdf pages 4 to 5.
jorg
You need to select input resistors (or resistive dividers) that will drive the correct voltage or current into the chip, when fed "normal" modular voltages. Then you design protection circuits (e.g. diodes) which will prevent the inputs from seeing voltages or currents in excess of the absolute maximum ratings.

On CEM / AS datasheets, the sample circuits typically assume voltages that don't resemble modular norms such as Euro. Tailoring the resistors to match your target system is up to you. They couldn't possibly know, back in 1979 or so, that there would be a Euro format with typical CV in the range of 0 to +5, or 0 to +10, or -5 to +5, or whatever seems to be the most common today. And of course there are many other formats and voltage norms out there, not to mention non-modular applications which will always use whatever is convenient.
twigathy
Brilliant, thanks both! I'm going to revise the schematic & PCB with some pots and get power going to CV inputs as well as attenuators for in and out (May as well). Then figure out the reverse-polarity/misplugged power connector circuitry. It'd be nice to be able to design the power input part exactly once and copy it over to other PCBs in the future smile
twigathy
If I've got this right... 330ohm resistor + 5v reference source + 12v input -> 21mA going to ground.

Tomorrow will be figuring out what the current controlled resonance input should look like. A little more searching suggests an opamp might be the way to go - something along these lines maybe - https://wiki.analog.com/university/courses/electronics/text/chapter-4# precision_voltage_to_current_converter

DozenCrows
Your 5V based CV pot circuit looks rather strange to me. I would connect the 5V to pin 3, and use pin 2 as the output CV. The pot will then give you the expected 0V to 5V range.

Then connecting that to the tip pin of the jack may also not work how you expect - you want to sum the pot CV with the jack CV, so you could use a summing amplifier circuit with an op-amp - e.g. https://www.electronics-tutorials.ws/opamp/opamp_4.html.
DozenCrows
If you want an example of a frequency CV input to the 3320, take a look at the schematic in this article: http://www.synthdiy.com/files/2002/MIF.pdf. The bottom left corner has a mix of pots and jack inputs feeding an op-amp which is then connected to pin 12 of the 3320.

This is an inverting summing amplifier, as increasing the voltage on pin 12 actually lowers the filter cutoff frequency - so it needs to work in reverse here if you want higher input CV to increase cutoff frequency.

The trimmer R5 can be used to offset the frequency CV, e.g. to get it into a desired range - such as -25mV to +155mV (as per the 3320 datasheet) for CV of 0V to 10V respectively.
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group