These informations are all over the place, so let’s summarise them here.
“Mandatory Fix” (even if your module seems to work):
replace R42 (220R) with a resistor between 1K and 4K7. With 220R the LM4040 is way out of its maximum rating and causes wobbling of the pitch, or it could just be destroyed. Look here
Easy (no excuses to not do them if you build a new 1222):
replace the L7805 with a L7808 because the Arduino is powered thru Vin not +5V.
add a resistor between pins 4&5 of the 3340 to reduce the ringing on the square out. It may affect the PWM pots range… look here. Some sources recommend 10M but some recommend as low as 1M.
Use a low temperature coefficient capacitor for C6. The BOM specifies ceramic, but that can lead to serious temperature sensitivity. Instead use a polyester film cap, or even better, polypropylene (with a temperature coefficient of about -200 ppm/K).
Possible improvements:
adding 1K output resistors (not absolutely necessary but “good practice”).
fixing the discrepancies between output levels and DC offsets of the different waveforms.
adding a regulator for the negative supply of the 3340 ?
C5 is on the wrong side of R6.
fixes to reduce PWM detuning
Comments:
An improved jack PCB could add the 1K output Rs, make all waveforms output levels 10Vpp, and provide the negative supply for the 3340.
GND and the eight jack signals are already on the existing connector.
We’ll need to find an easy place to pick up +/-12V to power the OpAmps and the negative -5V supply.
The Eurorack power connector is just in the right place Let’s pick +/-12V right on the leads of the ferrites/10Rs/schottkys/whatever you mount there.
The negative supply can connect to a pin soldered in place of one of the legs of R6 (which will be left out).
It should be possible to bodge C5 on the other side of the PCB. I didn’t try this so it may interfere with the components already on that side.
PWM detuning:
That’s a bad one to fix
There is this. Not easy to bodge into an existing PCB.
The AS3340-HYB is a small DIP-sized PCB which is supposed to be a drop-in replacement for the AS3340 with much reduced PWM de-tuning effect.
Please comment if you have tried it.
There is also a discussion about adding a trimpot but I can’t find it right away.
Ideas/suggestions/whatever to improve this post are welcome.
Especially someone who want to implement the improved jack PCB idea
I like how the Kassutronics 3340 VCO handles the outputs (with a sin output added). But I like Sam’s inputs and controls better. But more range on the octave switch would be even better.
So I’m working on a hybrid design, so far only on paper. Well, SSD. Won’t get done soon but it’s on my list.
Once I have my three #1222 working I won’t make another 3340 based VCO for some time…
Too much hype, average results, not worth the trouble…
Next VCO on the list is Braids… but only after I have a full working “LMNC Voice/BasicPatch” (VCOs/Mixer/Filter/VCA/ADSR/LFO/Mult) and some additional modules (very-simple-MIDI2CV/filters/mixers/etc) to be able to make good use of Braids. Still I add a few of the components to every order I place
If anyone is interested here is level adjustment for all waves, with additional pulse wave reshaper to make it a lot sharper. Trimpots are used to recenter the wave perfectly at 0, as it always bothers me if its slightly off. You can remove the trimpots and replace them with ~242.2K resistor if you don’t care about that.
And here is sine shaper with commonly available parts:
In the following month or two I will release a new module that uses lots of good ideas from other designs with some extra bells and whistles. Like LFO and wave-folder:
I tried setting up something like that a while back but the wiki editor was too confusing for me. I liked the old days when you just used the simple Wikipedia syntax. Everybody thinks they can make an “easier” wiki editor with the result that I’m confused as hell because nothing works right.
I read every post on this forum. Feel free to tell me if you want something made a wiki. Although, many users with enough trust level have been able to do so. Discourse automatically grants higher levels of access over time, so as you stay using it, you will be able to if you are not right now.
hey! cool good information! sounds like a mk2 with the rotary pot on pcb needs to be on the cards.
there is a mixer output for the link port coming that deals with the offsets but also doesn’t have to incase a mk2 comes along. so ok here goes!
would be cool making it the same looking from the front ill just change up the back and also add a sine wave (I never use one… but lots of people ask for it!)
ill get on it next week. ill post up a rough adjusted schematic… thingymajiggy.
regarding Pwm may do it with an external comparator? as it seems there is a opamp left in my list … 2 tl074’s 1 tl072. this may negate a few components (always a fan of keeping component count low)
if a sine is added there is space for 1 jack socket… choices?
EDIT “”"" So after having a bit of a think. maybe a sub output just a square wave 1 below could be good. Or because there is scope and room for 2 knobs still with lots of space. maybe even a wave shaping output or a morphing output (however between standard waveforms the outcome in my opinion isn’t that interesting, so wave shaping could be funky)
Let’s pick +/-12V right on the leads of the ferrites/10Rs/schottkys/whatever you mount there.
