SMR 4 Filter tuning feedback

some feedback about the filter tuning :

i first made a average tuning with the tips on the website.
and refined them afterwards.
my filter seems to be tuned OK, but i would like your advice :

with no soud, and no modulation to the filter :

i first tuned the resonnance pot >

no auto-oscillation at 60
starts after a few seconds at 61
almost (1 second) immediate at 62
and immediate at 63
(on my second shruthi, no oscillation at 61 too, same for the rest)

my CV/oct is tuned OK (auto-oscillation follow note pitch on the whole keyboard)

i activate 1 oscillator (for example : osc 1 = square)
i match the pitch of the note played on the keyboard with auto-oscillation at cutoff 35 by ajusting the low trimmer.

this is a compromise beetween not having the Loww setting too high (HF noise when cutoff 127) and a good range.
but maybe a bit high

i turn off the oscillator
with these settings, i have auto oscillation at 40hz with :
cutoff : 0
resonnance : 127
(playing any key under C2)

let me know if this sounds right to you :wink:

> i activate 1 oscillator (for example : osc 1 = square)
> i match the pitch of the note played on the keyboard with auto-oscillation at cutoff 35 by ajusting the low trimmer.

i was wondering about this when i tuned the filters on my two ssm boards: it would be useful to have a convention about at what cutoff setting the cutoff frequency matches the oscillators’. as long as you only have one shruthi and programm you patches only for that single shruthi, it doesn’t matter. but as soon as you exchange patches with other shruthis (or in our still somewhat underused patch sharing thread), a convention for this would be helpful, because otherwise patches with high filter resonance will sound very differently on different shruthis.

but how did you arrive at cutoff = 35? that seems pretty low to me.

i agree :wink:

cutoff @ 35 matching osc pitch with full resonance means matching the first harmonic…
so without resonance you have almost no sound but something like a sinewave.

at first i set this with cutoff @ 1 matching the pitch of the sound, but that was way too low, so i played with the sounds and the filter (just programming sounds) and find a setting that useable.
and it appears that it was 35

by the way, the tracking has to be properly set on the whole keyboard first - i really spend a lot of time on that :wink:
that can influence how the low setting will sound (maybe unclear here - you get me ?)

with my two ssm shruthis, cutoff freq = oscillator pitch (with osc range = 0) is at cutoff = 56.
with my cem shruthi - which i tuned only once after completion and haven’t retuned yet to match the ssm ones - it’s at cutoff = 60

so this shows that currently, shruthi users tune their synths rather differently. :slight_smile:

@pichenettes: how do you handle this? at what cutoff value does the vcf cutoff frequency match that of the oscillator (rng = 0) on the shruthis that you sell as fully assembeled units? for what filter tuning are the shruthi’s factory presets meant?

> @pichenettes: how do you handle this? at what cutoff value does the vcf cutoff frequency match that of the oscillator (rng = 0) on the shruthis that you sell as fully assembeled units?

It depends on the filter board. Depending on the tolerances of the caps used and the filter board, the filter may or may not “crash” at high cutoff / resonance. I usually trim the range trimpot so that the HF crash zone is avoided ; so it is never exactly the same. Factory presets come from different people and different machines, some of them were even programmed on a SSM2044 filter board - I regret there’s no “right answer”.

What I have learned is that I should get rid of trimpots in future projects ;D

noooo don’t! trimming some trimpots is a great way of spending your time, especially if you have ocd! :smiley:

Too late!

aaaaaargh - only one trimpot! how am i going to spend my evenings now?!

Tuning the contrast of the LCD :slight_smile:

naah, that’s not the same.

i’m afraid i’m going to have to find me another diy project to obsess over. hm, maybe an open source laser cannon might be the right thing for me. i’m sure there’s a lot of fine tuning to do on that one. pew pew!

you can bet your ass there is on a laser…

@pichenettes
You give us just one Trimmer? You are a penny-pincher!

hello guys :wink:

i know you are all fed up with those “filter tuning” threads, sorry for making that one more.

with my tuning settings, i still feel when having the cutoff at 127 that i get this HF noise, highs are pretty “messy” and harsh with some of the waveforms (PWM for example)

That’s not anoying with a lot of sounds, usefull on percussive sound (pizz / pluck kind of get an edge you can’t get anywhere else) but that’s bothering me with some of them

I just wanted to know if i’ve done something wrong somewhere… or if it’s normal

anymay i’m really happy with my two shruthi, the wavetables rocks :wink:
and the whole thing is quite powerfull :wink:

This has nothing to do with the filter tuning, it’s aliasing.

Only square / saw / triangle are band-limited.

Don’t mean to hijack this thread but it seems silly to start another since its about the same filter board tuning.

I tuned my first SMR-4 and I thought I did a good job, but now I notice that my resonance acts like a volume knob. When I turn it down the sound is completely gone. When I turn it up a little bit the sounds comes up with little resonance but also little volume.

Have I made an obvious mistake in the tuning? How can I balance out the oscillation volume with the low resonance volume?

It’s not related to tuning. Maybe I should put that in the build instructions but even if you don’t tune the filter, it’ll work as a expected - not optimally, but it’ll work.

You probably forgot to add the poles selector jumper, or have a weak solder point somewhere in the filter. I bet that no signal is going through the whole filter core ; the only thing you hear is the loudness compensation (a fraction of the original signal is sent to the output as resonance increases, to compensate the annoying loudness drop of 4 poles filter).

Ah thanks I think I found it! The jumper was there, but it appears I forgot to solder 2 of the 3 pins of the pole selector :slight_smile:

Yep that was it. Works like a charm now. The output is a bit louder without resonance, but I guess that’s normal.

Yes, this is normal, the ideal compensation curve is not exactly linear.