Tuning the SMR-4 filter board

I had a bad cap on my shruthi, the power supply, so before really getting in to tuning it, I replaced both 220s, to make sure I have clean power.

I spent some time tuning today, and it really does sound better… my samples are closer to the ones posted on the web site than ever before. I still think it could sound better though.

Is there a video of someone tuning the shruthi? I think this would be very useful. Start with the “self-test”, if all goes well move on to showing someone turning the pots, with audio of what to listen for. I know this might seem really simple to someone with more experience, but for the “first synth” crowd it could be really useful, and will ensure that more shruthis out in the wild, really do sound their best.

While I’m mainly concerned with the SMR-4 filter board, this would be great for all the filter boards.

The epiano patch is velocity sensitive and must be played very hard. I think I had my keyboard set to a concave velocity curve because in retrospect, you really have to hammer it to play it fine. It works better with cutoff set to 22 or 23.

As for the trimmer it doesn’t make sense to give a value because it may vary a bit depending on the value of the filter capacitors. And it really help to have all these turn, since the sweet spot for resonance and for V/Oct is on a half turn.

This is something I have to do again too, as I don’t think I quite nailed it first time around.
I guess if the trimmers were the normal cheapies, it may have been a little easier, as it would be possible to specify a vague guestimate of “If in doubt, set it at 3 o’clock” for example. Maybe a rough turns count for the current multi turn trimmers could work instead, but those clicks are hard to feel/hear…

Another thought, is there perhaps an approximate resistance that can be measured across them in circuit that could help give us something to aim for too?

Anyway, I think a little more time spent doing it by ear will pay off, so i’ll get to it soon!

I think you need precise pots like the ones in the kit. For example, when tuning reso, I go from silent, to tone, to distortion, in about 1/4 of a turn. It took me a while to figure out what I was looking for however… since I had never done this before. I think it sounds great now, most of my patches are pretty close to the pre-loaded patches.

I went through the preloaded patches again, most of mine sound good however, when I play sample #22 piano… I get little to no sound below E4… so I think my cutoff is wrong(I think)… and I’m not sure how to set this one up now. Should I just load that sample and turn lowwww pot till I get sound on this patch? The directions for tuning this pot, were a bit more subjective.

Reso has me a bit confused. I keep turning it up until it starts to resonate, but it gets carried away with itself pretty quickly, so then I have to back off the trimmer by quite a long way, like long before the point it started resonating in the first place. Meaning im not sure where it was supposed to be left.

The loww trimmer setting was a best guess too, as I couldn’t seem to hear what it was doing very well. Anyway, i’ll give it another go, and hopefully get closer to the mark… Im sure i’ll get there eventually.

Re: normal trimmers and multiturns… Multi’s are good for settings that need to be 100% on the mark. Like tuning oscillators (I wish I had them in my x0xb0xes), and setting very precise voltages. But the Shruthi filter boards appear to be a little more free than that. Still, the multi’s are better quality, thats for sure.

adjusting the cutoff to 23 get me sound down to C4, even when i really bang on the keys, below C4 I’m just getting a soft plop, and a little ringing…

Should I adjust the cutoff (lowwww) or just keep it as is? I’m pretty close on all the other patches… I think this will be the last adjustment I need… better yet, should I adjust the lowwww pot with the patch set to epiano, #22?

Yes, use epiano (with cutoff to 23) and trim it in such a way that you don’t have to bang the keys like hell to get something. Note that this patch is meant to sound like a 80s piano for notes above C4.

So the plops, and lack of sound below C4 is expected for this patch? If so I’m done… all the other patches sound sweet… and really close to the samples on the website… at least by my ear.

Seems to sound better when I jack the cutoff to ~80

and #55(NES bass) is not putting out any low notes either… cutoff is very high on this one… and the notes that do sound are totally distorted. adjust the pot?

80? When I set the cutoff to 80 on this one it sounds like an horrible gameboy sound. Weird…

First with cutoff = 22 (good sound) ; then with cutoff = 80 (does it sound like an e-piano? Hell no! ; then moving back form cutoff = 80 to 22

thanks oliver, that is pretty much what mine sounds like, though mine is a bit on the quiet side, and silent left of C4.

maybe I’m obsessing over tuning the cutoff… it took me so long to get the res pot right.

I’m going to post a few samples too, that should clear it it, maybe I’m just being crazy.

Oh dear, this is getting depressing! I retuned mine, and if anything it might be worse. The e-piano test sounds pretty rough compared to Oliviers demo :confused:

Have a listen?
First half is with filter at 23. Second half, filter is increased to 80 (Mind your ears!) and wound back down to 23…
http://tinyurl.com/epiano

This doesn’t sound that bad…

sounds much better than mine

ok here is mine… with cutoff at 23, like said only getting sound from the right of C4.

http://soundcloud.com/bitracer/epiano

better yet… here a a few samples… and I’m not sure if I’ve just been tinkering too long, and I need a break… or this actually sounds great?

some stuff simply does not provide much sound in the low range… I’m guessing that is normal?

It really looks like there’s something wrong… On some samples there’s some clipping (try lowering the gain on your soundcard) ; on some samples there seems to be some heavy aliasing (hint that the output signal is not as smooth as it should be - maybe one of the 33nF CV smoothing cap badly soldered?). I don’t notice anything wrong in the low end. Does it sound like that only during recording or does it sound better when played straight to monitors?

I’m not recording the samples I post online on some magical hardware, just a FA-66…

that is what it sounds like, and my gain is definitely all the way down. All the 33nf caps are soldered in nicely… though I’m willing to replace them all if you think that will help?

Is it possible I damaged something when I had C29 in backwards? I replaced both 220s with new caps to be safe.

I’ll put the board under a magnifying glass tomorrow and check every single solder joint… but from what I can see it all looks good.

Thanks for the feedback on mine. I guess i’ll leave it alone now then. Although to my ears it appeared to sound different. It seemed to go from ok but quiet, to loud and shrill.
Anyway! I get to do it all over again soon when I nail together my SSM board.
Im looking forward to hearing how it sounds compared with the SMR-4.

oliver, I’m really stumped, I just went through every solder joint on the board… everything looks good, still the board sounds bad…

I checked every resistor, every cap, I went through and removed every chip, and re-seated them in the sockets
I re-checked the power, its spot on, -5V, 0V, and 5.02V at each pin.

I’m not sure how to troubleshoot, I don’t have an o-scope. Would this be sufficient?

I’ll replace all the caps… with really good ones if you can tell me where to source them.

Also, I’m assuming that messing with the four pots on the board could not be the problem here? I’m pretty sure R3, R23, and R35 are correct… R34 I’m not sure, is it possible I’ve over adjusted them.

Is it possible I just have a bad part… what should I look at next?