4 pole mission, Please help i can't sort it
  • Hello,
    This is my first DIY project, was pretty confident but i am facing a problem i can’t resolve. Liquid and Wobbly flavors are sounding exactly the same, MS ans Whacky are sounding the same too. I checked everything, all in place, checked the soldering and everything is OK too.

    I bought the cleartune app from appstore on my iPhone to calibrate it and can’t understand how to do that, i am using iRig to plug synth into my iPhone. When i turn the V/Oct trimmer i can see the note of the resonance changing and When i turn the Range trimmer all i hear is a change in pitch going up or down the further i turn.

    Don’t know what to do, please can somebody help me ?

    Thank you.

  • Something to add, the resonance is working good, fiter cutoff too.

  • You should hear a difference between liquid and wobbly only when the resonance is very high, close to the maximum value. Check the soldering of the 4053 and of the 47k resistor.

    Regarding tuning, have you seen the V/Oct tuning section from the manual?

  • I just checked the soldering of 4053 and the 47k resistor, they are okay. Even when the resonance is set to max i can barely notice any difference. And whacky is exactly the same as MS, no wobbly introduced.

    I have seen the manual yes, sorry for the question.

  • If you have a scope, check what happens on pin 4 of the 4053 when you switch from liquid to wobbly. You can also probe this point with a cable hooked to the tip of a jack.

  • I have similar behaviour with mine. I thought it was just the subtlety of the different modes, that I wasn’t noticing. I will be interested to see what the resolution is.

  • You are welcome fleetway76. Anyway, i didn’t expect i had to do that kind of troubleshooting for a first kit. I feel lost. Can i ask where is that pin 4 on the 4053 ? And probing with a jack means just putting the tip on that pin, grounding the sleeve and listen on my headphones ?

  • I had a look at the datasheet for the 4053 so I think I see what it is trying to do. I will get my scope out over the weekend to see if I can track down the issue. I guess either the 1 pole output isn’t getting to pin 3, the logic signal isn’t getting to pin 9, or the chip is kaput (or something further along the chain). Should be easy to test either way,

  • Thanks freetway, waiting for your input :)

  • Never connect headphones, not only this is disrupting the circuit, but you can damage your hearing.

    Set the volume of mixer/amp to 0, probe the circuit, and gradually increase volume.

  • How do you mean never connect headphones? When probing the circuit?

  • It seems to me that busygirl is trying to directly to hook up headphones to circuit points. DON’T DO THAT.

    First, headphones behave like a small resistor to ground. There are points at while this will load the circuit and make it behave very differently from what it would be with the headphones not connected – this is not what you want. There are points at which it might damage a chip, because some op-amps can only output a small current without heating too much. You want the probe to be connected to a high-impedance input (most soundcards or mixer have “high-Z” switches to raise the impedance of the input, but even without those, the input impedance are above 10k and this is enough to cause only a minor load change in the circuit).

    Secondly, some headphones don’t like high voltages. Some points on the circuit, especially if it’s not working correctly, might be at +5V or so and this large DC shift might damage the headphones.

    Finally, your ears, please be careful about your ears.

  • Right, Yeah! By probe I mean oscilloscope. It would be hard to debug the system without one in this case.

  • I havent had a chance to check this yet, but persumably carefully shorting pins 3 and 4 of the 4053 would provide a quick test to rule out problems in the audio path.

  • When i short those two pins the resonance disappear.

  • Hmmm, does it look like there are shorts near the 4053?

  • So far I can tell that my chip 4053 appears to be working. pulling pin 9 high seems to close the switch. I see about 0.4v ac on pins 3 and 4 when the switch is closed. I can detect a very subtle change to the resonance characteristics when the wobbly mode is selected, but nothing like the guy who put a demo on youtube. Is there a particular patch that is good for demonstrating the resonance characteristics? Specific waveforms or whatever? I am going to hook up the oscilloscope and investigate further.

  • Hmm. I am kind of at a loss. I get all of the connectivity I expect, and my scope shows the lp1 waveform at the south side of the 47 k resistor. It all seems to check out, but I dont hear the resonance change markedly. I can get it to sound slightly different with different settings of cutoff and resonance, but I have to search for them. My scope shows the waveforms at the output look nearly identical between liquid and wobbly. Any ideas?

  • That’s because i have seen that video on youtube about those mode at the end of it. I expected a lot of change in the sound, but i barely notice it. Wobbly is a little bit more gritty, and Whacky, i wonder how it is any different from MS, the manual says it is a mix of wobbly and MS i believe..

    I verified every joint in the whole board, i didn’t see anything wrong seriously. I will check again around 4053.

    But look at this video at the end when he presents the flavors of the synth:
    http://www.youtube.com/watch?v=rSjlwYc7hr0

    Is it real ?

  • You could try putting a smaller resistor there (say 22k) for a more intense effect.

  • Instead of the 47k one ? I Will try.

  • let me know how that goes

  • Okay it did the trick really. i did put a 16,5k one because i didn’t have a 22k. Each flavor is very different now. Wobbly is full of character and Whacky, from patch to patch, can catch the Wobbly in the mix more or less. But FAR better now.

  • I wonder, did this guy in the video change the resistor for a smaller one too or is it expected to work like that with the 47k one ?

  • What I heard in the video seemed pretty consistent with what I have built.

  • Weird. I guess it might just be normal variance in the components, but I wouldn’t have thought that it would have made that much difference.

  • I don’t think it’s part variance – I still think there’s something wrong with your board… but what?

  • The only thing I can think is the level of the LP1 output is not as high on our boards relative to the rest of the signals feeding into the resonance opamp. I think my meter was showing about 0.58v ac. I used the bom from reichelt but I subbed a few caps with ones from farnell to get yellow dipped mlcc rather than the mulitcoloured reichelt ones,

  • Perhaps the normal resonance feedback is weaker for some reason. I there anything I could check there? It seems unlikely as I was able to get it to self oscillate to set the scaling,

  • I don’t know much about electronics, but i can say i did everything shown in the instruction assembly. And if something isn’t working, it isn’t about my soldering or mismatched component. I thought at first i made something wrong until i checked a thousand times. Maybe some component is dead somewhere ?

  • Could this be something to do with the 2164 chip? It seems to be controlling the frequency of the cutoff. I guess our problem is more related to the amplitude of the signals rather than the frequency though. Could it be a bad pin connection in one of the op-amps? I used machined sockets for the opamps rather than pressed pin ones. I know some people recommend against using any sockets for them. I might try popping in different TL072s when I can be bothered to open it up again. Just out of interest, The progression from 4 pole to 1 pole doesn’t sound exactly smooth somehow. It is like the 3 pole setting sounds kind of ‘out of place’. The thing is, even with only 2 different modes of resonance, it still sounds great! It is only really my feeling that something might be wrong in there that is bugging me. If anyone has any other suggestions I would be glad to hear them

  • At this point, recordings of the “init” patch at each pole would help ; or scope traces.

    The SSM2164 doesn’t control the filter cutoff, it is the filter.

  • I will put these together when I get a chance. Do you mean recordings of the various lp 1-4 flavours or recordings from probing the test points on the board?

  • I have put back the 47k resistor in place, i can notice that i have to push the res and the cutoff at max to hear the very high frequencies change between liquid and wobbly.

    Waiting for your input fleetway :)

  • i think he means sound samples or scope shots from the test points

  • I am not near my shruthi, but looking at the schematic, something dawned on me. I missed what was happening. I thought that the wobbly mode switched the 1 pole output into the actual resonance feedback loop, but now I see that it is actually modulating the cutoff.

    When I was probing with my scope I checked the output of IC2B and it looked very noisy. Would you expect the output to be constant (or at least varying at audio rate)? Perhaps something is getting the opamp at IC2B to self oscillate and masking the effect of the LP1 FM.

    Perhaps dropping another TL074 in there would affect things.

  • Perhaps it is an unfortunate interaction between the values of the scale and range trimmers that we have. They seem to be the other inputs to the cutoff value.

  • I guess it could even be noise on the negative rail from somewhere. Sorry, this is turning into a bit of a brain dump. I will stop now!

  • I wil try with another TL074.

  • I would only try one if you have one available, dont go out and buy one. I could well be totally wrong.

  • @fleetway76: The output of IC2B has some traces of the PWM carrier at 39kHz, this is normal. I don’t see how tuning settings could have an impact on that. I don’t believe in the supply noise theory either.

  • Yeah, I was getting a bit carried away with the noise on the power supply idea. I forgot about the noise suppression and decoupling caps all over the place. I am just mystified that I could see the LP1 signal at the summing junction, but not at the opamp output. Its very strange. The tuning settings only came into my head as they are the other components that feed into the CV summing node.

  • Hi, i replaced both 4051 and 4053, nothing better. I Will try another 47k resistor later today.
    By the way should i hear something when i turn the range trimmer ? If i understand correctly it should affect the cutoff range right ? I hear a very high pitched noise going up or down when res and cutoff at max, as i turn the trimmer.
    Wonder if this can point me in the right direction.

  • What you ear when turn the range trimmer depends on the patch programmed on the synth.

    Turning this trimmer is equivalent to increasing / decreasing cutoff by a few units. Nothing radical – and this is why not tuning the filter is not such a big deal. It’s not like there’s a combination of trimmer settings that make the filter work, and another combination of trimmer settings that break it…

    During the filter tuning procedure, the synth must be programmed to play a pure sine wave from the filter self-oscillation. I am not sure why you put cutoff and resonance to the max for the trimming procedure, as this would put the sine wave close to the upper limit of the audio range. Stay with cutoff = 64 as per the instructions.

  • Okay thank you for the info :) i Will start from scratch, ordered a new pair of Pcb :)

  • there’s no way back now :)
    welcome to the wonderful lifestyle of polyshruthi-ism

  • FWIW I ordered a ‘real’ (hopefuly!) SSM2164 from some dude on ebay. When I get it and slice open my Shruthi to fit it, I will do a bit more probing and record the levels and take traces from various points. Now I have got my head around what is meant to be happening, I should be able to gather more useful data.

  • I have never observed any difference between SSM2164s and V2164s. It’s not “cloning” or whatever, it’s legit second-sourcing.

  • Ah but it will have an analog devices logo on it!

    Seriously though, thats a bummer. I thought it might have had something to do with it. I will stick it in there anyway and see what happens. Probably swap some of the opamps out just to see if it makes any difference. I suspect it wont, but if I am in there anyway…

  • If anything the older vintage ICs will be less reliable. They had to do everything by hand back then from the chip layouts to checking for overlap.

  • SSM2164s are not “vintage” ICs.