Anushri troubleshooting
  • A little surprise, when I fired up the Anushri in the lunchbox, it did not work anymore! To be specific, I get a very low level signal from the phone and 3.5mm jacks that disappears at low volume pot settings, comes in at medium pot settings and then becomes only a thump with the volume pot fully CW. Any hints as to where I might start looking?

    Thanks

  • Voltage checks?

  • Getting to that… Kinda bugs me that it worked fine, I must’ve hit something.

  • @piscione: I don’t think you have a cold solder joint on your DIY bus board, but there may be a short or grounding issue with the plugs your Anushri is attached to. Have you tested it in another plug on your bus board?

    Looking at the picture of the back you posted in the census thread, you may have a bridged solder blob on the rightmost most connector. Just right of the blue wire connecting the top 4 or 6 pins. That is the one you have your Anushri plugged into. It could just be a reflection in the picture though.

  • @audiohoarder, thanks for the reply. I’ve been using it without the box, just plugging the Anushri directly into my modular rack with another ribbon connector. Olivier, all of the voltage points in the Troubleshooting section check out fine. I’m going to try the audio connections next.

  • I started checking the audio points but noticed the 7905 is too hot to touch for long and the 7805 is too hot to touch at all. All of the voltage points still check out fine. I’m powering this from the Euro connector.

  • Voltage, yes, everything is fine. Audio, partially, that’s a bit trickier.

  • At which point in the signal chain things go wrong?

  • This just got more interesting. I fooled with it this morning and I was getting sound as before but very faint, and if I raised the volume, all I got was a thump or click. I put the front panel on so I could see what I was doing, and I was going to take some video. It started working again! I’m going to re-flow all of the solder joints and see what happens.

  • When you turn up the volume it sounds like the gate is leaking into the audio path. I’d look for discontinuities around the gates and envelopes first. Then look around the volume pot.
    I’ve had some issues with my Shruthi power supply, so maybe try another one?
    Before you racked your Anushri did it use a 12V power adapter?

  • @audiohoarder, which ICs are the gates and envelopes? I’ve checked the volume pot, considering replacing it just in case. I did not try a 12V power supply, this Anushri went into the rack from the start.

  • > @audiohoarder, which ICs are the gates and envelopes?

    Surprisingly, none, it’s all digital :)

  • Darn those digitals!

  • There is only one way of solving that: check the signal path point by point until you find at which point the signal chain is broken or the control signals are garbage. This narrows down the problem to 1-2 ICs and at most a dozen passives.

  • I can’t seem to get any sound out of any of the points on the PCB but I did notice something odd, just by, uh, luck. I had the thing plugged in yet again to see what it was doing, and there is a 100v 4.7uf cap near the non-euro power supply input, that I happened to touch the top of, and got a buzzing sound through the speakers. So I touched the tops of a few other caps and no buzzing. Touch that one again, buzzing. Not sure if that means anything really, since I’m using the Euro power supply. I don’t have any 100v caps around unfortunately so I’ll need to see what I can find tomorrow.

  • 100v cap is uhh.. A bit high rated for Euro applications, is it not? 16-24v caps should be fine as well, if you have any of those around..

  • if it physically fits id always go for higher voltage ratings as the MTBF will increase significant when running under the specified voltage…..

  • Well true, but in a pinch, a 16v cap will do the job(for a fairly long time as well).. Not disputing that a higher rated cap is better, but I’m fairly sure we’ve all been at a point where kludging it to see if it works was an acceptable solution :D

  • > I can’t seem to get any sound out of any of the points on the PCB

    What about the control voltages?

  • The voltage check points indicated in the trouble-shooting section all check out as expected. It also responds to MIDI just fine. There does not seem to be any continuity between GND, +12 or -12 but the regulators continue to get very hot.

  • Can’t touch them hot, or “oooaaww that’s kinda hot better remove finger before burning it” hot?

  • Better remove my finger and turn it off before it cooks sort of hot.

  • That’s normal, mine is the same sort of “warmth”.. So doubt it’s a short :)

  • Okay. I could not find any indication of a short so that’s good.

  • Can I use a polarized cap in that position?

  • The filter doesn’t have a big DC offset so yes, you can do it – it might have a shorter life, though…

    I don’t think this has anything to do with your problem. The NP caps are coupling the different sections of the synth (oscillator to filter, filter to VCA). But you said the oscillator doesn’t work, right?

    You should really start to troubleshoot the VCO... Please report your observations on points A, E, F, G, H…

    What about the digital oscillator? What do you have on point U? And by selecting the DCO on the main panel (instead of the 2 sub settings), and increasing its volume, does it get through the filter? Point M, N, O?

  • Osc works, I actually think it’s the VCA. I’ll see if I can find something to replace that cap and check the points you mentioned Olivier. Thanks.

  • I’m confused, because a few posts ago, the symptoms were:

    “I can’t seem to get any sound out of any of the points on the PCB”

  • Sorry Olivier, let me try that again. I can’t seem to get any sound when I use some audio probes but if you listen to this video at a fairly high volume, you can hear what I’m getting out of the audio out phone jack. MIDI does work, I can’t hear the filter frequency doing anything but resonance (kinda wild actually) works.

  • There are dozens of possible failures that would result in the output giving only faint clicks/noise. You’re hearing a DC offset through the VCA, or board noise, or whatever hum/noise your recording setup has because you’re cranking the gain up to hear something. And there’s very little useful information in that. That’s why my troubleshooting instructions emphasize on finding the location of the problem on the board by probing intermediate points rather than trying to infer things from the output.

  • Okay. Point A, I get a click. E, F, G, H nothing.

  • Point A is a CV, no sound is expected here:

    “”“
    VCO CV. With default patch, C3 = 2.048V, 0.5 V/Oct. Affected by ENV or LFO VCO modulations, mod-wheel… Might be scaled or offset if software tuning is used.
    “”“

    So do you get something 2V-ish for middle notes, changing by 0.5V every octave?

    E, F, G, H are the VCO signals. So is the VCO dead or not? You wrote “Osc works”, so I assumed these points were OK…

  • I wrote “Osc works” because I get a sound from the Anushri output jack that responds to MIDI. But I do not get any sound from E, F, G or H. I assumed it was from the VCO.

    The VCO CV does change as expected, around 2v at C3, increasing by .5V every octave.

  • Which ICs are playing the role of analogue oscillator? And, only semi-related, I’m surprised at how difficult it is to find 4.7uf, 100v, non-polarized, radial capacitors.

  • > Which ICs are playing the role of analogue oscillator?

    IC6B does temperature compensation ; IC6C exponential conversion, IC7B is the integrator ; IC8 the reset comparator ; IC10A inverts the saw ; IC9B / IC10B generate the square and IC10D mixes everything.

  • I figured you’d say something like that! Could any one of these (or several or all) be the point(s) of failure? I still find it odd that everything worked fine at first. I may never know but it would be nice if I could figure out what I did to cause this thing to suddenly go south.

    The sound that I’m hearing that responds to MIDI, is that the digital osc?

    I’m heading off to Boston for a few days, maybe we can take this up again on Monday?

  • > Could any one of these (or several or all) be the point(s) of failure?

    Or a passive they are connected to… Is the LM311 output OK?

    > The sound that I’m hearing that responds to MIDI, is that the digital osc?

    Yes, directly or through whatever defective coupling is happening (the filter doesn’t work either, right?).

  • You’re right, the filter has no effect. The output of the LM311 is point E on the troubleshooting photo, and there is no output.

    Thanks Olivier

    Randy

  • So, getting back to this, Olivier, can you give me any more hints as to which part (or parts) would be worth checking?

  • IC6B ; IC6C ; IC7B ; IC8 and the passive components surrounding them.

  • I’m trying to read the schematic and figure out what to look for, really new at this sort of thing. From what I can see, referring to the V2164 (IC6), it looks like the Anushri uses pin 7 (in 2), pin 5 (out 2), pin 10 (in 3) and pin 12 (out 3). The only pin that registers any voltage is pin 12, which gives me a constant reading of +4v, no matter which key I hit.

    Interestingly, if I hold one probe on ground and touch the other to pin 13 or pin 4, I get the same sound I would get by shorting the hot and gnd of the output jack. Next up, IC7.

  • > The only pin that registers any voltage is pin 12

    These are current inputs and outputs.

  • Olivier, should I be seeing something at pin 7 of the 2164 when I play a MIDI note? At rest it shows whatever the last value was, hit middle C and it’s -0.0039, add an octave -0.0032, another octave -0.0029.

    The lower lead of R58 (closet to the trimpots) reads 0v no matter which note I hit but the upper lead shows the same values as above. Lower lead of R59 is -4v, lower lead of R60 is 2.-ishV at middle C, increasing by .5v for each octave. Upper leads are the same as R58 (and pin 7). Whatever is going into R59 and R60 seems correct if I’m reading the schematic correctly. Lower and upper leads of R57 show the same as the upper leads of the others (and pin 7). R56 shows 4v at the bottom, 0.25 at the top.

  • > should I be seeing something at pin 7 of the 2164 when I play a MIDI note?

    As I said, the 2164 is a current input chip, so the voltage on the input doesn’t say anything. Same for the output. You would need to probe the input/output currents and that’s quite complicated to do since you need to rewire the circuit.

    > The lower lead of R58 (closet to the trimpots) reads 0v no matter which note I hit but the upper lead shows the same values as above

    That’s normal, the lower lead of R58 is connected to the front panel VCO CV input jack.

    > Lower lead of R59 is -4v,

    That’s normal. The exact value depends on the position of the R74 (VCO offset) trimmer.

    > lower lead of R60 is 2.-ishV at middle C, increasing by .5v for each octave

    All good!

    > R56 shows 4v at the bottom, 0.25 at the top.

    All good!

    Next step: voltage on IC7 pin 1 (or IC6 pin 11, they are the same). When you play a middle C, then one octave up, then one octave down.

  • That’s interesting, a “current input” chip. I have alot to learn. Pin 1 of IC7 at C3 is 1.35v, at C4 is 1.16v, at C5 is 0.96v. You can call me “grasshopper” but that reference is probably before your time.

    I checked IC8 as well since it is also part of the VCO piece, and, keeping in mind my lack of knowledge, this one seems kinda screwy. Pin 1 is -5, pin 2 is 2.8, pin 3 is 2.5, pin 4 -5, pin 5 is 5, pin 6 is 5, pin 7 seems to be floating, just random values below 0 and pin 8 is 5. And the 7805 is nearly too hot to touch, I can’t leave my finger on it for my than 1 second.

  • > Pin 1 of IC7

    Readings are OK.

    > I checked IC8 as well

    All pins except 2 and 7 are OK. Voltage on pins 2 and 7 is not DC, so a volt-meter won’t take you anywhere. You need a scope – or at least record them with your audio interface.

  • I get a loud staticky crack from pin 2 and a very quiet staticky crack from pin 7 when the positive lead hits the pin, but pressing a key has no effect.

    I’ve been using the ground near the right hand 4822 so just for fun I thought I would see if there was anything on the adjacent VCO and VCF holes. Pressing a key using the VCO hole gives me a thump similar to the VCA envelope being nearly shut, and using the VCF hole gives me a louder version of the same thing.

  • Have you posted a photo of your board?

  • Keep in mind it was working when I first built it and then stopped working. These are not as sharply focused as I would’ve hoped but if they are not good enough I can do it again. One other option is that if you know of anyone in North America who could figure out what is going on with this, I don’t mind shipping it to someone. It’ll cost me some money but at least the Anushri doesn’t end up in a drawer.




  • Did you reflow all joints and add more solder where needed? It looks a little light on solder in spots. Thst could be the light though.

  • Yup, reflowed everything. I do tend to solder a bit light but after going over everything again it should be fine. What really bugs me about this is that it worked fine when I first fired it up!

  • Any conclusion to this situation? I have similar problems and would very much like to know what happened to yours. Thanks!

  • I checked everything, spent a few weeks, very frustrating, and eventually gave up and sent it away to be looked after. A second set of eyes (qp’s in this case) noticed a small flaw and fixed it. Here is what he said:

    “One of the IC socket header pins had a big gap, so the TL072 wasn’t making contact with the pins. Desoldered it and put in a posh socket.”

    I am currently waiting for it to arrive.

  • Things working and not working intermittently could point towards ICs not sitting in their sockets properly (or bad sockets), which seems to be the case here for one of the opamps.

    @corpusjonsey: check whether your ICs sit deep enough in their sockets. Re-seating might help (check the sockets after you take an IC out).

    I’ve had ‘wobbly’ TL072’s in the past: they only have 8 legs, so can move a little if your case gets thrown around (‘travels’), and the sockets are of the cheap variety. The bigger ICs are usually OK.

  • I saw that on qp’s instagram. I stlll can’t really see it though. Good to hear it’s working again though.

  • Great thanks everyone.

  • @qp did indeed get my Anushri sounding like it should again, as if there was any doubt he would. Now I just need to calibrate it. I started with the VCO offset and can’t even get it to track an octave. Any hints would be appreciated.

  • I’m so close to getting this thing working. Olivier, can you tell me which components affect the pitch tracking of the VCO?

  • Page 3 of the schematics, everything before IC7 pin 6.

  • And the saga continues…

  • Coming into R60 I get 2.0v, 2.5v and 3.0 volts (or thereabouts) when I hit a key, one octave up and another octave up. So, I’m going to assume whatever is coming into R60 is fine.

    If I put the positive lead on the other side of R60 the sound changes from a nice round tone to something distorted with added vibrato. This also happens on the out side of R58 and R59. So, I’m also going to assume there is something screwy with whatever is feeding R58 or R59.

    It looks like the Offset trimmer is feeding R59 but I’m not sure what is feeding the offset trimmer. R50 is fed from the VCO CV In jack I think so I would expect it to be okay if nothing is coming in to that jack.

  • > If I put the positive lead on the other side of R60 the sound changes from a nice round tone to something distorted with added vibrato.

    This is normal, IC6 pin 7 is a current input – the equivalent of a virtual ground. It is extremely sensitive to even small currents.

    Pin 3 of R74 is a Vref at -4.096V. Small deviations from that are OK.

    I realize you haven’t described the problem. “and can’t even get it to track an octave.”. Does the synth jumps by 10 octaves every time you move one octave up? Or 1.0001 octaves? Which interval do you get when the trimmer is set to its minimum position, and which interval do you get when the trimmer is set to its maximum position?

  • Lowest (CW) ~9th
    Max (CCW): ~9th

    I replaced the trimpot just in case, no difference.

  • @piscone, man, I dropped the ball on that one! I shoulda tested the v/oct when I had it. Didn’t even think about that. Many sorries!!

    @sammy123, if you look reeeeaall close you can see the top left 3 pins tension tabs aren’t in contact with the pin. Those sockets are ok and cheap, but I’m starting to lean towards the nice open frame sockets. Cost a bit more, but pretty damned worry free.

  • NP @qp. It works fine with software tuning but I want to use it for Eurorack.

  • So voltage on IC7 pin 1 does not change when you turn R75? Does turning R74 at least shifts it up/down?

  • Voltage on IC7 (TL072), pin 1 does change when turning R75 (v/oct) and the pitch of the incoming note changes too. Turning R74 also changes the voltage on pin1 of IC7, and also affects the incoming pitch, actually more than turning R75.

  • So how comes you still have a 9th on both sides of the trimmer? Offset is to transpose, Scale is to increase/decrease the interval.

  • That is what I am trying to figure out! I’m going to do some video of this, but I’m thinking it’s time to get back to playing more acoustic piano.

  • I did a short video of some playing some scales while turning the two trimmers, first R75 and then R74 and I think a bit more R75. Please tell me I’m doing something stupid.

    Video again

  • Take a piece of paper.

    Set the V/Oct trimmer to the minimum value.

    Make a table, with, for each line:

    • Voltage at IC7 pin 1 for a C2 note.
    • Frequency measured by a software tuner.
    • Voltage at IC7 pin 1 for a C4 note.
    • Frequency measured by a software tuner.

    Give 2 turns to the trimmer and redo the measurements.

    In addition, measure:

    • Voltage at the pitch DAC for a C2 note.
    • Voltage at the pitch DAC for a C4 note.

    That will be very useful to understand what’s going on.

  • Just a stupid question (again), minimum value is full CCW or full CW?

  • It doesn’t matter… Just turn it fully in one direction and then turn by turn in the other. What I want to understand is:

    • the relationship between interval and voltage at IC7 pin 1.
    • the relationship between voltage at IC7 pin 1 and the DAC voltage.

    That’s something I can chew on. A video or soundclip showing how broken things are doesn’t say much…

  • Minimum IC7 Pin 1 C2 1.626v, 86hz
    Minimum IC7 Pin 1 C4 1.2v, 436 hz
    2 turns IC7 Pin 1 C2 1.61v, 91.6hz
    2 turns IC7 Pin 1 C4 1.189v, 457.5 hz
    Minimum DAC (front of R60) C2 1.55v
    Minimum DAC (front of R60) C4 2.55v
    2 turns DAC (front of R60) C2 1.55v
    2 turns DAC (front of R60) C4 2.55v

    I assumed the DAC voltage could be measured at the in side of R60. I posted the video for a sanity check, so you could potentially say I was actually doing the whole thing wrong.

  • I should have been more clear about this, but I wanted the procedure to continue until the end of the trimmer is reached…

  • Put it in a spreadsheet, hope it’s what you had in mind. Attached Anushri Log v1.xlsx.

    Anushri Log v1.xlsx
    11K
  • My analysis is that you probably have some tolerances biting each other in the wrong way – and/or maybe something with the trimmer as it seems to have a residual resistance even in its extreme setting.

    The numbers indicate that the best tuning you can achieve is 1 octave + 0.8 semitones between C2 and C3. In the worst case, it would be 1 octave + 2.1 semitones. This contradicts the “Lowest (CW) ~9th ; Max (CCW): ~9th” you have posted, and this is why I was very puzzled.

    I would suggest replacing R61 by a 91k resistor ; and R75 by a 20k trimmer.

  • When I got the ~9th, the Anushri had been on for several hours, maybe that made a difference. Also, I was not using the tuner, just a close piano. Next time (although I hope there will not be a next time), I will know a better method to use.

    Not sure which parts I have around but I will make the adjustments and post back. Since I’m going to be wandering around to get parts, do you have any other suggestions for things I should pick up, just in case?

    Thanks Olivier.

  • Put 92k into R61, all I had around. Big difference! Not quite perfect but certainly playable. I also put in another 10k trimpot I had around, just in case there was something not quite right with the one provided with the kit. I’ll need to hit the local parts store to see if they have the 91k resistor and 20k trimpot.

  • And the fun continues, this time with the VCO CV. C2 = -1.03, C2 = -.02, C4 = 0.98. It is jumping 1 v/oct, just the wrong volts. I also ran a patchcord from a MIDI2CV module to the VCO CV in jack on the Anushri, no effect, no sound. The MIDI2CV is operating correctly.

  • I don’t understand what’s wrong here. Could you explain what you expect?

  • I’m not sure what to expect but I assumed two things. One, if I hit a note on a MIDI keyboard plugged in to the Anushri and that note results in the Anushri producing the pitch that represents that note, I figured an external VCO hooked to the VCO CV out would also produce a pitch of some sort. Two, if I send a CV to the CV In of the Anushri, it would react to that CV by producing a pitch. At the moment, neither of those two things are happening.

  • Wait, I moved a few things around and it does work. Actually, forget it for now, I need to have a better look and try and figure out what I’m doing wrong first. I may have just misunderstood the manual.

    Boy I wish I knew more about this stuff!

  • > I figured an external VCO hooked to the VCO CV out would also produce a pitch of some sort

    It would not necessarily produce the same note. The standard is that pitch goes up by one octave when voltage goes up one volt and that’s it. There’s no absolute mapping between a voltage and an actual note number. It doesn’t matter anyway because your external VCO has its own frequency knob that influences its pitch too. The CV input on a VCO is merely a “transposition” CV.

  • Yup, I know, that’s why I wrote “produce a pitch of some sort.” Also why I wrote “forget it for now.” It was my misunderstanding of something else and I reacted too quickly. I’m nursing a really bad cold or seasonally allergies, not sure which, but I’m thinking even less clearly than usual.

    Now that I’ve had a chance to fool with the Anushri a bit ‘though I gotta say it’s a great sounding little synth. Lots of variables in a small package and really nice filter. It’s mounted in my lunchbox and I’m going to pair it with an STO, EG, LFO, Mix/VCA and perhaps another filter for the STO.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion