Juno Chorus

[Midphase]

Thank you for this info. However in the "clone" schematic R67 and R70 don't correspond to the R83 and R84 on the original Roland schematic as far as I can tell. However in the clone, R75 and R76 do (I think) which would be a a match to the Roland ones as far as values.

I have been studying the Roland schematic, but it's a bit challenging to compare 1:1 since in the Roland schematic, the entire input section is assuming that a signal is coming from within the synth so there are no level adjustments, and even the depth and rate knobs don't exist in the original.



P.S.

"Are you using a log (audio) pot for RV5? What you are describing would happen if you used a linear pot (or you have a log pot wired backwards). "

I am using a LOG pot, but I do appear to have it wired backwards, I'll change that. This is one of the many things that I'm hoping to improve in my design, to make things a bit more clear for noobs like myself!

 

[dmp]

However in the "clone" schematic R67 and R70 don't correspond to the R83 and R84 on the original Roland schematic as far as I can tell. However in the clone, R75 and R76 do (I think) which would be a a match to the Roland ones as far as values.

You are right - I was looking at my own notes on the schematic. I used 22k for R67 & R70, but that opamp stage was not used in the Juno. Juno starts with R75, like you pointed out. So I lowered the gain of the input stage.

and even the depth and rate knobs don't exist in the original.

Correct. I still think it would be cool to duplicate the I, II, and I+II switch options, in addition to the pots. But the pots are much more versatile in comparison to a real Juno.

Check your values of R81 R61 R62

 

[Midphase]

You are right - I was looking at my own notes on the schematic. I used 22k for R67 & R70, but that opamp stage was not used in the Juno. Juno starts with R75, like you pointed out. So I lowered the gain of the input stage.


Correct. I still think it would be cool to duplicate the I, II, and I+II switch options, in addition to the pots. But the pots are much more versatile in comparison to a real Juno.

Check your values of R81 R61 R62

Thank you for all the help so far. I have some ambitious customization ideas for this and I'm hoping that perhaps we could figure it out together so that the entire community could benefit.

- Regarding the potentiometer suggestion from your previous post as to create a variable resistance for the inverting amp, in practice it didn't work and only yielded very distorted results. Perhaps I was doing something wrong, but if I understood your instructions to run an 100K pot in series with the 10k resistor and short pin 2 and 3 of the pot together.

- I ended up changing the value of R83 and R84 to 47K, 10K as in the schematic yielded too low of an output while 100K in the original Roland schematic was way too hot.

- I am running into a really bizarre issue that I've been having since the first build. Basically the very first time I power up the unit I get no voltage on the positive rail when I probe the voltage regulator. However if I unplug, and then plug it back it it works just fine. It's almost as if the sudden burst of voltage is weirding something out. Someone suggested changing out the D1 and D2 diodes in favor of something beefier, as well as using 50V electrolytic capacitors which I did, but the problem persist. It really has me stumped. I tried all sorts of things after testing out the circuit for shorts. Replaced the voltage regulator several times, changed out the diodes and the electrolytic caps. Any idea of what might be causing the issue?

- Next on my agenda is to figure out a way to have this be able to run on 12V as to make it Eurorack power friendly. I tried swapping out the regulators to 7812 and 7912 and it does work, but the chorusing effect is not as pronounced and I am picking up some clipping in the signal. Any thoughts on what else I might need to look at in order to get it to work? According to the data sheets of the 3009 and 3101 IC's, they should be ok running at +/- 12V, and I know the TL082's shouldn't have any problems with that.

- Lastly, I'd like to be able to have CV control the depth and rate along with the knobs. Looking at some of the Moritz Klein videos on YT it seems like it should be doable, but I'm a tad apprehensive since everything so far has been considerably more challenging than initially expected.

Anyways, I appreciate any further thoughts you might have. Many thanks!

 

[dmp]

- Regarding the potentiometer suggestion from your previous post as to create a variable resistance for the inverting amp, in practice it didn't work and only yielded very distorted results. Perhaps I was doing something wrong, but if I understood your instructions to run an 100K pot in series with the 10k resistor and short pin 2 and 3 of the pot together.

- I ended up changing the value of R83 and R84 to 47K, 10K as in the schematic yielded too low of an output while 100K in the original Roland schematic was way too hot.

Strange - assuming you wired it like this sketch, it should work. Probably would want to use a dual 100k audio pot so the L and R are equal gain.

- I am running into a really bizarre issue that I've been having since the first build. Basically the very first time I power up the unit I get no voltage on the positive rail when I probe the voltage regulator. However if I unplug, and then plug it back it it works just fine. It's almost as if the sudden burst of voltage is weirding something out. Someone suggested changing out the D1 and D2 diodes in favor of something beefier, as well as using 50V electrolytic capacitors which I did, but the problem persist. It really has me stumped. I tried all sorts of things after testing out the circuit for shorts. Replaced the voltage regulator several times, changed out the diodes and the electrolytic caps. Any idea of what might be causing the issue?

I'm not sure why it's behaving strangely, you could change it to a standard design and see if the problem goes away. Full wave rectifier with 1n4007 and add diodes back across the rectifier. Here is a good example (that uses adj regulators, but the 7815 / 7915 will be similar). But I've never had this problem with my build. I am using a wall wart 15vAC plugged in to the enclosure with the pcb rectifier/regulator.
Are you using the pcb or do you have this all breadboarded? Your problems seem to be implementation rather than design.

Next on my agenda is to figure out a way to have this be able to run on 12V

Nothing jumps out - I think this should run on +12/-12. You'd expect less headroom and I'm not sure what headroom the design has with +15/-15. What level of signal are you running into it for the testing?

Lastly, I'd like to be able to have CV control the depth and rate along with the knobs.

Doable - but advanced. I'd suggest start running simulations in CircuitLab. Easy for a beginner to start learning whats happening. Start with the basic blocks, like the inverting/non-inverting opamp stages. Then move on to the more advanced parts, like the oscillator that sets the speed/depth.

 

[Midphase]

I'm not sure why it's behaving strangely, you could change it to a standard design and see if the problem goes away. Full wave rectifier with 1n4007 and add diodes back across the rectifier. Here is a good example (that uses adj regulators, but the 7815 / 7915 will be similar). But I've never had this problem with my build. I am using a wall wart 15vAC plugged in to the enclosure with the pcb rectifier/regulator.
Are you using the pcb or do you have this all breadboarded? Your problems seem to be implementation rather than design.

Thank you! I need to figure out the power issue first since it does affect my testing, and when the positive rail isn't outputting voltage the audio gets crazy distorted (probably because the opamps have no positive headroom).

I've build this on PCB's that I got made from JLCPCB, although some of the newer portions of the circuit I'm building on a breadboard. But for testing, I'm generally keeping everything in the original PCB, and aside from some of the resistor tweaks to the output that I mentioned above, everything is like in the original design.

Here is the power supply section:



A few things to mention:

First of all, my understanding is that regulators need at least a couple of volts above the regulated output voltage in order to function properly. Wouldn't feeding in a 15V AC not give the regulators enough input to work with?

In addition, someone pointed out that the 1n5819 diodes and the 25v C1 and C2 caps are under specced for an AC power supply that could be providing as much as 45v peak to peak. I replaced the diodes with 1n4004's and the caps to 50v rated ones to be extra safe.

As I mentioned, from time to time (and far too often to call it a nuisance), I will get absolutely no voltage on the positive rail when I test with my multimeter on pin 3 of the 7815. Repeated plugging and unplugging fixes this as I eventually get a steady 15v output from 7815 and the unit works as described...but it's driving me nuts.

The soldering is clean, been soldering a ton of circuits in the past 3 years and I have a pretty good hang of what I'm doing. I don't see any accidental bridging or potential shorts, although the best explanation that I can think of is possibly a short caused by an intermittently malfunctioning transistor.

I will continue to test and see if I can figure out exactly what is happening. I'm also thinking about building out a second PCB and see if the issue appears on that circuit as well.

Lastly, regarding the output volume control, I wonder if perhaps something like this would be a simpler solution as a way to regulate the level?






Many thanks, and let me know if you have any other thoughts about what I'm trying to do.

 

[dmp]

First of all, my understanding is that regulators need at least a couple of volts above the regulated output voltage in order to function properly. Wouldn't feeding in a 15V AC not give the regulators enough input to work with?

In addition, someone pointed out that the 1n5819 diodes and the 25v C1 and C2 caps are under specced for an AC power supply that could be providing as much as 45v peak to peak. I replaced the diodes with 1n4004's and the caps to 50v rated ones to be extra safe.

15VAC is rms (root mean squared). The peak voltage is sqrt(2)*15 = 21v
This means with no current draw, the rectified voltage on C1 will charge up to 21v max, and on C2, -21v max. The regulators should have enough voltage to create 15v and -15v. This is the standard way a power transformer is sized for a regulated DC PSU. As current is drawn by the circuit, the pre-regulated rectified voltage will decrease, but I expect it should be OK (and my build has not had this problem you are describing). The capacitors see a maximum of 21v each, neither will ever see +v to -v (42v) because they are tied to the ground. Diodes are similar.

1n5819 is not underspec'd but switching to 1n4007 is cheap and a more common part to use in this application.

For the output pot, as long as the op amp can drive that pot as a load, it will give you a fader. Give it a try.

 

[dmp]

Have you verified that you have AC voltage when you turn it on and do not have DC voltage? Maybe the transformer is glitching with a internal fuse or something. If you do - then measure the input DC voltage to the regulator

 

[Midphase]

15VAC is rms (root mean squared). The peak voltage is sqrt(2)*15 = 21v
This means with no current draw, the rectified voltage on C1 will charge up to 21v max, and on C2, -21v max. The regulators should have enough voltage to create 15v and -15v. This is the standard way a power transformer is sized for a regulated DC PSU. As current is drawn by the circuit, the pre-regulated rectified voltage will decrease, but I expect it should be OK (and my build has not had this problem you are describing). The capacitors see a maximum of 21v each, neither will ever see +v to -v (42v) because they are tied to the ground. Diodes are similar.

1n5819 is not underspec'd but switching to 1n4007 is cheap and a more common part to use in this application.

For the output pot, as long as the op amp can drive that pot as a load, it will give you a fader. Give it a try.

Ok...some new developments -- I just realized that I'm only getting chorus on one channel (the P3 output, but no chorus effect on P4, just a clean output). At first I thought that the SK30 might be causing the issue, but when I switched it with the one that is working same difference. Then I thought perhaps either my 3009 or 3101 is messed up...but once again when I swapped them I ran into the same issue. P3 has chorus, while P4 just sounds as if the effect is completely bypassed (i.e. no difference between the effect switch on or off). Adjusting the trim doesn't really yield anything different, although I know the trimmers work since I used them to calibrate the waveform.

At this point, since I do have a few other PCB's, the best option I think is for me to build a second one and see if I get the same issues with another one. If I do then I'm not quite sure how to proceed next.

 

[Midphase]

Have you verified that you have AC voltage when you turn it on and do not have DC voltage? Maybe the transformer is glitching with a internal fuse or something. If you do - then measure the input DC voltage to the regulator

Wouldn't that cause both the positive and negative not to work? The -15v always works on the first try, it never fails. The +15v is the one that is working intermittently, which might imply a fault somewhere in the positive rail before it hits the regulator, but at this point I swapped out all of the components between the AC in and the 7815 and yet I still get the same issue.

I am definitely getting voltage to the regulator. It appears that both the 7915 and the 7815 are getting about 24v from the wall wart (AC to AC 16V unit), but pin 3 of the 7815 is showing -.8v or some nonsense like that.

P.S.

I also have an AC to AC 12V transformer which appears to deliver about 17v peak to peak. When I plug that in there, I get the same results. 7915 is outputting -15v while 7815 is outputting -.8v. Plug, unplug and like magic get 15v from the 7815 -- so weird!

 

[Midphase]

I just had a weird thought -- is it possible that the transistors I'm using have a non-standard ECB pinout?

I was under the impression that all A1015's would follow the same pinout (which is as far as I can tell 1.E, 2.C, 3.B), and same with the SC1815's, but could it be possible that some brands would have a different pinout? It seems incredibly unlikely, and I'm not sure it's possible to check, but the thought did cross my mind.