Juno Chorus

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Hi there!
After letting this project on a shelve for a while, i came back to it last night and found out its working somehow!
I have a few issues tho :
- i made the 220pf// to R24 mod but still have an annoying noise that fluctuates with modulation rate (i guess thats what dmp called robot noise?)

- it has high pitched noises increasing when i engage the depth knob - or when this one is fully off…
Could this be due to the fact that, for now, i housed it in a plastic box with metal faceplate?
MN3xxx series BBD Analog Shift Registers generally have two MOS source follower outputs. One is the delayed audio signal, the other a duplicate of the actual signal delayed one half clock tick. Some designs use a trimpot to balance the DC offset caused by different Vgs in the two source followers which generates a clock frequency square wave. If the clock frequency is high enough above audio, a simple LPF can scrape out that clock frequency component. If one of the two outputs is not connected the clock frequency component can be a couple volts p-p.

Unlikely to be related to chassis.

JR

- and lastly, i’d like the rate LED but can’t figure it out… could someone help me with completing this drawing with needed connections?


Thanks very much!!
 
Hi!
So i made all the mods, but still the same "issue"...
Attached is a small recording to showcase what i get :
First is a small dry sequence and then i move the knobs, starting with depth fully CCW and ending with input fully CCW...
Is this normal sounding to you?
Thanks for your time!

edit :
JohnRoberts, dmp, thanks for all your explanations!
Sometimes i'm amazed by my brain... how slow it can be!
It took me a while to understand what you were talking about, even by reading daily your different posts....
TRIMMERS, YES! TRIMMERS!
But after tweaking them all ways around, i couldnt notice any change in the unwanted noises - or any change at all.
Maybe they are bad, so i will change them and update my progresses here
Thanks again!
 

Attachments

  • test chorus.mp3
    2.3 MB
Last edited:
Hi!
So after changing the trimmers, i still have the same behavior…

I noticed that r81 (r81 on gligli pcb = r85 on original schematics) was in fact 22k in original schematics, so i’m wandering why did you change this one to 10k, dmp?

Not sure it can affect my issue buy i’m curious!

Thanks!
 
The original had 10k (R15 I think) in the schematic in the service manual
This sets the bias of the transistor in the buffer of the input signal going into the MN3009. I don't think it would cause the robot noise we've been talking about.
 
Isnt it the green part that we are talking about?
I’m far from secured with my understanding of schematics…

But yeah this is just curiosity..

Do you have any recommendations on what i could try to tweak? Maybe a different cap value in parallel with r24?..

Thanks!
 

Attachments

  • IMG_0048.jpeg
    IMG_0048.jpeg
    201 KB
Post the BOM you used with updates to all the parts you changed, showing what all your installed values are? I can have a look at it
 
Well the screenshot i sent this morning comes from this website :

https://www.florian-anwander.de/roland_string_choruses/
He has other schematics there, pretty nice source of infos on Roland’s choruses btw!


I do see the link in your posts yes and i checked it out. i saw some differences and it seems the net is full of various schematics…

For the BOM, i used the one Gligli provided and did the exact same mods as you explained here except for the switch that i didnt use.
I bought almost everything from tayda so the sk30a or maybe sa1015/sc1815 could be somehow bad - but i guess its not related to my robot noise issue
 
I can't help if you don't post what parts you used. Assuming you used all the right parts isn't going to help figure out the problem.

Similarly, if you don't post what schematic you used it's hard to be on the same page. That page you linked to has a juno 6 chorus schematic and it agrees with mine, not yours. I'm not going to hunt through all the other schematics there to try to figure out which one you've been looking at because you can't be bothered to just answer the question.
 
Hi dmp,

Sorry for the schematics link, i thought i posted both links but only one made it thru…
Here is the juno-60 schematics i took the screenshot from :

https://www.florian-anwander.de/roland_string_choruses/JUNO-60_schem_chorus.jpg
I didnt really use it to build it, just after, to check if i could spot something that would fix my issue.

I used gligli’s bom provided here :

https://github.com/gligli/juno-chorus-clone/commit/86dc4d6e32addefe5640765b14b110dfda580ac2
And followed it exactly, then just did the mods you explained here, a few messages before. No more no less.

I’m sorry if i didnt answer your question the way you expected. If there is still some misunderstanding i could post a picture of my pcb?

I was a little confused with all those different revisions schematics, but, as stated on the github, this project is based on juno-60
 
OK, juno 6 and 60 use the same chorus, the 60 does the switching in some different ways to allow recall. But I can refer to 60 part numbers to keep things clear. The R19/R24 and adding 220pf would be the same for the 6 and 60. A way to diagnose this is to see if you have the robot noises in both right and left as the issue only appears in one side (as it is reversing the phase of the chorus modulation) for the R19/R24 phase inverter. The way 'space' is provided to a stereo chorus effect like this is to simply invert the phase of the modulation to one of the L/R chorus chips.

You might also try lowering the level into the chorus and see if you still have the "issue"
Reroute wiring to the In/Out and any pots to see if it has any effect.

Other than that you'll probably have to try to rule out any other way the modulation is leaking into the signal. How did you connect RV1 & RV2? Maybe post a picture. If the ground lug of those pots are connected to a different ground location it could possibly let the modulation leak in. Should be close to p2 p5 of U1
 
Thank you very much for those leads!
I won't be able to experiment on this the next weeks but will definitely do it at some point and hopefully will come back here with good news!
 
Hello all...this is my first post here (although I'm quite active on Mod Wiggler). I'm in the process of building this, as well as customizing some elements of it, and ideally improving a tad on GliGli's schematics and documentation (if nothing else for clarity).

Has anyone else built this recently? If so, any thoughts you might share? I got mine working somewhat, although I suspect not everything is 100% quite yet. I'm awaiting on the shipment for new 2SK30's. The ones I'm using right now are from Tayda, but someone on Reddit suggested using 2SK30A GR's instead so that's what I'm waiting on.

Other than that, I am curious about some of the mods already made in this thread such as the LED and the True Bypass.

Anyways, I'd love to know if anyone else has gotten this to work perfectly, and if they'd be willing to share some additional info. Thanks!
 
Update - I'm starting to wrap my head around some of the idiosyncrasies of this build. What I have noticed is that the so-called "robot noise" mentioned in this thread might be related to the MN3101's (or in my case the CT3101 from CabinTech) which was not quite well seated in the socket. After futzing with it for a bit, and making sure that all of the pins had a solid connection, I got rid of the rhythmic interference-like noise that I was getting.

My next challenge for this is to add a guitar pre-amp and a switch between that and the regular line level input, right now trying to play a guitar through the chorus results in far too low of a volume. I also would love to have an output level control as well, in which case I would go back to using 100K resistors for R83 and R84 in order to get a much louder output that I could then turn down as necessary. At the moment, with the 10K resistors in place, the chorus sucks quite a bit of volume vs. the bypassed signal level.

Any thoughts on how to best tackle any of this?
 
I posted how I fixed the 'robot' noise but it seems there are multiple kinds of noise this build can have - if the clocking leaks into the audio.
I don't have a problem with levels when plugging a guitar directly in. I often use it that way. Then I turn the input down for a higher level, like a synth.
Compare to the chorus section of the Juno 6. Maybe you have something off still. I think R83&R84 are supposed to be 100k?
One thing I'd recommend is use TL072 instead of TL082. I must have gotten a bad batch of TL082 and they kept failing. The TL072 is spec'd to have lower noise anyway.
 
I posted how I fixed the 'robot' noise but it seems there are multiple kinds of noise this build can have - if the clocking leaks into the audio.
I don't have a problem with levels when plugging a guitar directly in. I often use it that way. Then I turn the input down for a higher level, like a synth.
Compare to the chorus section of the Juno 6. Maybe you have something off still. I think R83&R84 are supposed to be 100k?
One thing I'd recommend is use TL072 instead of TL082. I must have gotten a bad batch of TL082 and they kept failing. The TL072 is spec'd to have lower noise anyway.
Thanks, I was actually contemplating replacing the 4 TL082's with 2 TL074's...not quite sure how it would affect things, I guess some people would be even more upset at this being less and less of a true clone.
 
I think R83&R84 are supposed to be 100k?
I actually used 10k for these, as the gain was higher than I wanted with 100k.
When I'm using this with a guitar level input, I am connecting it to an amp after. So it is guitar level in / guitar level out. If you want it to bring guitar level up to line level, then a switch on the output might be the way to go, to give it 20 or 30 dB gain.
 
I actually used 10k for these, as the gain was higher than I wanted with 100k.
When I'm using this with a guitar level input, I am connecting it to an amp after. So it is guitar level in / guitar level out. If you want it to bring guitar level up to line level, then a switch on the output might be the way to go, to give it 20 or 30 dB gain.
I think I'm sold on going with a guitar preamp to the circuit along with a switch. Seems to work pretty well on my breadboard, but I'll need to update the schematic.

Was also thinking of going back to the 100K resistors on the output side, but adding an output level knob to further fine tune the adjustability.

I was able to calibrate the circuit with a single generator and an scope which I think got me to the ballpark of where I need to be on the waveshape clipping.

I'm determined to update this design a bit as well as provide some clearer instructions. If there is an interest maybe I'll offer it as a kit or something.

In the meantime, I do have some extra PCB's if someone else wants to build one from the existing schematics.
 
Ok...latest development, I put back 100K resistors on R83 and R84 to match the original schematic (although technically R86 and R87 should also be switched to 1.5K values.

My thought was that it would be cool to be able to be able to add an output volume potentiometer. I thought it'd be as simple as wiring two 100K pots in place of either the R83 and R84 or perhaps R86 and R87 but neither gives me the desired effect and rather results in some nasty distortion.

Anyone have any ideas on the best way to add an output level control to this circuit?

The other issue is that the input level pot is really touchy in the sense that it only delivers a useable level when the dial is set pretty low. I was thinking that perhaps I should change the value of the input pot (RV5) to make it so that I don't have to keep it really low all the time. Any ideas? Would I just go to a lower value on the pot, say from 100K to 10K perhaps?

Many thanks for all the help!

P.S.

I'm attaching the schematic I'm working off of:
 

Attachments

  • Juno Chorus Clone Schematic.pdf
    155.1 KB
I thought it'd be as simple as wiring two 100K pots in place of either the R83 and R84 or perhaps R86 and R87 but neither gives me the desired effect and rather results in some nasty distortion.
You could wire a 10k resistor in series with a 100k pot, where the wiper of the pot is attached to the leg of the pot, making it a variable resistor on the opamp feedback. At one extreme, the feedback is 110k and the other 10k. That would give you a big gain adjustment on the output opamp. It is a simple inverting opamp with the dry and wet inputs mixed, through 47k and 39k respectively.


The other issue is that the input level pot is really touchy in the sense that it only delivers a useable level when the dial is set pretty low. I was thinking that perhaps I should change the value of the input pot (RV5) to make it so that I don't have to keep it really low all the time. Any ideas? Would I just go to a lower value on the pot, say from 100K to 10K perhaps?
No, that's not how it is working electrically at all. The output of U9A is seeing 100k as the load to ground. The wiper adjusts the level from nothing to full. You don't want to change the loading of the U9A (which is what changing the pot value would do). 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). Or It could be that your circuit just has way more gain than you need.

I'm attaching the schematic I'm working off of:

I built mine by comparing that schematic to the juno 6 / 60 schematic, carefully component by component. For instance R67=22k and R70=22k in the juno 6, giving a U9B gain = 2. In the schematic you posted they are 33k and 10k, giving gain=4.3 (edit: this is a non-inverting opamp stage). Read and understand how a inverting / non-inverting opamp gain stage works, and compare your schematic to the juno6. It will take some time to understand, but you'll really be able to sort your problems out once you learn how the circuit is working.
 
Last edited:

Latest posts

Back
Top