Roland Jupiter 8 VCO1 drift

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, lisa, Joe.

User avatar
donepe
Common Wiggler
Posts: 84
Joined: Sat Mar 21, 2015 3:41 pm

Roland Jupiter 8 VCO1 drift

Post by donepe » Thu Jun 21, 2018 10:49 pm

my Jupiter 8 has an out of tune VCO1 on Module Board C

the synth has been fully calibrated and tuning on all other Module Boards on both VCO1/2 are rock solid and only drift by -1 or +1 cent.

VCO1 on Board C goes flat by -12cent after i have hit the compu-tune button

I have confirmed this behavior follows the Board in question and that specific VCO1 circuit so that rules out a mod con issue.

I have also replaced a few bad 4558 op amps in the VCO circuit but the problem still persists. I should mention that the CA3046 IC was replaced.


any ideas of where i should be troubleshooting would be much appreciated!

cheers

User avatar
Boogie
Veteran Wiggler
Posts: 507
Joined: Thu Nov 10, 2016 12:56 pm
Location: Belgium

Post by Boogie » Fri Jun 22, 2018 3:44 am

Hi donepe
both VCO1/2 are rock solid and only drift by -1 or +1 cent
I was just wondering how one can measure -1 or +1 cent.
Any special equipment, or is it the Jupiter 8 that tells it is like that.

Thanks
Boogie.
-------------------------------------------------------
“Digital generated ‘analog’, is missing its analog soul and therefore isn’t analog any more.”
Jupiter 8 clone: https://www.muffwiggler.com/forum/viewt ... p?t=209322
Minipops 7 clone: https://www.muffwiggler.com/forum/viewt ... p?t=199360
ClapTrap clone: https://www.muffwiggler.com/forum/viewt ... p?t=202248
www.nabla-instruments.com

User avatar
donepe
Common Wiggler
Posts: 84
Joined: Sat Mar 21, 2015 3:41 pm

Post by donepe » Fri Jun 22, 2018 9:53 am

i use the software strobe tuner by Katsura to check tuning

User avatar
guest
Super Deluxe Wiggler
Posts: 4511
Joined: Mon Aug 19, 2013 11:49 am

Post by guest » Fri Jun 22, 2018 12:15 pm

whats the nature of the error? is it a scale error or an offset error? (are all notes off 12 cents, or does it change over the scale?)
openmusiclabs.com

User avatar
Boogie
Veteran Wiggler
Posts: 507
Joined: Thu Nov 10, 2016 12:56 pm
Location: Belgium

Post by Boogie » Fri Jun 22, 2018 12:23 pm

donepe
great !
Thanks :guinness:
-------------------------------------------------------
“Digital generated ‘analog’, is missing its analog soul and therefore isn’t analog any more.”
Jupiter 8 clone: https://www.muffwiggler.com/forum/viewt ... p?t=209322
Minipops 7 clone: https://www.muffwiggler.com/forum/viewt ... p?t=199360
ClapTrap clone: https://www.muffwiggler.com/forum/viewt ... p?t=202248
www.nabla-instruments.com

User avatar
tobb
Super Deluxe Wiggler
Posts: 1072
Joined: Sat Nov 26, 2011 5:41 am

Post by tobb » Fri Jun 22, 2018 5:20 pm

a cap in the sample/hold circuit maybe?

User avatar
donepe
Common Wiggler
Posts: 84
Joined: Sat Mar 21, 2015 3:41 pm

Post by donepe » Mon Jun 25, 2018 10:29 am

guest wrote:whats the nature of the error? is it a scale error or an offset error? (are all notes off 12 cents, or does it change over the scale?)

all notes on VCO1 on that voice module are flat around 12cents

it is somehow related to the compu "tune" when I push the tune button, all other voices tune perfectly and this bad voice drops from around 6 cents flat to 12.

the board has been carefully calibrated in sequence with the others (vco tuning etc)

I should mention VCO2 on the same voice module tunes perfectly, so the issue lies somewhere with a bad cap, ic or op amp in the VCO1 circuit

User avatar
guest
Super Deluxe Wiggler
Posts: 4511
Joined: Mon Aug 19, 2013 11:49 am

Post by guest » Mon Jun 25, 2018 10:52 pm

there are a couple of possibilities here: 1. there could be some slew issue, which i think tobb was getting at with the s/h suggestion. if the computune acts quickly, before the circuit can settle, it will mistune. so this would suggest a capacitor in the CV summing stage, or perhaps the stability cap in the exponentiator. or 2. there is some offset being added during computune which is not present during normal play. the CV summer gets fed from the MOD circuit and X-MOD circuit. im not sure what the defaults are for these during computune, but if they are not zeroed out well, they could add an offset.
openmusiclabs.com

User avatar
donepe
Common Wiggler
Posts: 84
Joined: Sat Mar 21, 2015 3:41 pm

Post by donepe » Tue Jun 26, 2018 4:38 pm

thanks for the useful info! I will be replacing these caps for sure :)

DesolationBlvd
Common Wiggler
Posts: 54
Joined: Tue Oct 29, 2019 9:17 am

Re: Roland Jupiter 8 VCO1 drift

Post by DesolationBlvd » Tue Aug 04, 2020 9:48 am

Bumping this because I am now experiencing this issue with my Jupiter-8. The autotune always leaves one voice's VCO 1 noticeably flat (only objectionably so in the top octave), and swapping the card causes the flat voice to move with it.

Swapping out IC17 4053 (it returned from service recently, so they are socketed) from a good board to the wonky board had no effect.

Edit to add: I grabbed the CV right before the VCO in question, it's the same as any of the other VCOs.

DesolationBlvd
Common Wiggler
Posts: 54
Joined: Tue Oct 29, 2019 9:17 am

Re: Roland Jupiter 8 VCO1 drift

Post by DesolationBlvd » Tue Aug 18, 2020 7:14 am

Sorry to bump and double-post, but I just found another insight on my tuning issues. The bum oscillator is much closer to in-tune when it is set to pulse/square than when it is set to saw. I confirmed the tuning by using the master tune to get it dead-on with the square, and then switching it to saw made it go flat. The JP-8 VCO is saw core, but it looks like the square (VCO 1) / pulse (VCO 2) is used for the autotune. The control voltage input is not affected.

This discrepancy is not consistent across oscillators.

TL:DR - Any reason why changing the waveform would also throw the pitch off?

User avatar
guest
Super Deluxe Wiggler
Posts: 4511
Joined: Mon Aug 19, 2013 11:49 am

Re: Roland Jupiter 8 VCO1 drift

Post by guest » Tue Aug 18, 2020 12:56 pm

if its only going off at high notes, have you tried adjusting the VCO linearity trim?
openmusiclabs.com

DesolationBlvd
Common Wiggler
Posts: 54
Joined: Tue Oct 29, 2019 9:17 am

Re: Roland Jupiter 8 VCO1 drift

Post by DesolationBlvd » Tue Aug 18, 2020 9:31 pm

guest wrote:
Tue Aug 18, 2020 12:56 pm
if its only going off at high notes, have you tried adjusting the VCO linearity trim?
I tested again tonight. The oscillator is in tune at 220, 440, and 880 Hz when set to square, and is consistently flat at those same pitches when set to saw - the speed of the beating at each pitch seems to suggest the pitch is getting offset when set to saw somehow. It doesn't seem to be a linearity thing.

User avatar
guest
Super Deluxe Wiggler
Posts: 4511
Joined: Mon Aug 19, 2013 11:49 am

Re: Roland Jupiter 8 VCO1 drift

Post by guest » Tue Aug 18, 2020 10:46 pm

the square is derived from the saw, so this is a bit odd. can you take a probe point from the saw before the waveform switch? that way you can see if the drift only occurs when the waveform switch is activated (set to square, but listen to saw).
openmusiclabs.com

DesolationBlvd
Common Wiggler
Posts: 54
Joined: Tue Oct 29, 2019 9:17 am

Re: Roland Jupiter 8 VCO1 drift

Post by DesolationBlvd » Wed Aug 19, 2020 7:06 am

Probing at the inputs of IC7, the waveform selector, it does look like the saw is still flat (or rather, the square is still sharp compared to saw), regardless of what the selector is set to. At the very least, the pitch of the incoming waves is not affected to my eyes on the scope.

User avatar
guest
Super Deluxe Wiggler
Posts: 4511
Joined: Mon Aug 19, 2013 11:49 am

Re: Roland Jupiter 8 VCO1 drift

Post by guest » Wed Aug 19, 2020 1:59 pm

are you using a tuner or your ear? do any of the other sound different between square/saw? does this square sound brighter than a square on another voice?
openmusiclabs.com

DesolationBlvd
Common Wiggler
Posts: 54
Joined: Tue Oct 29, 2019 9:17 am

Re: Roland Jupiter 8 VCO1 drift

Post by DesolationBlvd » Wed Aug 19, 2020 3:11 pm

I used a tuner to play the 220/440/880 Hz test tones I tuned the square to (so I could determine that the saw was flat in comparison). I haven't compared the triangle.

I'll have to test the squares again, this time with the filter wide open.

User avatar
guest
Super Deluxe Wiggler
Posts: 4511
Joined: Mon Aug 19, 2013 11:49 am

Re: Roland Jupiter 8 VCO1 drift

Post by guest » Wed Aug 19, 2020 4:01 pm

the reason i ask, is because the square is derived from the saw, so if they simultaneously sound off, then perhaps its a perceptual thing. the presence of higher harmonics changes the percieved fundamental. if you look at them simultaneously on your scope, are they locked to one another?
openmusiclabs.com

DesolationBlvd
Common Wiggler
Posts: 54
Joined: Tue Oct 29, 2019 9:17 am

Re: Roland Jupiter 8 VCO1 drift

Post by DesolationBlvd » Wed Aug 19, 2020 8:47 pm

guest wrote:
Wed Aug 19, 2020 4:01 pm
the reason i ask, is because the square is derived from the saw, so if they simultaneously sound off, then perhaps its a perceptual thing. the presence of higher harmonics changes the percieved fundamental. if you look at them simultaneously on your scope, are they locked to one another?
What I meant was that I tuned the master tune such that the wonky oscillator was in tune, no beating, on the square, and then switching to saw gave me noticeable beating.

The saw and square weren't locked to each other.

User avatar
guest
Super Deluxe Wiggler
Posts: 4511
Joined: Mon Aug 19, 2013 11:49 am

Re: Roland Jupiter 8 VCO1 drift

Post by guest » Wed Aug 19, 2020 9:34 pm

if pin11 and pin14 on ic7 are not in sync, then there is something really wonky going on. can you post a scope shot?
openmusiclabs.com

DesolationBlvd
Common Wiggler
Posts: 54
Joined: Tue Oct 29, 2019 9:17 am

Re: Roland Jupiter 8 VCO1 drift

Post by DesolationBlvd » Mon Aug 24, 2020 5:42 pm

I took a closer look, and they were actually roughly in phase.
IMG_0349[1].JPG
IMG_0350[1].JPG
You do not have the required permissions to view the files attached to this post.

User avatar
guest
Super Deluxe Wiggler
Posts: 4511
Joined: Mon Aug 19, 2013 11:49 am

Re: Roland Jupiter 8 VCO1 drift

Post by guest » Mon Aug 24, 2020 6:18 pm

if you have a second channel on your scope, you can see both at the same time and get a really good idea of how in or out of phase they are. set the scope to "chop" mode (sometimes called different things), and trigger off the square wave. then select the sawtooth as the output and see if they stay in sync or not. its possible that the opamp driving the sqaure or sawtooth has some output stage issue that is really dependent on impedance, so switching to that output makes a difference, but if the pitch seems off at the analog switch pin, even when its not selected, then im not sure what it could be. what frequency did the beating seem to be when you compared to the master tune?
openmusiclabs.com

Don T
Super Deluxe Wiggler
Posts: 1282
Joined: Sun Feb 02, 2014 3:11 pm

Re: Roland Jupiter 8 VCO1 drift

Post by Don T » Tue Aug 25, 2020 2:02 am

Suggestion: Place the scope probe at the output of the VCO core, before the square wave shaper. Switch back and forth between saw and square to see if wavelength changes at all.

DesolationBlvd
Common Wiggler
Posts: 54
Joined: Tue Oct 29, 2019 9:17 am

Re: Roland Jupiter 8 VCO1 drift

Post by DesolationBlvd » Tue Aug 25, 2020 11:58 am

Looked at the output of pin 6 on IC 3A, and switching between waveforms had no effect.

Don T
Super Deluxe Wiggler
Posts: 1282
Joined: Sun Feb 02, 2014 3:11 pm

Re: Roland Jupiter 8 VCO1 drift

Post by Don T » Wed Aug 26, 2020 3:45 am

DesolationBlvd wrote:
Tue Aug 25, 2020 11:58 am
Looked at the output of pin 6 on IC 3A, and switching between waveforms had no effect.
Ok, scope or DMM, Probe pins 8, 9, and 11 of the 3046, and repeat same test for each pin, looking for any changes from saw to square.

Also, not sure if you have seen this site, but it may be useful:

https://jup8restoration.wordpress.com/2 ... rd-repair/

Post Reply

Return to “Music Tech DIY”