JUNO-106 PWM Duty Cycle Calibration Issue

sweetbeats · September 27, 2025, 1:47am

Add tags before posting: brand + model + (optional) fault
e.g., korg, models-korg-m1, no-power
If your model isn’t listed: use models–tbd (e.g., models-korg-tbd) and post the exact model in the pinned “Request a new model tag” thread. We’ll add it and retag.
When your issue is fixed, please mark the reply that solved it (click the ✓ “Solution” on that post).

• Model & revision/OS: Roland JUNO-106
• Symptoms (what changed, when): PWM Duty Cycle calibration: scope reveals combination sine/square wave when probing TP 8~13 rather than clean square wave. See attached pic. This is my first time calibrating this JUNO-106. It followed me home, in need of lots of love, it’s now been comprehensively refurbished/restored and I’m going through the calibration procedures. All good until the PWM Duty Cycle calibration.
• Tests already done (voltages, parts replaced): it’s been completely recapped, power supply refurbished/upgraded, completely refurbished panel board (new sliders, new tactile switches, etc), new key contacts, bender board repaired, it has all new AR voice and wave modules now installed…it’s been gone through. Voices are all working, I can hear duty cycle modulation when the PWM slider is adjusted as well as when VR31 is swept, but I can’t get an appropriate square wave to display on the scope. I double checked using my scope’s calibration test point and get a beautiful and appropriate square wave, so it’s not the scope or user error with the scope. IC17a pin 1 is a non-modulating sine wave…traced that back to pin 2, same…kept going upstream through IC18a pin 14, then pin 12 to IC23 pin 4…all non-modulating sine wave…any other suggestions?

• Photos / audio clips:

Travis · September 30, 2025, 4:26pm

Is the filter all the way up? If the filter were down that would explain why you’re seeing a mostly sine-like wave. TP8 is after the filter and VCA so that would make sense. Make sure you’re using the Test Program as described on page 18 of the service manual.

My other thoughts are: those signals on IC17, IC18 and IC23 should not be sine waves. If you are seeing a 60 Hz sine wave or some multiple of that frequency then there is a ground loop (or no ground connection) and you are actually reading modulation from the mains voltage. If that’s the case then you need to use a different ground point with your oscilloscope probe. I think there’s a test point somewhere on the board labeled Analog Ground or similar, attach your ground clip on the probe to that. Warning: never attach the ground probe to something that isn’t grounded, it can damage the synth and/or scope.

sweetbeats · September 30, 2025, 7:25pm

Thanks so much for the reply.

Yes I’m following the Service Notes to the ‘T’, so am in test mode, bank 5 is selected, PWM is switched on…basically everything on the control surface is set as prescribed in the chart on page 18, row 5 for the “PWM 50%” test. So for the filter the frequency slider is set to 10, and the keyboard slider is set to 10…everything else is 0 and the polarity is normal. Pic below.

I attached the scope probe ground reference to the analog ground test point in the upper left section of the module board, and I’ve verified this test point has a low impedance path to the chassis ground.

I’m going to re-check the waveforms at the relevant outputs and inputs of ICs 17, 18 & 23 again.

FWIW this is how the control section appears when I’m conducting the PWM duty cycle test:

Travis · September 30, 2025, 9:19pm

You’re welcome! Sounds like you’ve got it set up right. The outputs on ICs 17 and 18 should look more like DC voltages, no waveforms at all. They are the CV signals to control different parameters of the voices and should be a stable voltage corresponding to where the potentiometer is set on the front panel (unless the LFO is enabled for that parameter, then you would see the LFO there).

As a baseline you can measure the +12 V power on the voice board and make sure that reads as a steady 12 V DC voltage. If you see a sine wave there too then it’s definitely something to do with the oscilloscope. Perhaps try a different probe.

sweetbeats · September 30, 2025, 11:48pm

I haven’t rechecked the inputs/outputs of ICs 17, 18 & 23 yet, but I can affirm the +/-15V and +5V rails are rock stable all the way to the ICs’ respective power inputs. I spent a decent amount of time with the recapping and upgrades on the power supply…improved the heat transfer capability between the semiconductors at the heat sink, new power transistors for the +/-15V rails and new M5230L regulator, new +5V regulator, added crowbar protection at the outputs of the +/-15V rails, all calibrated and checked according to the Service Notes and is stable and happy under load…very low ripple at the outputs.

I’ll re-check the CV path back from the output of 17b with the scope and report back. BTW I understand what you are suggesting about the probe, but it’s a newer probe I’ve used on a ton of work, scope displays proper and beautiful square wave when I probe the scope’s calibrator. I don’t believe it’s the scope or the probe. But I’m keeping an open mind.

Again, thanks much for the engagement.

sweetbeats · October 1, 2025, 7:26pm

So I should be happy when a problem disappears, but I don’t like it when I don’t understand what changed. This is to say, after re-scoping the power rails, and the relevant inputs and outputs of ICs 17, 18 & 23 (which, BTW, now look like nice flat lines when AC coupled, instead of sine waves), the outputs of the voice modules now produce a nice square wave when doing the PWM duty cycle checks. I was able to successfully complete step 10 of the calibration procedures in the Service Notes. But I hate not knowing why the problem just disappeared. Previous to this I tried multiple scope modules in my Tek TM 500 setup, and multiple probes, all with the same faulty result. Now it works. Anyway, the only wrinkle is it looks like my duty cycle waveform is inverted. So for step 10-3, the 95% duty cycle check, of course the instructions present the waveform below:

My scope display looks as posted below, which looks inverted.

All controls on the control panel were set as per the Service Notes, including the filter phase control. I’m thinking I don’t need to worry about this, but, me being me I can’t just truly let it go so I’m mentioning it here.

sweetbeats · October 1, 2025, 7:26pm

synzoneau · October 5, 2025, 5:44am

You don’t need to worry about this. It’s got to do with the triggering on your scope. It looks fine

sweetbeats · October 7, 2025, 1:53am

Ahhh okay. Gotcha. Thank you.

So I finished all calibration steps including the chorus bias.

Question: when the chorus is switched on it’s definitely noisy. But I can really hear the noise in the headphone output. Is this normal? A. for it to be much more pronounced in the headphone output compared to the line outputs, and B. for it to be noticeably noisy when you switch either chorus mode on?

synzoneau · October 13, 2025, 2:08am

Yes it is totally normal for headphone output to be noisier than the line outs. You’ll find it’s the same on most synths. Whether the chorus is too noisy is a matter of preference, the BBD chips age with use and a little bit of noise is actually better than none but that’s my personal taste. If it sounds like a day on a surf beach when you engage it and you haven’t even pressed a key it’s time to change them. You get the reissue Xvive3009 from Cabintech they work and sound exactly the same but you must do the full calibration to get the best results.

sweetbeats · October 13, 2025, 3:34am

Thank you…very helpful.

When you state “full calibration” you’re not talking about the voice module calibration right? Just the calibration steps related to the chorus circuit? I just completed a full calibration, and I certainly understand needing to re-check/adjust the chorus bias if the BBD chips are replaced, but does that necessitate recalibration of the voice modules?

synzoneau · October 13, 2025, 4:05am

Yes just the chorus circuit needs to be recalibrated if you replace the BBD chips

sweetbeats · October 13, 2025, 4:31am

Cool. Thanks so much!

sweetbeats · October 26, 2025, 4:48am

FWIW I replaced the BBD ICs (8 & 10) with the Xvive MN3009 direct from Cabin Tech…decent price, fast ship, sounds great, and the noise floor with the chorus engaged is much lower. Thanks for the guidance.

Icetooth · February 17, 2026, 9:50am

Hi, nice work done and good to hear that the chorus replacement chips worked out so nice! I’m currently going through fixing some problems with a Juno-106 I came by. I wonder if you please can give a description on how you build the crowbar protection and how it was installed, any pictures? I would like to do something like that myself since I’ve read how much damage can be caused if the M5230L breaks. BR Icetooth

sweetbeats · February 22, 2026, 11:29pm

Yeah for sure. It’s really simple.

just replace the reverse polarity protection diodes D1 and D2 with 17V zener diodes (1N5354). The zeners will still provide the reverse polarity protection, but will also prevent output voltage from going over +/-17V.

Icetooth · March 2, 2026, 9:10am

Thanks sweetbeats! I know too little about this. I thought something like explained on Axotron would be needed to either blow a fuse or turn off the output. I wonder what will typically happen if the ML5230L failes? Will it mean that TR1 and/or TR2 would be fully opened, what will happen with the 1N5354 in such a case (and the rest connected to the ±15 railes), and how will I know that something is wrong? BR Icetooth

sweetbeats · March 4, 2026, 7:25pm

I respectfully disagree with your input voltage calculations. I’m not sure how you arrived at 17.1VDC input voltage…

To calculate AC RMS voltage to DC volts you multiply the AC RMS volts by the square root of 2 (which is 1.414). And keep in mind while the voltage from the mains transformer secondary to the center tap is documented as 19VAC, the swing from secondary to secondary is then 38VAC. That 38VAC to rectified raw DC then is about 53.7VDC. And this assumes your mains voltage is the rated standard. The schematic assumes my mains voltage is 117VAC when in fact it’s usually 121-123VAC, sometimes even 124VAC. So that means, in my case, the rectified raw DC could be as much as 57VDC. Yes there is some voltage drop across the bridge, and there may be minimal voltage drop across the main filter caps, but with the total drop across both sides of the bipolar input, it is still probably about 52VDC. I could measure but can’t easily open my JUNO-106 up right now. So what that means though is the Vcc+ and Vcc- inputs are each seeing, in theory, about 26VDC. So that’s what could pass to downstream systems if the regulator fails…and yes that all depends on how it fails and I don’t know the common failure modes for the M5230L. But regardless it seems like super cheap insurance to put the zeners in and ensure the maximum output from the +/-15V supply is +/-17V.

Icetooth · March 4, 2026, 8:02pm

Trying to learn and understand, hope you have patience with me. As far as I can understand the Zener Diod would lower the audio DC to 17V if it starts to exceed 17V after the M5230L (looking on the positive side of it). It does this by opening up a current flow from Cathode to Anode. Now what I like to understand is will it make the fuse (F2 and F3 for the negative part) to blow and thus hopefully saving the downstream systems? Can the 1N5354 handle all the current without blowing itself before the fuse does? Also I wonder if one could use 1N5353 16V instead or is it too close to 15 so with tolerance in regard it might be on the trigger to early and drain some current?

sweetbeats · March 5, 2026, 4:15am

“As far as I can understand the Zener Diod would lower the audio DC to 17V if it starts to exceed 17V after the M5230L (looking on the positive side of it). It does this by opening up a current flow from Cathode to Anode.”

Yeah something like that. It might be more accurately stated the zeners limit the maximum voltage at the output of the +/-15V bipolar supply to +/-17V.

The M5230L is a dual-tracking bipolar voltage regulator. Its job is to ensure stable and constant output voltage, in this case +/-15V. It also does some ripple filtering. If it fails the voltage could go low, or go to zero, or it could be unstable, or it could go high, and as mentioned in an earlier post the limit of that high voltage is up to whatever it is at the input, which could be, in the case of the JUNO-106, as much as +/-26 to 27V. The zener diodes ensure the output voltage can be no more than +/-17V.

“Now what I like to understand is will it make the fuse (F2 and F3 for the negative part) to blow and thus hopefully saving the downstream systems?”

Installing the zener diodes have nothing to do with the current protection fuses F2 and F3.

There are several things that need to be safeguarded:

Reverse polarity (making sure the + rail can’t go - and vice-versa). The output diodes (whether rectifiers or zeners) do this.
Over-voltage…ensuring the output voltage cannot exceed a safe level. Installing the zener diodes in place of the rectifier diodes at the outputs accomplishes this.
Over-current…ensuring that, in a failure state where the powered system or systems draws too much current, which can permanently damage the system components electronically or mechanically, the system is cut off from power. Fuses blow because the current drawn by the device is too much.

F2 and F3 will blow if there is too much current flowing, regardless of the output voltage. The are in the system for that reason and adding extra protection to limit output voltage does not change how they will respond to an over-current condition. Nor will an over-voltage condition necessarily include an over-current condition. Electrical components can be damaged if the system voltage exceeds their limits, or the amount of current exceeds their limits. Installing the zener diodes adds protection to prevent the regulator from allowing an over-voltage condition in a failure state.

“Can the 1N5354 handle all the current without blowing itself before the fuse does? Also I wonder if one could use 1N5353 16V instead or is it too close to 15 so with tolerance in regard it might be on the trigger to early and drain some current?”

The 1N5354 is a 5W part. That means it’s rated for about 300mA current at 17V…about 330mA at 15V. The fuses F2 and F3 are supposed to be 500mA rated. So now you might be thinking the zener diodes will fail before the fuses blow in the event of an over-current condition. But that’s not likely the case. Electrical components are usually rated to last some amount of time at their maximum electrical and environmental limits. Like an electrolytic capacitor might be rated to last 1,000, 2,000, even 10,000 hours at its maximum rated electrical limits like voltage, current, ripple, and maximum environmental conditions like temperature…which could be 85, 105, 125 or more degrees celsius…and components may last for short periods of time above these limits, since they can tolerate long periods of time at those limits. The fuses, however, are designed to fail instantly when their limit is breached. The fuses are designed to fail before anything downstream fails. Including the 1N5354 parts.