Jupiter 6 resonance hack

The Roland Jupiter-6 Analogue Synthesizer
User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Jupiter 6 resonance hack

Postby mima85 » Thu Apr 05, 2012 2:52 pm

Hello everybody. My name is Michel and I'm new on this forum.

First of all I apologize for my bad engRish, but I'm italian so english it's not my mother tongue.

I have a quite technical question.

Since some days I'm the proud owner of a Roland Jupiter 6.

It's a great machine and I like it so much, but I'd like to know if there's something that can be done to make the resonance eating less the sound's body. I know this is relative to the 24db/oct filter, which have a strong cutoff slope, and to the filter circuitry implementation, but I'd like to do something, if possible, to achieve this aim.

I have to recalibrate the instrument and I'm used to do these things (I restored/recalibrated myself almost every keyboard I have in my studio), and reading in the service manual I saw that there's a specific resonance tuning procedure (the 4th step on the relative chapter), on which there's a hand-drawn waveform which looks like a filtered squarewave with strong resonance. On the "WHAT ADJUSTED" field it's stated "Amount of feedback for a proper regeneration". The test point where it's requested to hook the scope it's the R21 on the jack board or directly the output jack, so I think that this tuning it's being done directly on the sound the synth produces.

The calibration instructions say that the 1st oscillation immediately after the wave's rising/falling edge must touch the 0v line on the oscilloscope.

So the main question is: if I tune the calibration pot to make that oscillation farther from the 0v line than closer (so I think reducing the amplitude of the resonating oscillations), there's a possibility to reduce that "body eating" effect when having high resonance? Doing this could affect the correct working of the machine?

Comparing on my oscilloscope/spectrum analyzer software the filtered waveforms of my Jupe and my Juno 106 having the filter set in the same way I could see that at high resonance settings the Juno's waveforms maintain their basic shape (so the sound have more body), while the Jupe's waveforms "flatten out" too much. Hence my idea.

If this sort of "hack" is applicable I don't worry about loosing some of the synth's original sound character (if the synth will sound in a better way, of course) as it's reversible simply by tuning the trimmers like said in the service manual.

I'll do the calibration procedure in the next days, in the meantime I'd be curious to read some of your experiences, if someone did something similar.

Thanks to everybody :D

davemoog
Posts: 4
Joined: Fri Sep 23, 2011 9:10 pm

Postby davemoog » Sun Apr 08, 2012 3:21 pm

Sorry I don't know the value as it was done by experimenting. But I think (it's been a while) this is what I did. I added a resistor between the centre and the ground on the resonance slider. This does not reduces the amount of resonance at maximum but changes the curve of the control voltage so that you have more control of the amount of resonance through the lower positions on the slider - resonance comes on slower.

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Sun Apr 08, 2012 6:37 pm

I did right now the modification I was talking about.

Well, I can say I'm quite satisfied of the results :)

Now (for me), while remaining aggressive, the resonance it's much more usable, and to my ears that modification even changed slightly the whole way the filter behaves, also without resonance, making it warmer and smoother. The machine now have more punch also on bass sounds.

The only drawback, for people which need them, is that are no longer possible those screaming, ear-piercing, extremely high resonance sounds, but I don't care about because I don't use them. Also because sounds where the resonance shines are still possible, and on high reso sweep sounds I could hear a very little but pleasant overdrive on the resonated harmonics when the filter reaches lower cutoff frequencies.

So if you want your Jupiter 6 to sound a bit more lush, smooth, warm and punching on basses, just follow the 3rd step on the calibration procedure, but instead of making the resonance oscillations touching the 0v line on the oscilloscope, make them going farther. You'll also see on the scope that doing this, while reducing the resonance oscillations, will increase the amplitude of the squarewave's "body". Exactely what I had in mind.

Make sure that all the voices are being calibrated in the same way, comparing their waveforms in the oscilloscope and adjusting them to look the same. Eventually try different adjustments to get a proper (for you) sound.

Keep in mind that the resonance still lowers the overall volume of the sound, but with this modification it eats less sound's body, so just turn the volume pot higher when using high-resonance sounds :)

Have fun with your Jupes :D

nebula
Posts: 87
Joined: Sat Feb 24, 2007 3:38 am

Postby nebula » Wed Apr 11, 2012 2:38 am

Are you sure your mod didn't just reduce the overall resonance?

I read somewhere (maybe here) that the JP-6 was designed to have excessive parameter ranges. To my ears, the resonance on the JP-6 at full is about twice as much as you can get from other synths. When I want punchy bass I remember that and just don't turn the res up so high.

Same thing with the clicky envelope. Adjust your attack so it just disappears and you'll have roughly the fastest attack available on many other synths.

The downside to all this is that all of the sliders are WAY more responsive than most other synths. Programming the JP-6 is an exercise in subtlety - if you really move those sliders a lot you are most certain to blow right past the sound you are looking for.

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Wed Apr 11, 2012 7:37 am

nebula wrote:Are you sure your mod didn't just reduce the overall resonance?


Yes, my mod reduced the overall resonance, but from oscilloscope observations I could see that the waveforms, while still being flattened out when using high resonance on the filter, aren't being longer flattened out so much like the synth does with the factory calibration, when using the same amount of resonance (not according by the slider's position but by the spectrum analyzer). So there's a real gain in sound's body and not only a mere change in the reso's slider range. That change in filter behaviour can even be immediately seen on the oscilloscope while doing the calibration, as stated previously:

mima85 wrote:You'll also see on the scope that doing this, while reducing the resonance oscillations, will increase the amplitude of the squarewave's "body". Exactely what I had in mind.


This gain in sound's body is also present when using the filter without resonance. The first sound I made when I got the Jupe was a simple brass patch (without resonance), and I was surprised how it sounded better, fuller after doing the mod without doing any change to the patch itself. This also happened on other sounds without resonance I made, including basses, which are now punchier.

Keep in mind that this mod can be done according to everyone's taste, if one still wants an "intrusive" resonance just have to turn the trimmers accordingly. And restoring the synth to its factory settings it's just a matter to recalibrate it following the service manual's instructions.

So experiments can be done to reach everyone's ideal sound.

nebula wrote:I read somewhere (maybe here) that the JP-6 was designed to have excessive parameter ranges. To my ears, the resonance on the JP-6 at full is about twice as much as you can get from other synths.


That it's true, maybe you read it on gearslutz, on this or this thread.

nebula wrote:When I want punchy bass I remember that and just don't turn the res up so high.


As above.

nebula wrote:Same thing with the clicky envelope. Adjust your attack so it just disappears and you'll have roughly the fastest attack available on many other synths.


Yes actually there's a mod that prevents this clicking (by forcing a slightly longer attack time to shut up the clicks) but I don't want to do it, as very fast attack sounds are no longer possible with this mod and to restore the original status you have to re-solder in place the original capacitors (or components with the same values).

nebula wrote:The downside to all this is that all of the sliders are WAY more responsive than most other synths. Programming the JP-6 is an exercise in subtlety - if you really move those sliders a lot you are most certain to blow right past the sound you are looking for.


True, actually if the only benefit with this simple mod would be a mere change in the resonance's slider range I'd restore the synth to the factory tunings. But I could see and hear real (for me) improvements in the sounds, so I thought that sharing this could be useful for people who would their Jupes sounding more warm.

nebula
Posts: 87
Joined: Sat Feb 24, 2007 3:38 am

Postby nebula » Tue Apr 17, 2012 1:41 am

... and I'm sorry if I seemed dubious .... it sounds like you know what you're talking about. I didn't mean to get all soapboxy.

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Tue Apr 17, 2012 6:39 am

Don't worry, it's all right :D

Just answered to your questions, I admit sometimes I dwell too much while answering, but it's just because I want to be as much informative as I can. I didn't want to argue, trust me :wink:

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Wed May 09, 2012 11:09 am

I just found another mod (this time some soldering works must be done) here:

http://fa.utfs.org/diy/rolandjp6/

There are a couple of resistors to be soldered, it looks very interesting.

It seems that it works on the VCA control voltage by sending to it part of the resonance CV thru a 1 MOhm resistor, to make the amp to reduce less the overall sound's volume.

In the next times I'll try it to see how it works together with my custom resonance tuning.

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Thu May 10, 2012 7:31 pm

I did the mod, then some tests. I could hear real improvements in the filter's response, now the sound isn't being longer lowered in volume as much as like an unmodified Jupiter 6 does. And with my custom tuning the sound looses less body than with the factory specified tuning.

Here you can find a wave recording a filter test: http://dl.dropbox.com/u/2345020/jup6_filter_test.wav

Initial settings are:

- VCO: 1 sawtooth alone
- VCF: cutoff and reso at minimum, EG intensity at maximum
- VCF EG: Decay about at half slider, all the rest at minimum
- VCA EG: flat envelope

The test recording consists of 4 filter sweeps:

- First sweep: cutoff to minimum, rising and lowering the resonance alone
- Second sweep: resonance at maximum, rising and lowering the cutoff
- Third sweep: resonance at half, rising and lowering the cutoff
- Fourth sweep: resonance at miminum, rising and lowering the cutoff

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Sat Nov 10, 2012 12:16 am

a little update on this: instead of a 1 M resistor I used 2x 820 K in parallel, so it's like having a 410 K resistor. Well, the results are a lot better! Now there's no more thin out at all, just be careful with extremely high resonance settings because when the filter resonates, especially on low frequencies, could overdrive the VCA and the signal starts to clip. But fixing this it's just a matter to reduce the VCA level for patches which use the filter in this way.

I also did the VCA smoothing circuit mod described here which removes the annoying envelope's click noises and improves a bit the synth's bass response.

I'm very satisfied of my improved Jupiter 6!

User avatar
Gil Sicuro
Posts: 97
Joined: Mon Jan 14, 2008 3:03 pm

Postby Gil Sicuro » Sun Nov 11, 2012 3:05 am

mima85,

How farther is the resonance wave from the 0v line on the scope, in volts?

I was thinking myself about another mod on the JP6 - what about patching the triangle wave output from the VCO IC and making one octave below? That would make a programmable sub oscillator.

Oh, can you post a sample of yout non-resonant brass patch?

cheers

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Sun Nov 11, 2012 10:01 am

Gil Sicuro wrote:How farther is the resonance wave from the 0v line on the scope, in volts?


I don't remember, I should redo the measurings but I have to wait for my friend (which is an electronic engineer) to have some spare time to le me using his oscilloscope.

Anyway I did this job in a very empirical way, just by pushing out the resonance as farther as the trimmers could. As there were some very slight differences between the trimmers, I searched which was the trimmer which kept the resonance closer to the 0v line and I tuned all the others according to it, in order to having the waveforms looking the same for all the voices, to prevent differences between them.

Gil Sicuro wrote:I was thinking myself about another mod on the JP6 - what about patching the triangle wave output from the VCO IC and making one octave below? That would make a programmable sub oscillator.


Hmm I'm not sure this is a good idea, because if for any reason a triangle together with another waveform playing the same note on one oscillator is needed, this wouldn't be longer possible. It depends by what you do with the synth and what are your requirings, but I wouldn't suggest this as a mod to be done by all to improve the Jupiter 6, it's too dependent to everyone's tastes.

Gil Sicuro wrote:Oh, can you post a sample of yout non-resonant brass patch?


I'll post it soon.

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Mon Nov 12, 2012 9:16 pm

Here's the brass sound test. In this recording there are 6 samples of the same patch at different levels of resonance. The first 3 are within the first half of the slider, the 4th at the middle, the 5th at 3/4 and the 6th at full resonance.

This track was entirely recorded keeping the synth at the same volume level, the only thing I change is the resonance level.

As you can hear there's no longer the thinout of the sound's body while raising the resonance but instead a slight reduction of it, and in general a volume increment because of the resonance. And I prefer a lot more this instead the automatic volume lowering done by an unmodified Jupiter 6. At least it's me that I have to turn down the volume or the VCA level if the signal clips.

I also did a test with a pad/string sound here. There are 2 sounds based on the same patch, one with low cutoff (the pad) and one with the filter opened (the strings). There's just a slight touch of resonance to give the sound more contour. As before the volume level wasn't touched between the 2 samples.

So now who can say that the Jupiter 6 it's not able to do lush pad/strings? :wink:

No effects or EQs are used at all. Just the Jupiter 6 alone, recorded in mono.

analog1
Posts: 12
Joined: Fri Nov 07, 2008 10:16 pm

Postby analog1 » Wed Nov 28, 2012 7:52 pm

Nice sounds and presets, imagine the same sounds through a roland chorus or evendide effect !

User avatar
mima85
Posts: 35
Joined: Thu Apr 05, 2012 2:06 pm

Postby mima85 » Mon Dec 03, 2012 12:43 pm

My Jupiter 6 thru a Juno 106 chorus? Sweeeeeeeeeet :mrgreen:


Return to “Jupiter-6”

Who is online

Users browsing this forum: No registered users and 0 guests