Kronos - external control of drawbars question

[ABECK]

I'm considering a Kronos 73, but I want to check my assumptions about some of the external control options. If I were to set up a combi using the CX 3 engine for one of the parts, I'm fairly certain I could set that part to a specific MIDI channel and control it from an external controller. I was wondering if I we're to use something like an Axiom controller, could I map those drawbars to the CX3 part? If so, would I still be able to use the faders on the Kronos to control parameters for the other parts? It doesn't seem like an odd assumption, but I thought I'd check it out.

[J. Dan]

My understanding is that it requires SysEx to control them, not simple MIDI CC's. I haven't looked at it in detail.

[allan_evett]

It's SysEx. I considered doing that, at one time. But I wondered: after spending the money for a Midi Solutions box, would the SysEx commands address the CX3 engine within a Combi, or only in Prgram mode. All Korg tech support could say at that time was, 'try it'....

 

Anyway, Busch figured out the exact codes and sent them to me. I'd be glad to PM them to you, if you wish.

[OB Dave]

I always thought it was SysEx too - if you Google you can find a few references to this - but a very reliable source recently told me otherwise: that the CX 3 engine (and therefore the Kronos) will respond to regular MIDI CCs. I would love for somebody to test this out to confirm this. My understanding is that the CX-3 and Kronos drawbar CC mapping is: Drawbars 1-9 are controlled by MIDI CCs 16-24 (decimal). This is the same CC mapping used by the Nord Electro upper manual, and both manuals of the C1/C2.

 

If anyone can confirm the Kronos responds to these, I'll buy you a beer. And the boys in the RedKey lounge can attest to my good taste in beers.

[wmp]

No free beer for me. Google let me down and I'm lazy. If I had a Kronos I'd check with a MIDI monitor like Midi-Ox. Drawbars should be simpler than they are. Various controller schemes are enough without all the sysex and nrpn stuff. Controller to nrpn mapping requires more than filtering and packing. You have to remember the state of all drawbars and combine one incoming controller with the state of the rest before transmitting. Possibly tricky business in heavy traffic.

This Midi Solutions box may or may not have all it takes to program it. It may or may not get confused.

 

Perhaps the Ocean Beach DB-2 will do all that.

[burningbusch]

This is what I sent to Allan. Controllers from Novation and M-Audio, possibly others, do support sending of SysEx, but you need to double check specs. I see that what I sent Allan was related to upper drawbars. Controlling the lower is going to look similar save a byte or two difference.

 

OK, I pulled these from the Kronos using a MIDI Monitor. I've been fooling around with an iPad app called MIDI Touch and I've verified most of these. You will see that the second to the last byte pairs is the variable. They all need to be 1 byte variables.

 

[basic] Leakage Level F0 42 30 68 43 0B 7F 00 04 00 00 00 xx F7 (x=0 - 99)

 

[Percussion] Enable (Default: Off) F0 42 30 68 43 1D 00 00 06 00 00 00 xx F7 (x=0 - 1)

 

[Percussion] Level F0 42 30 68 43 1D 00 00 06 01 00 00 xx F7 (x=0 - 1)

 

[Percussion] Decay F0 42 30 68 43 1D 00 00 06 02 00 00 xx F7 (x=0 - 1)

 

[Percussion] Harmonic Switch F0 42 30 68 43 1D 00 00 06 03 00 00 xx F7 (x=0 - 1)

 

[Vibrato/Chorus] Upper Enable (Default: Off) F0 42 30 68 43 1D 00 00 06 04 00 00 xx F7 (x=0 - 1)

 

[basic] Expression Level F0 42 30 68 43 1D 00 00 05 02 00 00 xx F7 (x=0 - 99)

 

[Rotary Speaker] Rotary Switch (Default: On) F0 42 30 68 43 1D 00 00 06 0D 00 00 xx F7 (x=0 - 1)

 

[Rotary Speaker] Rotary Speed (Default: Slow) F0 42 30 68 43 1D 00 00 06 0F 00 00 xx F7 (x=0 - 1)

 

[Drawbar] Upper Drawbar1 Level F0 42 30 68 43 1D 00 00 04 00 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper Drawbar2 Level F0 42 30 68 43 1D 00 00 04 01 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper Drawbar3 Level F0 42 30 68 43 1D 00 00 04 02 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper Drawbar4 Level F0 42 30 68 43 1D 00 00 04 03 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper Drawbar5 Level F0 42 30 68 43 1D 00 00 04 04 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper Drawbar6 Level F0 42 30 68 43 1D 00 00 04 05 00 00 xx F7 (x=0 - 8)

 

[Drwabar] Upper Drawbar7 Level F0 42 30 68 43 1D 00 00 04 06 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper Drawbar8 Level F0 42 30 68 43 1D 00 00 04 07 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper Drawbar9 Level F0 42 30 68 43 1D 00 00 04 08 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper EX Drawbar1 Level F0 42 30 68 43 0B 06 00 00 09 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper EX Drawbar2 Level F0 42 30 68 43 0B 06 00 00 0A 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper EX Drawbar3 Level F0 42 30 68 43 0B 06 00 00 0B 00 00 xx F7 (x=0 - 8)

 

[Drawbar] Upper EX Drawbar4 Level F0 42 30 68 43 0B 06 00 00 0C 00 00 xx F7 (x=0 - 8)

 

Busch.

[OB Dave]

Perhaps the Ocean Beach DB-2 will do all that.

Well, that's actually why I chimed in. I recently added DB-1 support for the weirdo Hammond Suzuki MIDI scheme. Support for Korg SysEx could be also be added but first I wanted to confirm whether it's really necessary since I'm almost out of code space on the DB-1's little microcontroller.

 

I dunno about chorus/vib/percussion/etc, but my understanding is that for drawbar commands, the Korg will respond to either Sysex or MIDI CCs 16-24. If that's the case, my work is done. Would be great to confirm that.

[wmp]

You will see that the second to the last byte pairs is the variable. They all need to be 1 byte variables.

 

They can't be. That's sysex. F0 and F7 enclose a sysex message. The next few bytes are manufacturer id (historic midi DOH!), then model, device id, and other envelope information.

 

The whole drawbar set is encoded into one byte every time you move any drawbar and sent in a sysex message.

[burningbusch]

You will see that the second to the last byte pairs is the variable. They all need to be 1 byte variables.

 

They can't be. That's sysex. F0 and F7 enclose a sysex message. The next few bytes are manufacturer id (historic midi DOH!), then model, device id, and other envelope information.

 

The whole drawbar set is encoded into one byte every time you move any drawbar and sent in a sysex message.

 

Obviously it's SYSEX. I stated that above. Why would the entire drawbar set (I assume you mean the state of nine drawbars) need to be sent every time? THAT makes no sense. While it's true initially when the controller and Kronos first start they would not be in sync, but after the drawbars are moved they would be in sync and remain so for the duration using the individual SYSEX strings I posted above.

 

You can argue with this but I verified with a MIDI monitor and M-Audio controller.

 

Busch.

[AnotherScott]

Perhaps the Ocean Beach DB-2 will do all that.

Well, that's actually why I chimed in. I recently added DB-1 support for the weirdo Hammond Suzuki MIDI scheme. Support for Korg SysEx could be also be added but first I wanted to confirm whether it's really necessary since I'm almost out of code space on the DB-1's little microcontroller.

 

I dunno about chorus/vib/percussion/etc, but my understanding is that for drawbar commands, the Korg will respond to either Sysex or MIDI CCs 16-24. If that's the case, my work is done. Would be great to confirm that.

This reminds me that there was some discussion somewhere about there not being a way for you to create a cost effective version of your device with dedicated controls for percussion and such. I was thinking, though, for controlling something like VB3 where any way to get physical access to percussion settings would be desirable, maybe the faders could even be repurposed. I'm thinking of perhaps a mode where, if the top button is held down (or toggled), the drawbars send out alternate CC commands... like, the first drawbar would be Percussion On/Off, the second would be Soft/Normal, third would be Fast/Slow, fourth would be 2nd/3rd, where you would move the drawbar up to position 0 for the rocker's "up" position and down to 8 for its "down" position. Additional drawbars could be used for other parameters a user could define, key click, whatever. Maybe one drawbar could handle C/V, where 0 is off, and pulling the drawbar out could move you through the 6 possible settings, and maybe the user could define full out to be his favorite one.

[Jamie Finstad]

Perhaps the Ocean Beach DB-2 will do all that.

Well, that's actually why I chimed in. I recently added DB-1 support for the weirdo Hammond Suzuki MIDI scheme.

 

Does that include the SK1? I'm trying to figure out how to use the DB-1 to control the lower manual on the SK-1.

 

Jamie

[wmp]

A closer look at the hex indicated that byte ten is probably your drawbar number. I just took a quick look at the sysex messages and thought Korg might be doing it like Hammond. The XK-3 only uses nine values for nine drawbars and transmits the state of all nine drawbars in an nrpn message. That'd be difficult and require an external controller to remember the state. It seems the Kronos is a simple matter of mapping sysex to controllers. That may not be necessary if the Kronos can be set to send controllers.

 

For all I know it could be a law that no two manufacturers are allowed to do it the same way. It takes highly specialized drawbar engineers to sort all this stuff out for every make and model. I think that's Dave.

[burningbusch]

That's cool wmp. I need to clarify, I tested using Kenton Control Freak Live (discontinued). I don't know if M-Audio would work in this situation as you need some flexibility when working with sliders/knobs where a variable is involved. The Kenton gives really complete control. Just because the controller supports SYSEX doesn't mean it's going to work in this situation.

 

Busch.

[allan_evett]

Perhaps the Ocean Beach DB-2 will do all that.

Well, that's actually why I chimed in. I recently added DB-1 support for the weirdo Hammond Suzuki MIDI scheme. Support for Korg SysEx could be also be added but first I wanted to confirm whether it's really necessary since I'm almost out of code space on the DB-1's little microcontroller.

 

I dunno about chorus/vib/percussion/etc, but my understanding is that for drawbar commands, the Korg will respond to either Sysex or MIDI CCs 16-24. If that's the case, my work is done. Would be great to confirm that.

 

Very cool that, with the configuration software, the DB-1 will control anything that receives MIDI CCs. In addition to being a very useful with my Stage 2, live, it would be great to have a DB-1 (or more likely, DB-2 ?) control the TW organs in my Jupiter 50 - when using that 'board as my 2nd tier. AFAIK, Roland still hasn't assigned any CCs to the drawbar parameters in the JP-50; it was hinted at, for OS updates, when I spoke to tech support not too long ago. The tech I spoke with also wondered about SysEx; said that they were still waiting for all of the JP-50 SysEx documentation to come in from Japan.

 

SysEx support for both the Kronos and JP-50 would be cool. If it took creating a DB-2 to make that happen, I'd likely be first in line to purchase one. Having a box that could control multiple clone engines - both hardware and software, via MIDI CC and SysEx would be a fantastic tool.

[burningbusch]

Perhaps the Ocean Beach DB-2 will do all that.

Well, that's actually why I chimed in. I recently added DB-1 support for the weirdo Hammond Suzuki MIDI scheme. Support for Korg SysEx could be also be added but first I wanted to confirm whether it's really necessary since I'm almost out of code space on the DB-1's little microcontroller.

 

I dunno about chorus/vib/percussion/etc, but my understanding is that for drawbar commands, the Korg will respond to either Sysex or MIDI CCs 16-24. If that's the case, my work is done. Would be great to confirm that.

 

The manual for the CX3 organ shows that it receives/sends:

Upper drawbars CC# 16 24

Lower drawbars CC# 70 78

 

Plus all of the other functions (chorus, percussion, etc.) send/receive CC#s. I think we're looking at a situation where even though the sound engine is the same between the CX3 and the Kronos, the MIDI implementation is not. I could be that Korg provided this additional CC# support as the CX3 is a dedicated organ module.

 

I will try to get the Kronos to respond, but I'm not too hopeful.

 

Busch.

[AnotherScott]

Yes, I think the "real" CX3 uses standard MIDI for drawbars, and while the Kronos uses the same tonewheel emulation modeling algorithms, that "engine" is still being implemented on a different electronic and physical platform, and there are differences in various logistical aspects. Other differences are that the Kronos doesn't support the shallow triggering; that, by default, the footswitch does not toggle rotary speed as it does on a real CX3; the various buttons to enable percussion and such aren't set up to illuminate on enable and go dark on disable, etc. So it would not surprise me if the MIDI implementation were different as well.

[Kurt W]

OB Dave - Kronos only use sysex for drawbar control. Currently I am using a Voce Midi Drawbar to control the upper manual in an organ combination on the Kronos. The midi cc's from the VMD are converted via a Midi Solutions Event Processor to sysex. Advantage of the Voce controller is that it has buttons for percussion and V/C in addition to the drawbars. Disadvantage is that the Kronos does not provide (or not enough) power via midi to supply the Event Processor. This means that I need the additional Midi Solutions power adapter - 3 boxes to control the upper manual of the organ. Kronos is a fantastic workstation but not the performing tool I hoped for.

 

Also have your Ocean Beach DB1 and think it would be a better tool than my current setup i) if it handled the Kronos sysex drawbar implementation and ii) implemented support for both upper and lower manuals. For Nord equipment (at least Stage 2) the DB1 top button switches between the A and B slots (manuals). This is visualized on the Stage 2. For Kronos you should use the LED besides the top button to indicate whether the drawbars currently controls the upper or lower manual.

 

For me the Kronos control surface is quite bad for live stage use, especially for a combination including a dual manual organ. I am sure many others experience the same

and would be potential buyers of DB1 if you could implement this. Any chance you doing it?

[burningbusch]

Disadvantage is that the Kronos does not provide (or not enough) power via midi to supply the Event Processor. This means that I need the additional Midi Solutions power adapter - 3 boxes to control the upper manual of the organ.

 

The MIDI Solutions boxes are powered at their MIDI IN ports. The problem is not the Kronos, it's the Voce. Many of these small controllers don't provide power on their MIDI OUTs. To test, plug the MIDI out of the Kronos into the MIDI IN on the MIDI Solutions box. It will light right up.

 

Busch.

[Kurt W]

You are correct, the Voce is the problem as it is connected in front of the processor.

 

Just checked the Event Prossor directly connected to Kronos and it works fine. Sorry....

[burningbusch]

You are correct, the Voce is the problem as it is connected in front of the processor.

 

Just checked the Event Prossor directly connected to Kronos and it works fine. Sorry....

 

But you bring up a good point. If you need to use the MIDI Solutions box you might also need their power unit.

 

Busch.

[ABECK]

Wow - so, it sounds like the MIDI Solutions box would be necessary to make this happen. Good to know. A second keyboard is not critical for me, but would certainly make some scenarios easier.

 

Thanks - still deciding if moving to the Kronos makes sense.

[OB Dave]

Wow. This is what I get for not logging in for a few hours! Thanks for all the responses - I'll attempt to address them all here.

 

This reminds me that there was some discussion somewhere about there not being a way for you to create a cost effective version of your device with dedicated controls for percussion and such. I was thinking, though, for controlling something like VB3 where any way to get physical access to percussion settings would be desirable, maybe the faders could even be repurposed. I'm thinking of perhaps a mode where, if the top button is held down (or toggled), the drawbars send out alternate CC commands... like, the first drawbar would be Percussion On/Off, the second would be Soft/Normal, third would be Fast/Slow, fourth would be 2nd/3rd, where you would move the drawbar up to position 0 for the rocker's "up" position and down to 8 for its "down" position. Additional drawbars could be used for other parameters a user could define, key click, whatever. Maybe one drawbar could handle C/V, where 0 is off, and pulling the drawbar out could move you through the 6 possible settings, and maybe the user could define full out to be his favorite one.

It's probably possible to do this, but I'm reluctant to do so. I don't think operating it would be very intuitive. A function that wants to be controlled by a button, really ought to be controlled by a button. Whatever its limitations, at least the DB-1 is dead simple, intuitive, and totally bulletproof. I appreciate the suggestion - but it just seems weird to me to toggle drawbars to suddenly not be drawbars anymore but instead control percussion, vib, etc. I think it would drive me crazy.

 

I do think there's a need for a drawbar controller with full percussion/vib/chorus/expression pedal input/leslie toggle/etc. I guess that's what the B4D (made by Boehm) was supposed to have been, but it wasn't programmable, didn't use a standard expression pedal input, wasn't MIDI bus powered, and so on. The biggest issue for me was that if you changed patches it couldn't re-synch with your VSTi, which made it difficult to use in live performance.

 

I'm thinking such a device needs a USB interface in addition to DIN5 MIDI. I have some cool ideas to make it easier to use in live performance, I just don't know if I could bring the thing in at a reasonable cost. At one point I did start on a bill of materials but then got sidetracked with my regular consulting gig. I guess I should take that back up and see where it leads me. My gut feeling is that the thing would come in at around $400-450 retail. That seems high, but then the street price of the XM-2c is in the low $400s. So, I just dunno.

 

Does that include the SK1? I'm trying to figure out how to use the DB-1 to control the lower manual on the SK-1.

Yes. Actually, the request for this feature was specifically to allow you to use a DB-1 to control the SK1 and SK2 lower organ manual. This firmware has already been written and I'm in the middle of testing it. If you're interested in helping beta test it, please PM me your email addy.

 

For all I know it could be a law that no two manufacturers are allowed to do it the same way.

Ha, tell me about it! It's so unfortunate none of this stuff was ever standardized. What a mess. A colleague has been deeply involved in the development of the new HD MIDI spec that was just demonstrated at the MMA meeting, and several months ago I gave him a proposal for standardizing tonewheel organ controls. I guess I should follow up and make sure that made it in.

 

OB Dave - Kronos only use sysex for drawbar control.

OK, this is very good to know. I'll going to make the addition of Korg SYSEX support to the DB-1 a priority. Would you be able to help test the code when it's finished?

 

Also have your Ocean Beach DB1 and think it would be a better tool than my current setup i) if it handled the Kronos sysex drawbar implementation and ii) implemented support for both upper and lower manuals. For Nord equipment (at least Stage 2) the DB1 top button switches between the A and B slots (manuals). This is visualized on the Stage 2. For Kronos you should use the LED besides the top button to indicate whether the drawbars currently controls the upper or lower manual.

Hmm, that part is going to be harder. I'll see if I can make that work. I'm just about out of non-volatile storage. Currently in user-defined mode, the DB-1 only stores CC values for one manual. The additional storage for the other manual and the code to support it might blow me past the limit. That probably sounds completely ludicrous, but the heart of the DB-1 is a tiny little PIC microcontroller and it only has so much FLASH memory. If I'd known I'd be asking so much of it I would have chosen a bigger device!

 

The MIDI Solutions boxes are powered at their MIDI IN ports. The problem is not the Kronos, it's the Voce. Many of these small controllers don't provide power on their MIDI OUTs.

I think there's two things in play here.

 

1) The MIDI solutions boxes won't work off a 3.3v MIDI bus, only 5 volt ones.

 

2) The reason the Voce won't power downsteam devices is that signal ground on the cable is not tied to chassis ground (where the +5 is referenced). I asked Dave Amels about this once, and he pointed out the schematic in the MIDI spec and he's right - they use two different ground symbols, one for chassis ground and one for MIDI signal ground. In practice, most manufacturers use the same ground reference. I modded my V5 by adding a wire from MIDI OUT signal ground to chassis ground, and it was then able to use it to power an MIDI Solutions Event Processor. I'm guessing it would be the same with the Voce Drawbar Unit.

 

 

 

[Kurt W]

 

I'll going to make the addition of Korg SYSEX support to the DB-1 a priority. Would you be able to help test the code when it's finished?

I would be happy to test it.

 

Hmm, that part is going to be harder. I'll see if I can make that work. I'm just about out of non-volatile storage. Currently in user-defined mode, the DB-1 only stores CC values for one manual. The additional storage for the other manual and the code to support it might blow me past the limit. That probably sounds completely ludicrous, but the heart of the DB-1 is a tiny little PIC microcontroller and it only has so much FLASH memory. If I'd known I'd be asking so much of it I would have chosen a bigger device!

What about making a dedicated Kronos firmware so you not would be restricted to the user-defined mode?

 

The reason the Voce won't power downsteam devices is that signal ground on the cable is not tied to chassis ground (where the +5 is referenced). I asked Dave Amels about this once, and he pointed out the schematic in the MIDI spec and he's right - they use two different ground symbols, one for chassis ground and one for MIDI signal ground. In practice, most manufacturers use the same ground reference. I modded my V5 by adding a wire from MIDI OUT signal ground to chassis ground, and it was then able to use it to power an MIDI Solutions Event Processor. I'm guessing it would be the same with the Voce Drawbar Unit.

Will give it a try if this will work also with the Midi Drawbar

[Kurt W]

OB Dave - I added a wire as you described to connect chassis ground and MIDI OUT ground. Works perfect, the Voce Midi Drawbar powers the Event Processor as you said. Many thanks!

 

Very much hope you will find a solution for using DB1 with Kronos. The DB1 needs no external PSU and would mean that a single box will give you control over the Kronos drawbars. Even after your help to avoid the power adapter for my current solution it involves three boxes.....

[OB Dave]

You've convinced me of the need. I'll figure out a way to make it work gracefully.

 

While I was at NAMM I was notified of a situation up at Mammoth Mountain which required my immediate attention. So the past few days I've been evaluating the situation and while I think the urgency might have been overstated, I believe I need one more day to complete my evaluation and I'll be headed back to San Diego to figure out an elegant solution to the Kronos drawbar problem. Does anyone know if the PC3X uses the same SysEx scheme?

[allan_evett]

OB Dave - I'm trying to get the SysEx commands for the TW Organ engine in the Roland Jupiter-50. If I secure those, do you think there might be room for those strings as well ? Or might we be talking about the larger, more comprehensive drawbar box you mentioned earlier ? I'd be willing to pay the additional $100 - $130 over the current DB-1 price. I think that $400 - $450 would be a very fair price for a comprehensive drawbar device built for Nord Instruments, with programmable MIDI CC assignments, and SysEx strings for KB3, Kronos CX3, and JP-50.

[wmp]

Does anyone know if the PC3X uses the same SysEx scheme?

 

It's controllers. TFM says ...

 

http://b3and88.com/img/pc3-kb3-midi.png

 

... and so does my PC3x.

 

 

[Kurt W]

You've convinced me of the need. I'll figure out a way to make it work gracefully.

[AnotherScott]

I had a feeling Kurz would not require sysex. When it comes to MIDI functionality, Kurz always seems to do everything you want it to do, just the way you want it to do it. That's an area where they consistently lead over Korg, Roland, and Yamaha.

[allan_evett]

I had a feeling Kurz would not require sysex. When it comes to MIDI functionality, Kurz always seems to do everything you want it to do, just the way you want it to do it. That's an area where they consistently lead over Korg, Roland, and Yamaha.

 

Going slightly OT here, but this touches on one of my pet-peeves....

 

I agree, Scott. Functionality-wise, they're at the head of the pack in so many ways. One of many examples is how auxiliary outputs are handled. The additional outputs on the PC3 series can actually be routed per patch, at the patch level. Other than Nord, that one's spotty with other manufacturer's; when present, it's usually highly conditional. OB Dave offers configuration software for the DB-1, and is willing to work on implementing further SySex capabilities; I think we need more folks like that at the 'big three'.... At the engineering level the technology is available to allow for very flexible and almost open functionality; it's been that way for a while. Dave Amels produced a box, at least 15 years ago, that demonstrates the concept well: The Voce V3.

 

It's 2013; there's no excuse for gear in which the roads don't connect. And no sacred cows need protection . Absurdities that shouldn't be include: fixed pitch bend ranges, fixed split points, limited use extra outputs, limited MIDI CC routings / SysEx capabilities, and sub-par manuals (exceptions at the moment: the JP-50's and Stage 2's documentation).

Granted, I can understand the limits - due to cost containment - on entry-level, and even some low/mid-level stuff. But a lot of the head-scratchers occur with high-end gear.