Page 3 of 3

Re: *** NOOB Warning *** Attempting CODM - First Post

PostPosted: Sat Nov 19, 2011 11:16 am
by steve till
Another suggestion concerning expression..

Why not put the speakers for the virtual Swell stops inside the Swell chamber?
And of course the speakers for the virtual Choir stops inside the Choir chamber?

That solves two problems, first, all Swell stops come from the same location,
and second, the amount of expression is exactly the same for both the
virtual and the real pipe Swell stops. Same for Choir.

I tried that years ago with some old analog technology married to pipes,
of course we had lots less adjustability of the way the expression worked.
Still, the results were far better with the speakers inside the pipe chambers.

Re: *** NOOB Warning *** Attempting CODM - First Post

PostPosted: Sat Nov 19, 2011 11:34 am
by RWAbacus
Neither the Swell chamber nor the choir chamber has any extra room for speakers - regretfully. Both are packed with pipes.

I still have to operate those two sets of louvres under Hauptwerk's control. Therein lies the problem.

Rereading Mark's comment on how he controlled physical louvres, however, got me to thinking that I could do the same thing as he did adding the 8-contact roller back in the equation along with the 10k potentiometer - with another piece of Rube Goldberg's hardware linkage for so doing.

Each of the eight contacts could be recognized - by Hauptwerk - as a switch on input - much like a thumb piston (although not momentary contact). Then Hauptwerk could output to one of Midiboutique's output pipe drivers - although not a pipe, but a portion (1/8th) of the air motor. I think this would work. As the action is reversed, the 'pipes' would be reversed as well.

Am I correct in this thinking?

Re: *** NOOB Warning *** Attempting CODM - First Post

PostPosted: Sat Nov 19, 2011 1:29 pm
by mdyde
Hello RWAbacus,

I'm not sure I fully understand your idea, but in general Hauptwerk doesn't have a means in the CODM (or that you could apply to off-the-shelf sample sets) that would allow you to add arbitrary switch inputs or outputs that could operate or be operated by virtual swell pedals (continuous controls), i.e. that would translate between a set of MIDI switches and virtual continuous control positions.

I really think you'd do better to use appropriate circuitry so that your swell shades use conventional MIDI control change messages. It's usually best to try to make MIDI implementations as conventional as possible, for simplicity and maximum compatibility.

Real Pipe Organ Swell Shade Action

PostPosted: Sun Nov 20, 2011 2:09 pm
by RWAbacus
OK ... the solution seems to be at hand ...

In an e-mail to Jordan at, he responded that he can reprogram the firmware on the existing mdec32dd-n's board that's used to control the stop action of all the pipes so that the top 8 notes of that driver board (only) can act as a "bargraph" behaviour, whereby CC#7 (Volume) is converted to NoteON/NoteOff messages to keep the "bargraph" updated accordingly.

This behaviour is applied to the last 8 outputs regardless of the unit's DIP switch settings.

The reprogrammed firmware then replaces the existing firmware (it's socketed).

A most perfect solution in so many ways and so much better than that which I had thought until his email this morning - and at a most reasonable price of 50€ + shipping.


Re: *** NOOB Warning *** Attempting CODM - First Post

PostPosted: Sun Nov 20, 2011 2:25 pm
by mdyde
Hello RWAbacus,

Splendid - that sounds ideal.

Re: *** NOOB Warning *** Attempting CODM - First Post

PostPosted: Mon Nov 21, 2011 1:48 pm
by megafisc2
I agree. In my case, the roller/contacts existed so it was a simple solution. Here is the roller that was used:
I simply encoded the roller contacts as a note-on message and decoded them on the same mdec driver that I used for my stops. This then drives the eight stages of a Peterson RC150 shade motor. Certainly, had the roller not existed, I would have gone the way Jordan has suggested.

Please keep us posted on your progress.

Re: *** NOOB Warning *** Attempting CODM - First Post

PostPosted: Mon Nov 21, 2011 6:04 pm
by RWAbacus
I had the same rollers - albeit much older - and just took two of them out a few days ago (and cut all the wiring out too). The ones I removed included the diodes across each leg of the circuit to stop the transient voltage spikes. Without these, pipes would speak - and sometimes MANY pipes would speak - without any provocation whatsoever just by moving the swell and/or choir expression pedals! Before installing those diodes, I observed the blue electrical arcs on the contact roller and that's when the extraneous sounds occurred. This wiring isn't even connected to the pipes except circuitously through the power supply. Ghostly. Diodes stopped all those problems and those arcs.

I almost went with what you suggested; however, I'm not that far along in the CODM yet to make two more 8-contact (faux) keyboards in addition to the 4-manuals + pedalboard. You can't use your solution, I believe, without altering the organ definition, and you can't (reliably) alter someone else's organ definition since they probably didn't even use the CODM. Eventually I hope to go the full CODM route. For now, I'm using existing sets and disabling some/many audio samples and substituting real pipes for those ranks. And, that is easy to do in Hauptwerk without CODM.

Since the new console will be moveable, I didn't want to keep the existing wiring either (twenty more wires). Right now, there's but a firewire cable to all the pipe ranks and that's it. Regretfully, firewire is only rated to just under 5 metres maximum length - so a short tether at the moment.

Interestingly, we have the same Peterson RC150 shade motors for the swell and choir louvres.

Midiboutique's solution using a reprogrammed controller was just right for me; neat and compact. It's ordered.

I also ripped out a long contact roller with 32 contacts for the old crescendo pedal. A fun time was had by all.