Phyto

Pipe organ console : conception

43 posts in this topic

Dear all,

I’m really a newbie (a great one, and french so please forgive me for my explanations).

I have the project to create my own pipe organ console. The problem is that the project is quite huge and my knowledge quite small. New to electronics I wish I do this myself from the beginning to the end.

So it consists in a 3x61 keyboards + 32 pedalboard + around 200 stop switches and other buttons. Maybe later a swell pedal and a preset for combinations but this will be discussed in time.

I really don’t know how to realize this for the best. Which hardware? PIC or STM 32? What would be necessary to take in count all this with efficient throughput to enable aftertouch?

Finally this console must be linked to a PC for replay system, partition editing, etc…

Thanks for any help

1 person likes this

Share this post


Link to post
Share on other sites

I'd suggest you read this topic in the Miditzer Forum:

Building A 4 Manual Console

You'll need to register but it's free. That is a very long account of someone building a console. It will give you a better idea of what you are attempting. Maybe it will encourage you. Maybe it will discourage you. It will certainly enlighten you.

Share this post


Link to post
Share on other sites

OK I registered to see this great job, but my project is quite different. I'll go into this nice forum since the subject seems to be split.

Share this post


Link to post
Share on other sites

Which hardware? PIC or STM 32?

PIC will work and there is much information about using it in projects of the type you want to do. You will need multiple Cores that can be linked together with MIDI cables. I assume STM 32 will work too but I have no experience with it.

I would suggest doing a smaller project first to get familiar with the technology. MIDIfying an old electronic organ would be a good introduction.

Share this post


Link to post
Share on other sites

PIC will work and there is much information about using it in projects of the type you want to do. You will need multiple Cores that can be linked together with MIDI cables. I assume STM 32 will work too but I have no experience with it.

I would suggest doing a smaller project first to get familiar with the technology. MIDIfying an old electronic organ would be a good introduction.

Jumping in here...I'm thinking about a digital organ console to drive something like jOrgan or Miditizer. I have two 61 note keyboards from a Schoeber organ, and stop tabs from an 80's Baldwin. Sounds like I should start with MidiIO128 to midify the keyboards, one keyswitch per input. That leaves me 6 leftover. Anything reasonable I can do with 6 inputs and my stop tabs as a first step?

I also have a 13 note pedalboard with integrated switches from the Baldwin. I know that is a compromise, but I'd like to keep the physical size to a managable (portable?) level.

--tom

Share this post


Link to post
Share on other sites

Jumping in here...I'm thinking about a digital organ console to drive something like jOrgan or Miditizer. I have two 61 note keyboards from a Schoeber organ, and stop tabs from an 80's Baldwin. Sounds like I should start with MidiIO128 to midify the keyboards, one keyswitch per input. That leaves me 6 leftover. Anything reasonable I can do with 6 inputs and my stop tabs as a first step?

You'll probably want to add a second Core pretty quickly to handle the pedals. I'd wait till then to tackle the stops.

You probably don't have pistons on your keyboards. Pistons are more useful than stops for serious playing. I'd add pistons before I did the stops. If you are going to use the Miditzer, adding 11 pistons per manual works well. The 11th pistons are located well to the right and are used for Cancel and perhaps Memory Level. You could wire up 3 pistons on each manual until you get the second Core.

Once you have pistons, then you'll probably want indicating stops. You might do that by adding an LED over each stop tab and mechanically reworking the tabs to be momentary on. That will allow you to set a registration with a piston, see the settings in the LEDs, and then add and subtract stops by momentarily touching a stop tab.

Share this post


Link to post
Share on other sites

You'll probably want to add a second Core pretty quickly to handle the pedals. I'd wait till then to tackle the stops.

You probably don't have pistons on your keyboards. Pistons are more useful than stops for serious playing. I'd add pistons before I did the stops. If you are going to use the Miditzer, adding 11 pistons per manual works well. The 11th pistons are located well to the right and are used for Cancel and perhaps Memory Level. You could wire up 3 pistons on each manual until you get the second Core.

Once you have pistons, then you'll probably want indicating stops. You might do that by adding an LED over each stop tab and mechanically reworking the tabs to be momentary on. That will allow you to set a registration with a piston, see the settings in the LEDs, and then add and subtract stops by momentarily touching a stop tab.

Jim,

So, pistons are used to save combinations of stop key settings. right? (If there is a primer of organ terminology, please point me at it...) I have another "parts" organ (Gulbransen Premiere) with 6 buttons between the manuals, but everything is electromechanical. I don't believe that these would be reusable.

So, I program the pistons in the Miditizer software, then use the piston buttons to activate the specific combinations?

For what its worth, I have the 2 keyboards for the Gulbransen, but the "combination action" mechanics attached to the back make them quite large. The Schoeber keyboards are relatively compact.

Thanks,

Tom

Thanks,

Tom

Share this post


Link to post
Share on other sites

Is there a preferential hardware to enhance and overcome the 31250 bauds limit ? Do you ever had some problem of saturation or do you know people who had problem with ?

Is there a good solution ?

Maybe this could be a problem for some litterature.

Edited by Phyto

Share this post


Link to post
Share on other sites

Is there a preferential hardware to enhance and overcome the 31250 bauds limit ? Do you ever had some problem of saturation or do you know people who had problem with ?

I know that in todays modern world of Gigabit (or 10Gb) ethernet, 31250 baud sounds incredibly slow but each note is only 3 bytes so that means you can play over 1000 notes per second, which is more than you would ever need even for the most complex piece of music :)

Even if you are sending lots of CC messages, you would not really saturate it.

Of course MIDI over USB (as used by the CORE32) does not have this limitation......

Cheers

Phil

Share this post


Link to post
Share on other sites

That's why I asked the question. Not only for the only notes, but I thought the refresh of all the configuration every x ms does take part to the flux.

For a pipe organ, one can really imagine 30 stops playing together with 10 notes per second, refresh at least every 20 ms, that is to say a total of 3 x 30 x 10 x 50 = 45 000 bytes per second.

+ extra like couplings or transposition...it's easy to think about saturation.

That's of course for the active part of the instrument but I suppose the inactive one is under continuous survey and refresh too !

Share this post


Link to post
Share on other sites

For a pipe organ, one can really imagine 30 stops playing together with 10 notes per second, refresh at least every 20 ms, that is to say a total of 3 x 30 x 10 x 50 = 45 000 bytes per second.

+ extra like couplings or transposition...it's easy to think about saturation.

That's of course for the active part of the instrument but I suppose the inactive one is under continuous survey and refresh too !

The MIDI only has to transmit what changes. The MidiBox scans the whole organ at 1,000 times per second, sees the change, and then sends the MIDI Message that describes that change. If you are using the console to control a virtual organ then it is the virtual organ that does all the multiplication of the notes for the stops and couplers that are on. Even then it is only the changes that generate data so for your example that is 10 notes x 30 stops x 3 bytes = 900 bytes/sec and it normally does not have to be sent via MIDI although it could be.

The potential problem is the swell pedal. You have to design things so that the pedal doesn't "jitter" and send two alternating values as fast as the hardware can manage. When there is a problem of MIDI data saturation it is often traced to the swell pedal.

Share this post


Link to post
Share on other sites

So, pistons are used to save combinations of stop key settings. right? (If there is a primer of organ terminology, please point me at it...) I have another "parts" organ (Gulbransen Premiere) with 6 buttons between the manuals, but everything is electromechanical. I don't believe that these would be reusable.

So, I program the pistons in the Miditizer software, then use the piston buttons to activate the specific combinations?

Yes, your understanding of the pistons is correct. I'd suggest downloading the Miditzer and "playing" with it using the screen and mouse to get familiar with things. It is a true virtual organ in that everything is functional without the need for any real hardware. Of course, you need a real keyboard to really play live music but you can explore the function without it.

This might serve as a primer with respect to theatre organs: Learning about the Theatre Pipe Organ (TPO) and How the MIDITZER simulates a TPO

Section IV of the "Hauptwerk Installation and User Guide" has a primer for classical organs. I don't know if that is available for download separately from the download of the program, which is huge.

Share this post


Link to post
Share on other sites

Many thanks Tom and Jim for your explanations !

Last, what are the main hardware difficulties if I think about replay, partition editing, wifi console ?

For the swell pedal I can take it in consideration even if it'll be one of the last thing to construct. But you're right to make me think about before shuting up the pipes.

Share this post


Link to post
Share on other sites

Last, what are the main hardware difficulties if I think about replay, partition editing, wifi console ?

By replay do you mean recording a live performance and playing it back? If so, for a virtual organ that would be done on the computer side. The Miditzer provides a MIDI recorder/player built-in. I think the other virtual organs are now catching up on that feature. If not, you can use general purpose MIDI sequencers for that purpose but it is less convenient.

I don't understand what you mean by "partition editing."

By wifi console I assume you mean a wireless MIDI connection rather than MIDI cables connecting the console to the computer. I don't know of a Midibox project that provides a wireless connection. I know such devices are available and you could use one to replace the MIDI cables to the Midibox. Here's one example of a wireless MIDI device: Wireless MIDIjet Pro from Classic Organ Works

Share this post


Link to post
Share on other sites

Yes you understood what I meant by replay. However the organ will be real and not virtual so the sequencer is the solution. What do you mean by less convenient : more expensive ?

Sorry for the "partition editing" but I don't know the exact english term for this. A sort of score writer = music notes writer, a program that converts what you play into a score...if you understand this !

For the rest, I wish to avoid cables in the room since the instrument will be done of 3 parts :

- the instrumental one, containing pipes and electrical traction system (solenoids)

- the console containing the motherboard

- computer part for working on a desk.

At least wifi between console and organ would be great !

Share this post


Link to post
Share on other sites

For the rest, I wish to avoid cables in the room since the instrument will be done of 3 parts :

- the instrumental one, containing pipes and electrical traction system (solenoids)

- the console containing the motherboard

- computer part for working on a desk.

At least wifi between console and organ would be great !

Wireless is certainly possible but do you really need it? Very few organs have wireless consoles, virtually all of them have at least a power cable and signal cable of some sort coming from them. With a serial interface (MIDI or faster) you need nothing more than a CAT-5 cable between the console and where ever the pipes are being controlled from.

Share this post


Link to post
Share on other sites

Phyto, I didn't realize that your project used real pipes. That is likely to make my answers wrong. Tell us something about the pipes part of your project. What will be the flow of control from the console to the pipes? My experience with pipes is almost all Wurlitzer theatre organs that still use their original electro-pneumatic relays. My thinking will tend to be colored by that background. If you have something quite different, say a classical organ with direct electric valves under the pipes, I'll need to adjust my thoughts and comments accordingly.

Share this post


Link to post
Share on other sites

OK, I will tell you some more. Even if I don't know the precise technical terms, I hope you will understand.

The pipe part is very classical and can be comparate to an old mechanical instrument. So you are entirely right when you think about "classical organ with direct electric valves under the pipes". The solenoid with pallets are under electric command, but the transmission is not so direct from keyboards to the valves to avoid use of high voltage and the subsequent problems you can meet in such all electric instruments.

I would say only the transmission is electronic, using MIDI from console to the basement of the organ case where "power relays" are set. From these starts the effective electric transmission. Each pipe has an independant magnet inside him (in the chest) and can be controlled individually. But this is the very simple part, only wire.

To say little more about the project, it is not so huge since I plan 400 pipes, constituting about 4 pipe types of 100 notes each, that is to say something like a stopped diapason, a flute, a principal and a reed (if my terms are correct). Of course there will be multiple "extensions" and "transpositions" since each stop is playable in 8' 4' 2' 1' with all mutations (theorically, if it makes sense). Musically speaking, the real difficult part concerns harmonization that must be very clever to provide good effet.

Share this post


Link to post
Share on other sites

Wireless is certainly possible but do you really need it? Very few organs have wireless consoles, virtually all of them have at least a power cable and signal cable of some sort coming from them. With a serial interface (MIDI or faster) you need nothing more than a CAT-5 cable between the console and where ever the pipes are being controlled from.

Wireless, maybe not indispensable but so convenient...

What do you mean by MIDI or faster, MIDI over ethernet ?

Share this post


Link to post
Share on other sites

Wireless, maybe not indispensable but so convenient...

What do you mean by MIDI or faster, MIDI over ethernet ?

MIDI over Ethernet, MIDI over USB, or, depending on whether you need to interface to a lot of 'standard' gear, just speed up the MIDI data rate itself. Classic Organ Works / Midi Works of Canada does sell the MIDIjet wireless MIDI interface. I've experimented with wireless MIDI over the Xbee wireless modules.

MIDI is plenty fast enough for a pipe organ *if* the coupling/unification happens at the chamber end and not at the console end. A friend has built and I use MIDI pipe driver boards with 'built-in' unification. The information is a little dated (we've evolved a little past this point) but the concepts are still valid: YAPPI Board It's a PIC based board that can play a rank of pipes at any of the 'standard' pitches.

Share this post


Link to post
Share on other sites

How interesting...

Just a thing, I don't understand what you mean by "*if* the coupling/unification happens at the chamber end and not at the console end".

If you can reformulate it and what it supposes in terms of network, many thanks.

Share this post


Link to post
Share on other sites

How interesting...

Just a thing, I don't understand what you mean by "*if* the coupling/unification happens at the chamber end and not at the console end".

If you can reformulate it and what it supposes in terms of network, many thanks.

Unification requires playing a single rank of pipes at multiple pitches. Say I have a 97 note rank of stopped flutes that I want to play at 16, 8, 4, and 2 foot pitches.

Lets say the unification is handled in the console. If I turn on the four stops and start playing single notes, the processor in the console determines that four MIDI messages must be sent to the pipe chest. If I play a three note chord, 12 MIDI messages must be sent to the chest. And when I let up another 12 MIDI messages must be sent. For each key I press and release 4 Note On and 4 Note Off messages must be sent to the chest.

In the YAPPI boards that I mentioned previously the unification happens at the chest. When the 16, 8, 4, and 2 stops are turned on a MIDI message is sent to the chest. Then when notes are played, a single Note On and a single Note Off is sent to the chest for each note. The board at the chest figures out which pipes need to be played. Far fewer MIDI events need to be sent over the wire.

The location of the 'unification' processing has a large bearing on the amount of MIDI traffic required.

Couplers (especially between manuals) have the same issues.

Share this post


Link to post
Share on other sites

Thank you so much John, I now see how YAPPI is a great job even if I don't fully realize all possibilities : you need one YAPPI card for each rank or unit rank of pipe ?

Ideally, I need 100 outpouts for each 4 ranks.

For some practical reasons I'd like to have all keyboards and pedalboard with the same stops, and such a system is probably the best to create complex mutations ! So if all keyboards are identical (except coupling) maybe the solution is very easy and very economic in material ?

For example, a list of 40 stops that are the same for each keyboard and pedalboard. The only difference between them would be the coupling if it is use. Is this imaginable? If yes, it is a very flexible instrument I dream, with maximum possibilities.

Share this post


Link to post
Share on other sites

I now see how YAPPI is a great job even if I don't fully realize all possibilities : you need one YAPPI card for each rank or unit rank of pipe ?

Yes, we generally use one YAPPI board per rank of pipes. There is a mode where it can drive a straight (multi-rank) chest but flexbility is lost.

In a home organ the number of ranks is usually fairly small and the cost of the YAPPI board per rank is relatively low in the grand scheme of things.

Share this post


Link to post
Share on other sites
Ideally, I need 100 outputs for each 4 ranks.

For some practical reasons I'd like to have all keyboards and pedalboard with the same stops, and such a system is probably the best to create complex mutations ! So if all keyboards are identical (except coupling) maybe the solution is very easy and very economic in material ?

Your organ layout is going to be much more like a theatre organ than a classical organ even though it will have a classical sound because of the pipes you use. I mention that because as you study how your organ might be arranged, you should know that the information about theatre organs will be more useful than that about classical organs.

I realize that you are probably saying 100 outputs because it is a round number but you might as well get used to using the real numbers that will be involved. A basic rank of pipes is 61 pipes because there are 61 notes on an organ keyboard, 5 octaves + 1, and this is normally an 8' pitch. A rank that is extended to fully cover two pitches 16' + 8' or 8' plus 4' has 12 more pipes for a total of 73. For 3 pitches, generally 16 + 8 + 4 or 8+ 4 + 2, you need 85 pipes. For four pitches, almost always 16' + 8' + 4' + 2', you'll have 97 pipes.

Now way off topic with regard to Midibox, do you have your pipes or are they yet to be acquired? Assuming you need to acquire your pipes, I am going to assume you'll be using used pipes because your costs will be astronomical if you use new pipes. I think you may have some problems finding ranks of classically voiced pipes that are extended to 97 pipes and maybe even 85 or 73. Even in theatre organ ranks, not many ranks are available with 97 pipes in a rank. The pipes at the top end of a 97 pipe rank are tiny and they all tend to sound the same, at least to humans. The bottom 12 pipes of a 16' rank are huge, heavy, hard to locate, and expensive. You'll need at least 1 octave of 16' pipes to provide a bottom to the pedals, preferably a stopped rank (half length) so the low C will be under 10' when mounted on a chest at floor level. You can miter the pipe or mount it horizontally if you don't have 10 to 11 feet of height. By now I think you will have figured out that I don't think 4 ranks of 97 pipes is needed or even very practical. You can get 16' pitches on the manuals by making 16' tenor C stops from your 8' ranks; these play at 16' pitch but the lowest octave is not present. Pitches higher than 2' are always missing notes at the top end of the keyboard. You can do the same to get 2', 2 2/3', and 3 1/5' pitches from a rank that only has a 4' top end.

I'd suggest you use jOrgan to mock up your organ virtually and experiment with ranks and stops to figure out how your ideas are going to work out when it comes to actual performance. Things that look good on paper often turn out quite differently in reality.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now