The build of a VPO - MIDI fying the bass pedals [Need Advice]

[daveter9]

I am converting an Eminent Omegan 7900 into my first (GrandOrgue) VPO ;)

Currently, I am converting my bass pedals such that they send out MIDI signals over USB.
To midi-fy the bass pedals I am trying to wire the reed switches to the contacts of a 49-key USB (MIDI) keyboard (This seemed easier to me than setting up an Arduino with MIDI output).
I removed the keys from the keyboard to just leave the required circuits and wired up a few reeds switches.

The pedal seems to produce correct MIDI signals many times but it fails to send out a full-on (value 127) signal sometimes.

The above log shows subsequent value 0 events (and thus not producing sound) even though I am repeatedly pressing the pedal.
To me, it does not feel like there is a relation between the amount of force I apply to the pedal and whether it works correctly.
I did hook up a multimeter to make sure the reed switch always actives on a press.

My Electrical Engineering knowledge is not optimal but I am thinking it may have to do with resistance.
Does anyone have experience with hooking up a bass pedal to an existing MIDI keyboard or any Idea why the setup does not work as expected?

Thanks in advance,
David

[Some extra pics for context]

[tbeck]

The "full-on" value is actually velocity. I think some MIDI implementations use a Note on message with 0 velocity rather than Note off. Could there be a keybounce issue here? Did you check how the keyboard worked before wiring it to the pedal board?

[daveter9]

You are right, they use a Note on with velocity 0 to denote an off note. According to the MIDI standard this is perfectly legal. I am not sure what keybounce is supposed to mean. I did test the keyboard before wiring it by playing pieces on it for an hour, I did not notice any struggle during that play time. I just wired a momentum switch to the keyboard breadboard, sometimes the flip of a switch just does not send out a midi signal to the computer so I am starting to question the functionality of the MIDI keyboard. I think the next step will be to test a different midi keyboard with a wired momentum switch to it.

[myorgan]

IIRC, everything MIDI is controlled by an "event." Event types can be SysEx, Note, Control, etc. For each event, there is a value. For example, a Sostenuto Pedal is either 0 (off) or 127 (on), whereas things like Notes can run the gamut from 1-127, but 0 is always off.

When I use the term "Note Off," I'm referring to a command given in early sequencing programs to counteract something like a cipher (in organ speak). Sometimes it was termed, "All Notes Off." I'm not sure if sequencing programs still have that command or not.

Michael

[tbeck]

There is an all notes off command. It is sometimes referred to as the "panic" command. It is a CC (control change) with a decimal value of 120 and velocity 0. I see that there is also a channel parameter, so I don't know if it is a universal all notes off, or just for the specified channel.

[myorgan]

Daveter,

Welcome to the Forum! I hope you continue to participate here for a time to come.

The pedal seems to produce correct MIDI signals many times but it fails to send out a full-on (value 127) signal sometimes.

The above log shows subsequent value 0 events (and thus not producing sound) even though I am repeatedly pressing the pedal.
To me, it does not feel like there is a relation between the amount of force I apply to the pedal and whether it works correctly.

I seem to remember that some MIDI equipment tends to send repeated velocity values of 0 to make sure there isn't a stuck note-on condition. That could be what is happening, and was referred to in another thread recently. The velocity for an organ will be either 0 or 127 (off or on). There is no in-between unless your organ has velocity-sensitive keyboards or after-touch.

You should test different keys, and that would be informative. Your event log shows you are definitely getting a signal from one key. If you change the key, you would see the key number change, and you could verify the organ is sending multiple 0 values, and on which key.

Hope that helps a little.

Michael

[daveter9]

Dear Michael,

Thank you for welcoming me, I really appreciate the responses ;)

It does not seem like it sends repeated velocity values of 0 since they are only send on a switch event of states. I have tested different keys and they send midi messages with different key values, that works as expected. I wired a momentum switch to the keyboard and got some strange behaviour as I described in the comment to tbeck.

[daveter9]

At this point I am not really sure what the next step in my experiments should be except for trying a different keyboard. Does anyone have any tips and/or pointers?

[daveter9]

I just came up with a different solution. I wrote a small python script which sends MIDI messages to a MIDI loopback address, GrandOrgue receives these messages and plays the corresponding note. This opens up for cheap midi-fying of keyboards by using cheap USB PC keyboard. I am gonna try to source a cheap USB PC tomorrow and wire it to the pedal.

edit: There is no notable latency when playing a key, even when written in Python. I did not try playing multiple notes at the same time yet which I will try later on.

edit 2: playing multiple notes at the same time is no problem.

edit 3: It seems for windows at least the python 'keyboard' library does not support the device ID of the keyboard which means only one USB keyboard can be used, this can be fixed using other libs which are a bit harder to install.

[jbird604]

I wonder if the problem could be that the small keyboard you are trying to use has TWO switches per note (I see two "landing pads" per note in the photo), but the pedalboard only has one switch per note. I'm not certain that would cause a problem, but I know that many of the digital pianos that I work on will not play a key unless BOTH the switches are closed. The keyboard conveys velocity information by timing the difference between the closing of the two switches.

Now you of course don't need velocity information for a VPO, since organs don't respond to key velocity as a rule. But the small keyboard may REQUIRE that both switches be closed before it can send a proper note-on signal via MIDI.

[jbird604]

Most of the keyboards I've worked on did not have a common line for the two pads of each key, but yours could be different. If you can see with your eyes that both pads switch to the same sense line, then it might work just tying them together. But it might not, as we don't know just what kind of multiplexing the system is doing with the keys. There could be timing pulses of some sort on those sense wires, and tying two of them together could cause unexpected results. But the only way to know is to try it!

Certainly it is cheaper to use a second-hand keyboard instead of a purpose-built MIDI encoder. But I do see on Roman Sowa's website that he has a little 32-note pedal encoder for 33 euros (but you have to use a diode matrix to wire it to the contacts). He has an easier encoder that doesn't require a matrix and can accept up to 63 notes and four potentiometers (expression pedals) for 60 euros.

Keep up the good work, David. I hope you find a good way to do this at reasonable cost. I'm trying to do the same thing here, building a VPO on a small budget. I do love Grand Orgue.

[daveter9]

So for the record, the two pads are wired up individually on one and end together on the other end after going through a diode individually. But indeed we do not know what kind of multiplexing is happening so my testing concludes that this USB MIDI keyboard is just not suited for this implementation.

Thanks for the pointers on the MIDI encoders, I will take a look at them!

I did see your thread for your VPO on a small budget, cool project! Seems like you are mostly handling the audio output at the moment, it seems difficult. Grand Orgue is amazing, I never used HW but I do not see a reason why I should haha. Maybe I should never try HW to prevent falling in love with it.

[jbird604]

There are plenty of nice organs available for Grand Orgue, many for free. I'm sure HW is worth the extra cost in many ways, but I am happy so far with GO.

[daveter9]

I managed to get the pedal fully working using a generic 5$ USB PC keyboard. If anyone is interested on how to midi-fy basically any input (pedal/keyboards) using a cheap PC keyboard let me know and I will write up a tutorial post in this forum.

A small video of the finished product since I am proud of it ;) I will convert my two eminent keyboards with this method soon as well.

[tbeck]

I'm curious to see how you did it.

[jbird604]

I would love to see a tutorial. I need to MIDI up some pistons and hesitate to buy a $100 encoder for it.

[daveter9]

@jbrid604 wiring up pistons is even easier since GrandOrgue supports keyboard shortcuts for pistons out of the box.
I'll see if I can write up a tutorial soon.

[daveter9]

Which section is best suited for such a tutorial? I can't decide between the virtual-organs, MIDI-device and software or Electronic Organs Repair and Restauration section.

[jbird604]

I'd say MIDI device section.

[Admin]

My intent is to use the recently activated articles area of the Forum for tutorials and informational pieces, rather than the clutter of discussion areas of the Forum where topics become increasingly buried and difficult to locate with time. Posting articles at this time requires enhanced permissions that are available by submitting a request to me via PM.