Organ Swell Shades Control

MIDIfication
1

On most pipe organs, the volume of the pipes is controlled with swell shades.  These shades are similiar to window louvers or dampers and are controlled by a servo motor which opens and closes the shades in response to the setting of a potentiometer attached to a foot pedal (similar to the accelerator on a auto).

the MB Hardware Platform list contains an MF module which I assume is for “motor fader”.  Could this module be used to control a servo motor on swell shades? The pot is usually 10k.

Johnc

2

Hi Johnc,

is it a common DC motor with two inputs? Then it could work. You can doublecheck this by supplying the motor with 5..10V DC, it’s especially important that the motor changes the direction when the polarity is swapped.

Best Regards, Thorsten.

3

Thorston,

The motor required to move the shades will probably have to be a high torque unit.  Will that be a problem to drive it from the FM module?

Johnc

4

I think you should meassure the current of your servo motor at the startup moment.

The TC4427 (Dual High Speed Powerdriver) has a peakcurrent from about 1,5 A. If you meassure the same or less current you can take the MF Modul. If its more then 1,5A you need an other powerdriver with more power. (This is only abput the hardwareside. I don´t know if there is a big difference from the programming side for an servo motor.)

Greetz

Markus

5

I am back on the track of using the midibox MF module to drive a dc motor.  After researching the issues, Hbridges are available to provide for the increased output needed for a larger dc motor.  What I am trying to do is to do position control using a dc gear motor as typically done in the radio control area for model aiplanes and boats.  Basically, it involves a pulse generator, servo amplifier and H bridge. My big question is wether the MF module can provide the function of the servo amplifier, accepting the pulse input and driving the Hbridge and motor.  The motor will be an auto windshield wiper motor, 12vdc, reversible,with 60 RPM output, with a potentiometer connected to the output shaft for feedback. Does the MF module operate on midio128, 64 or what app software?

Anyone knowledgable in this area, please respond.

Thanks,

Johnc

6

The MF module operates on the midibox64E- the midibox lC / hn applications.

I think the midibox64e application is the best for your needs of these here listet.

You can controll up to 8 pots per core.

Greetings

7

Midiboxxer,

Thanks for the reply.

So I would need a core running the 64E app, and an MF module.

Since I will have only on rotary encoder in the form of a pot, and only one motor to run, can I do it with only one DIN?

The dc motor is a reversible, 12 volt motor,  only two leads.  I will add a pot on the output shaft to encode position. How does the feedback pot tie in to the MF module, or the core?

Since the MF module cannot provide enough drive for the motor,  I will use an Hbridge which has two inputs, one for each direction.  How does this tie in to the MF module?

Thanks,

Johnc

8

Since I will have only on rotary encoder in the form of a pot, and only one motor to run, can I do it with only one DIN?

Yes you can.

The dc motor is a reversible, 12 volt motor,  only two leads.  I will add a pot on the output shaft to encode position. How does the feedback pot tie in to the MF module, or the core?

This should answer your question.

http://www.ucapps.de/mbhp/mbhp_mf_interconnections.pdf

Since the MF module cannot provide enough drive for the motor,  I will use an Hbridge which has two inputs, one for each direction.  How does this tie in to the MF module?

That depends on what chip you are useing. Perhaps you could only change the voltage regulater and the power mosfet drivers with some with more power. This may be a solution for you:

http://www.robotroom.com/HBridge.html

Hope this will help you.

Greetings

Markus

9

Midiboxxer,

It appears from the datasheets for the TC4427 and the schematic for the MF board that the HC595 delivers two outputs to each TC 4424 on the MF board.  the TC 4427 is a dual mosfet driver rated 1.5 amps (the TC 4424 will do 3 amps) with no special circuitry for direction, etc.  consequently, it appears that any two input H-Bridge with the necessary rating to match the motor, will work.

What I don’t see in the references is where feedback from the pot comes into the picture.  There is input directly to the core which I assume is where the feedback comes in, however, I am at a loss as to what the signal is; variable dc voltage analogus to position of a pot as a % of 5vdc, a variable resistance value from the pot directly. If a pot is involved, what value is it, and is it linear.

Last question,  will the 64E apps software run the MF module, right out of the box, or is there an ini file. or some way to configure it?

I am not familiar with the ALPS devices or motorfaders in general, other then that they are motorized pots used in audio, etc.

I am getting closer to the realization that the MF module and core will do what I want them to do, but some extra input would help. 

Thanks for your help,

Johnc

10

The pot should be a 10K linear one. It have to be connected to the J5 port of the core.

Last question,  will the 64E apps software run the MF module, right out of the box, or is there an ini file. or some way to configure it?

You have to change the main.asm here:

; Although MIDIbox64E has been designed for rotary encoders, it can also handle with

; up to 64 pots/faders or up to 8 motorfaders.

; Pots and faders are mapped to the “encoder” entries 64-128.

; Example: if group width is 16, and group 1 is selected, encoders are using

; entry 1-16, and pots are using entry 64-70

; NOTE: morphing is automatically disabled if analog pots/faders are connected

#define DEFAULT_NUMBER_AIN 0

;

; you could enable the multiplexers here to test this application

; with a MIDIbox64 based hardware

; if 0: no multiplexers (a *must* when MF module connected)

; if 1: use multiplexers

#define DEFAULT_ENABLE_AIN_MUX 0

;

; motorfaders connected?

; if 0: MF module disabled

; if 1: MF module enabled

#define DEFAULT_ENABLE_MOTORDRIVER 0

;

; The well known motorfader calibration values – can also be changed “online” in the calibration menu

#define DEFAULT_MF_PWM_DUTY_UP   0x01 ; PWM duty cycle for upward moves  (see http://www.ucapps.de/mbhp_mf.html)

#define DEFAULT_MF_PWM_DUTY_DOWN  0x01 ; PWM duty cycle for downward moves (see http://www.ucapps.de/mbhp_mf.html)

#define DEFAULT_MF_PWM_PERIOD    0x03 ; PWM period                        (see http://www.ucapps.de/mbhp_mf.html)

and perhaps the “encoder entry” for the pot. But in your case I don´t think so.

Greetings

11

Midiboxxer,

Having gone through all recommended reference links, etc. I’m still a bit confused.

Permit me to review what my control device inputs and control device outputs are:

Control Box - The installation (core, MF card) is intended to operate as a stand alone device, installed in a small box, with terminal strip for inputs and outpus.  Power supply will be internal to the box.

Control devices:

Control Pot - The control pot is built in to the foot actuated swell shoe on the organ console remote from the location of the control box (core, MF card).  This pot sets the desired position for the swell shades (0-closed to 100% fully open).  No leds are needed to indicate position.

Swell shade motor Drive -  This includes the dc gearhead motor with arm attached to the output shaft which opens and closes the swell shades.  Movement is from 0 - closed shades to 100%- fully open shades, or rotationof the output shaft of 90 degrees.  Attached also to the output shaft is a rotary shaft feedback pot, linear taper, 10K which tracks the position of the output shaft.  So from the motor drive there are  7 wires,  3 for the 12vdc to the motor(2hots, ground); 4 for the connection to the feedback pot (top of pot, bot. of pot, wiper, ground).

Control Inputs:

Position setting pot -  Analog in on J5, term A0 of the core

Feedback pot  - Analog in on J5, Term A1 of the core

Outputs:

output to motor - digital out on J3, M1-1 and M1-2 on the MF card.

The core will have midi out and midi in connectors, however, operation of the installation will be by turning on the power to the core, etc.  midi channel selection is for setup, config, troubleshooting only.

Since the installation has minimum devices,  a custom pc board will be used to combine the core and the MF module components. With only one motor to drive, all should fit on one board.

What I don’t understand is the input side from the pots.  Do the leads from the pots connect directly to the terminals on the core? Do I use the 5vdc on the core PS to input a proportional dc voltage?  Same question for the feedback pot signal.

Which ini file for  64E software do I use,  the  MF ini?  How does the  software know which of the pot signal inputs is control input, and which is feedback input?  I haven’t found anything in the writeups, tutorial, etc. that say anything about feedback. Do the ALPs motor faders have a feedback pot tied in to the slider that is internally wired.  With only 4 wires to each motor fader from the core/MFboard, something is not right. 

Regards,

Johnc

12

Hi,

for a better understanding:

Normally the pot of the motorfader is used as feedback and controllpot. In best case it has an touchsensor too. So when the application runs in automatic mode, the fader moves when the core receives data from e.g. an sequenzer. Now, when you touch the fader, the touchsensor detected it and the software (midibox Lc /64e) stops the automatic mode, so you can do your changes.

For your needs it should be possible to connect the pots like you discribe and give them the same massages. You should ask someone here who already use this app with motorfaders to try this for you.

Greetings

13

Midiboxxer,

My understanding of how the motorfader works, and how it is used is weak. I can see how the fader itself can serve as feedback to the control loop, but it didn’t dawn on me that a voltage, current, or what have you, resulting from the the same resistance value could also be used for some other purpose.  Once the  parameter is digitized, it could serve both purposes.  The PIC software could easily handle that.  Looking at the interconnection diagram, on the ALPS fader, there are two leads for the motor and  three connections at the other end, one for the touch sensor, ground and one other.

is it safe to say that the one odd connection comes from the fader resistance itself?  Can anyone familiar with the ALPS motorfader confirm this for me? If thats the case, then all my questions will have been put to rest.

Hopefully, Thorston or comeone else will jump in here with some input.

Thanks,

Johnc

14

Midiboxxer,

Located data online for the ALPS motorized slide faders.  Good stuff.  the diagrams show terminals:

A and B on what appears to be the motor.

Term. 1 on the motor end of the device.

Term.  T, 2 and 3 on opposite end from motor. 

I suspect that A and B are for the motor.

T is the tap on the slide resistor,

Question:  Is term. 2 the bottom of the resistor, or is 3 the bottom?

Thanks,

Johnc

15

Henrygr, Audiocommander,anybody,

The thread on this subject seemed to die out, so i’ll start it up again.

I started a thread in the “midification” area titled " Organ swell control" which is a servo control project I am trying to put together using a core and mf module.  You can read the thread to get details on what I am doing.  I started out looking at the hardware and devices used with  R/C and robot control, but found that the older solutions using IC chips such as the NE544 are now out, as the chip is no longer available. Being a midibox user, I turned to a midibox solutiion which can accomplish with software, what the NE544 did with hardware.

Basically,  it involves controlling a 12vdc, gearhead, auto windshield wiper motor with an HBridge high current (MOSFET)driver using the MF module and  core; position selection with a rotary pot, and feedback from the output shaft of the motor using a rotary pot.  Most dc motors used in R/C and robotics do not have enough output torque, so thats why the wiper motor.  Not only that, the wiper motors are cheap. Midiboxxer has been giving me help with application of the midio 64E apps software, but I invite anybody interested to chime in, please.  The 64E ini file and tutorial are very much geared to other applications and provide bells and whistles that don’t apply to servo control.  Since I am not a programmer, or am I proficient in “C” or asm,  being able to use the 64E software out of the box, would be my choice. Problem is I can’t figure out where the software looks for the position selection input  and where it looks for the feedback control input, both of which should be dc voltages.

Regards,

Johnc

Stryd_one, All

My current midification project involves servo control also.  Not R/C  but same principles. 

PLease check the thread  “organ Swell Shade”. in the midification area.  I have some basic questions and reservations concerning the midibox “MF” module in servo contol.  My particular interest involves position feedback from the motor unit to the core, and if the existing 64E software provides from position feedback.  Also,  I would ask that anyone knowledgable in that area to make any comments of clarification, input, instruction, tips, etc. Thorston, if you come across this post, your input would be appreciated.

Thanks,

Johnc

I guess the main problem is that you have to make the PWM between 1 and 2ms .

It’s pretty hard (but not impossible, to my knowledge) to work under 1ms with MIOS.

So, maybe there could be a easier solution using core analog outputs, and a Ne555, like in this example :

http://www.epanorama.net/circuits/servo10v.html

what do you think ?

servo10v.gif
servo10v.gif550×343 5.11 KB

16

Bill,

Check out the following website, there are several good circuits to generate servo pulses;

http://www.uoguelph.ca~antoon/gadgets/servo2.htm: …servo3.htm:  …servo4.htm

All of these generate pulse trains in the 1 to 2 ms range for compatibility with the typical R/C servos.

You should be able to change the frequencies and pulse widths by changing component values.

also google  NE555 and you will find numerious sites with data concerning 555 nd 556 circuits.

Johnc