Voltage concerns

[OrganGrinder]

Hi

This is my first post, so I hope it is in the right place.

I am a newbie with electronics, done some minor stuff in highschool >20yrs ago but generally don't know much.

I am trying to midify an old organ of mine while preserving the orginal organ.

The organ currently uses 15V, 17V and 20V for keys, pedals and stops respectively.  I also understand the DIN and DOUT modules from SmashTV work on a maximun of around 5V.

So my questions are:

1) for DIN, how do I limit the voltage from the organ so I don't fry the core/DIN.

2) for DOUT, how do I increase the output voltages to the requirements of the organ circuitry.

Thanks in advance

OrganGrinder

[jimhenry]

On the DIN side a simple resistive voltage divider can lower the voltage BUT it might upset the analog circuits in the organ.

DOUT is more complicated because you need to know what is needed in terms of both voltage and current.  Further, there might be issues of noise being introduced into the analog audio circuits.

Before you get too far into this you need to ask yourself why you want to preserve the original organ sound and why you want to use DOUT.

These days you can get an organ sound that is probably far superior to what an analog electronic organ produces.  Are you holding on to the analog circuits for any reason beyond a sense that you shouldn't discard that much circuitry?  I've been wrestling with the issue for years with my Rodgers Trio organ, a high end electronic of its time.  Other projects have kept me from doing anything with the Trio but I have decided that when I get back to it I'm stripping out the analog electronics as not worth the extra complications to making the console into a MIDI controller.

Once you discard the analog electronics, DOUT would be pointless with nothing to control.  But if you retain the analog electronics, why do you want to control them with MIDI?  The only reason I can think of is so you can use a sequencer to drive the analog tone generation.  Do you really have analog tone generation that is so good and so unique that it is worth the effort to drive it with MIDI?

[OrganGrinder]

I don't want to go off topic into areas of why I am preserving the original workings of the organ or if using samples is better or anything else like that.

But I will discuss my motives for this and although I would appreciate constructive comments, I am not interested in a debate.

Anyway...

The organ that I am midifying was my late great uncles and is currently in near perfect condition for a pre 1980 organ.  Only part of one octive on the lower manual dosen't work right and it is not the contacts but elsewhere, and I am getting that fixed soon.

Anyway I was planning on donating this organ to the chuch which I attend, being that I am the only organist and the current organ is woefully inadiquate.

What I thought is that if I installed midi in and out to the organ, it would provide a level of future proofing the system.  It would provide several advantages:

  1. If through age the internals of the organ break down, it is still usable with an    external module.

  2. If I move elsewhere, the organ can still be used with a sequencer of some kind  (probably a laptop).

  3. The organ sound can be supplimented with sound modules or the roland piano which is already at the church.

I have already considered a radical midification (replacement instead of supplimenting), but have rejected the proposal for several reasons:

  1. Replacing the innards with not only sound modules but also the other equipment such as amplifiers and speakers would make the endeavour very expensive and beyond my means.

  2. I can't see that it is likely that I could duplicate the stops, traps, toys and effects that already exist on the organ.  If I change the functions of any tabs or buttons, I would also have to change the labels or panels on the organ to maintain the turnkey approach of the organ console, and being that I am inept at any kind of woodwork, I don't think that is a good idea.

Now I do have a Wurlitzer 950 which seems to be on its way out (many things aren't working right or at all and it is only getting worse) and that would be a good condidate to radically midify, but a working system which only needs a small fix seems to me like throwing the bady out with the bath water.

I hope this my motives clear

OrganGrinder

[jimhenry]

My suggestion would be to radically midify the Wurlitzer 950 first as that appears like it will be an easier project.  Once you have that under your belt you will be in a much better position with regard to the other organ.

If you want to proceed now with the other organ, I would still suggest only providing MIDI Out.  I don't dispute your reasons for wanting to add MIDI In but your self assessed electronic skills don't seem up to the challenge.  You need to understand the analog electronics to be able to add MIDI In.  The safest way to add the MIDI Out would be to add a set of contacts for every switch and key that is to be MIDIfied to keep the MIDIbox electronics completely separate from the original organ electronics.  It might not be necessary but you'll need to understand the analog electronics to know if you can borrow the signals for the purposes of MIDIfying the console.

Building a MIDIbox as a standalone project is not too hard.  When you start modifying the circuits of an analog organ you are now reengineering the organ.  That is a big step up from merely building a MIDIbox.  I hate to rain on your parade but I don't want you in over your head on an organ you are trying to preserve.

[stryd_one]

This is my first post, so I hope it is in the right place.

Hi 'Grinder :) Welcome aboard!

I just thought that seeing as you're new I should mention that taking Jim's advice could be very beneficial to you... he's very well versed on these kind of things. I know it might be a let-down if your initial idea might be too difficult for the meantime, but it's much better to suffer a little disappointment now and wait a while until you have finished learning about the organ's electronics, than it would be to ignore Jim's advice and bust up such a special treasure :)

[moebius]

Although I don't know about mechanical/electrical construction here... just kinda remember that these organs use fairly big currents in driving voices from the keyboard.

Why (DIN) input from the keyboard would be anything but tappin' contact voltage from the keyboard thru resistor driving a transistor acting as a switch for DIN?

I'm not a FET guy, but it also seems fairly easy to drive FETs from DOUT, like done with a "competitors" board:

http://www.geocities.com/JDPetkov/Hardware/mdec64/Vern/valvdriv.gif

(It's for driving organ valves, but with some modifications it should work for "switching" keycontacts to bus)

All this would need a common ground between organ and MB, but seems doable. So, I'd say learning curve is there, but it's not really that hard.

Moebius

[jimhenry]

Why (DIN) input from the keyboard would be anything but tappin' contact voltage from the keyboard thru resistor driving a transistor acting as a switch for DIN?

Analog organ circuits are all over the map.  It is possible that there is an audio signal being switched by the keys.  If it is DC, it is impossible to say what that signal feeds into and whether the minute pullup current on a DIN would upset the balance of the circuit.  It might well be an easy thing to do and it might not.  Unless you have the circuit diagram and you can evaluate the effect of adding the DIN circuit to an existing analog circuit, you just don't know.

[OrganGrinder]

I have the service manual to the organ, a Gulbransen 600 series.  Just because I don't have experience or much knowledge of electronics dosn't mean I can't read a curcuit diagram.  I have an IT degree majoring in Software Engineering and can read UML (very similar to circuit diagrams).  It's just the electroncs side that my knowledge is lacking.

Anyway back to the organ.  The keyswitches are directly powered from the power supply at 15V DC, the tone signal is not added until after the signal passes through the gate matrix (the next curcuit in line).

My thought was to include the lines to Din and Dout (isolated from each other with a diode) and connected to the keyswitch curcuit boards, I believe there is some curcuitry on the keyswitch boards in case of feedback from the gatematrix so connecting through the wires is probably not a good idea (not to mention difficult as they are bundled together).

If I had access to a scanner, It would be easy enough to give a diagram of the workings and where I was thinking of connecting the lines to Din and Dout.  Anyway the planned locations were near the keyswitches before too much additional circuitry got involved.

The idea for Din was that when a key is pressed, the current passing through the keyswitch would also go to Din (with appropriate voltage reduction) and continue without problems.

With Dout, the idea is first in software a check to see if the key if already pressed is made, if no Din signal then provide the Dout signal.  The Dout signal would be amplified (preferrably with the same power source as the keyswitch pcb) and sent through the pcb at an appropriate location where it wouldn't also activate Din but still activate the rest of the board as if the keyswitch was pressed.

From what I can work out, this is the only place I could include Din and Dout without interfering with the rest of the organ (as jimhenry has warned about) and still maintain the functionality I am after.

I will have to check again, but I think the stops use a conditioned signal so directly connecting Dout to them would probably cause trouble with the workings.

I haven't mentioned the peddle board yet.  The peddleboard is run by a 2in/13out diode matrix and should be easy enough to handle in software if I can't find a hardware solution for it.  ie check the in lines to determine which note is played by the respective out lines.

Organ Grinder

[jimhenry]

It sounds like you were being modest in your original post.  The newest version of SmashTV's DOUT board can use driver chips rather than current limiting resistors.  Perhaps you can find chips that will allow you to deliver the higher voltage you need to drive the tone generators.

[DrBunsen]

I wrote here about a shift register from Sparkfun that puts out up to 50 volts on 8 pins.  Could be a good DOUT replacement.

[seppoman]

Hi Dr,

just a small correction - the TPICs are sink drivers. That means they don´t "put out" any voltage, they just "take in" up to 50 V. If you want to switch anything with them, you´ll have to apply a common positive voltage to the objects (e.g. relais) and then the TPIC can switch on/off a connection to ground.

Another comment: It´s nice that Sparkfun decided to sell these ICs. But there are a number of other distributors for these ICs and the manufacturer (TI) has a very generous free sampling program - I got 5 ICs for free.

Seppoman

[OrganGrinder]

Hi all

its been a while since i gave an update on this thread.

first i have been working on repairing the gulbranson 600, so far with limited success.  this is mainly due to the fact that i am finding it difficult to issolate circuits to confirm that they are operating correctly.

so far, there was no problems with the keyswitches so no fixing there.

all the stops on the lower manual are working now, turned out the problem was one part of a lm3900 (a quad amp chip) which cause problems with the 4' stops.

the upper manual appears to be a much bigger problem.  it appears that i may have problems on the "gate matrix", this board with use of daugherboards (which are fixed and not easily removable) mixes the inputs from the keyswitches with the analogue signals from the tone generator before sending the overall signal to the appropriate flute and non flute stop controls.

now it is difficult to do anything to the gate matrix motherboard simply due to the large number of wires involed (5 octaves of keyswitches, 8 octaves of tone generator inputs, and 9 octaves of tone outputs - some 233 wires), and that is just to get complete access to the board (remove it to a workplace) so that i can properly diagnose and repair it.  as i mentioned before the daughter boards appear to be fully attached (soldered) to the motherboard, and each mother/daughter board interface has over 70 connections.

it looks like the possibility of repairing this thing is beyond my skill and probably beyond what is resonable to ask a service guy to repair.

on the bright side, i believe i have figured out how to suppliment the keyswitches with Din and Douts (and not blow the PICS).  unfortunately i am not able to scan images so an easily understandable explaination is not possible at this time (maybe if a borrow a digital camera, it might be good enough).  afterall i would have liked some feedback before i started cutting wires.

so to sum up, it looks like my plan to repair the organ and midify it may not work afterall - but i am not giving up yet, a working solution may still be there.  otherwise it will have to be a radical midification.

Thanks for reading me ramble

OrganGrinder

PS - pk, i share your pain at finding out what thoughless people have been doing to your efforts to provide them with usable organ music.

people now days seem only interested in what is popular and not what is good.  organ music is commonly seem as "old fasioned", but my response is that people now days are so impatient, wanting the quick fix (ok sounding guitar can be done in a few days hard practice, reasonable organ takes a fair while longer).  this is reflected in the music industry where the money is in low skill level good enough as opposed to properly developed quality music products (look at the training difference between rock musicians and classical (or jazz)).  the scary thing is that the normal person dosen't appreciate the difference. (anyone who begs to differ, consider the last time you sang with a pop melody instead of being deafened by THUMP THUMP THUMP)

PPS if anyone wants to discuss what make good music or any other phylisophical dogma, please start a different thread, i was only expressing an opinion with no intention of starting a debate.

[stryd_one]

ALL of them had been jettisoned in favor of contemporary service music, including the pipe organ.

That's really sad. I imagine that would have been intensely frustrating for you.... :( The shame about this is, that it's yet another example of the church not knowing enough about the subject they're supposed to be teaching the parishioners. At least historically speaking, the lengths of the pipes and the corresponding wavelengths of oscillating air pressures and the interrelationships and ratios between the harmonies that they create are all engineered in replication of (or as inspired by) God's design of the universe. The church clearly went to some effort to destroy that form of worship  ::) I don't claim to be a good christian or anything but I'm not teaching it either ;)

people now days seem only interested in what is popular and not what is good.

That's because they haven't learned the difference. It's a fine line between singing hymns and karaoke  ;)

[greenfox]

As some of you have already noted, I have already headed down this track (see Midifying Yamaha E5AR).  I made the decision to cut out the existing circuits.  This was an easy decision for me for a number of reasons.  A/ I never did like the sound of Yamaha organs.  B/ I paid next to nothing for the organ with this intent to Midify it.  C/ It had numerous electronic faults which would have taken a lot of work to fix and with experience in trying to do this can often cause new problems and no sooner things are fixed than something else falls over.

Having said this, if it was a Conn 650 I was planing to work on which was in good working order, I would have great difficulty in cutting its heart out.  I would like to find myself a Conn 650 some day.

There are 2 ways I would contemplate doing the Midification without destroying the original organ.  A/ chose one of the buss bars to use as the Midibox switching.  It will mean disabling the original sounds switched by just this one buss bar.  B/ install reed switches and magnets to create a new set of switches completely independent of the original organ circuitry.

I cannot see any way of using Midi signals to play the original organ circuitry.  You would need DOUT signals for every buss bar of every note.  I am also hard pressed to think of many reasons why this would be needed.

It is certainly a good idea to use the existing amplifiers and speakers if they give a good sound.  This was not practical in my case with the Yamaha as the sound through the amplifiers and speakers was very poor.

I have found the sound of Miditzer and the true-to-history function of the organ, far and away exceeds any electronic organ I have had the privilege to play.

[organizer]

OrganGrinder writes:

...so to sum up, it looks like my plan to repair the organ and midify it may not work afterall - but i am not giving up yet, a working solution may still be there.  otherwise it will have to be a radical midification.

Hi OrganGrinder.  I'm new to the forum and read your thread with interest.  I'm preparing to midify my recently aquired Gulbransen Rialto II.  Like you, I had some challenging repairs to make before I began.  I've been reasonably successful.  Replaced several of those LM3900's, as well as other parts.

My organ has a gate matrix which I suspect bears some similarity to yours.  I had several shorted diodes on my daughterboards which gave some quite bizarre symptoms.

Anyway, if you would like to contact me directly, I'll be happy to offer what assistance I can.  I have a service manual and was able to do a pretty good diagnosis by careful analysis of the symptoms.  It takes time & patience :)

Tony

[greenfox]

Since we only have a single pedalboard, it must serve dual duty: play as designed with the studio Conn ... and convert to MIDI when plugged into the touring replica of the San Bernardino 216.

That will require MIDI triggers either on the pedal clavier itself, or on the touring console it plugs into. Any preferences?

With my project I have put reed switches on a strip of timber under the toe end of the pedal board.  I have screwed doenut magnets to each pedal.  This means the pedal board is self contained with one DIN and a CORE (I am currently building a second DIN and banks of pistons).  I did this for similar reasons to what you mention.  I currently have 1x Casio keyboard and plan on a second and third.  I will use the pedal board and pistons with both the portable set up and the E5AR setup.

The Yamaha has tongues that press switches in the body of the organ (witch I do not use but have left there).  From my experience console organs either use the tongue method or reed switch and magnet method.

[greenfox]

The other project on my MidiBox to-do-list is: 3x analogue in signals on the CORE I use for pedals and pistons.

This function will give 2x Swell (Main & Solo) and 1x Crescendo pedals.

A couple of the members of the virtual organ forum have got this working.  It requires a few modifications to the MIDIO128 main hex.  (Aparently an analogue input card is not required as the CORE will accept I think 8x analogue in's direct on the PCB)

Talk about MidiBox growing into new applications.  It may be useful to wright this function into the standard MIDIO128 files so it can be easily activated and configured through the "mk_midio128.syx" configuration.

[OrganGrinder]

hi all

i have just midified one of the manuals of the gulbranson!!!!

once i got the hang of what i was doing it was easy.

i think i can get a scanner so i can show what i did to the schematics, on the organ i only needed to remove 2 resistors per key, cut a few wires and connect the wires from the din to the circuit.

now to figure out the pedal controllers - this could be a little harder as the pedal contacts work through an active matrix, but i'll see what i can do...

OrganGrinder

[greenfox]

good stuff. ;)