Imre's organ project

Nearing the end

2008-11-17T12:09:00.002-05:00

Well, the past couple of weeks went rather well. I put together everything, two manuals, pedal,stop switches, pistons (rewired...). They work fine, except for a few minor problems:

1. Some of the contacts are a bit bouncy. This is not a big problem for the keys, I can only hear that with the harpsichord sets, but are a bigger concern for the stop switches. Eventually, I want to make them pushbutton stlye, i.e., they return to their original position and the status is indicated by an LED. The current stop switches are just too bouncy to make it reliable. I'm experimenting with several different pushbuttons to see which one would work fine.

2. Occasionally, some notes get stuck. I haven't been able to discover the cause, it can be a short circuit, such that too many MIDI messages are generated and my interface loses some, but it can also be Hauptwerk missing some of the messages. I have the feeling that it is the first one.

Still to come are the expression pedals. And also, to make everything look nice. I plan to install a 19" LCD monitor on the music rack, so that I can play music in PDF. But I want to make it so that I can use old-school printed sheet music, too.

Ups and downs

2008-10-31T11:52:00.002-04:00

The HWce MIDI input encoder board has arrived. Just in time, because I had the lower manual and the pedal all wired up, ready to test. I had a hard time insterting my wires in the connector. Once I managed to connect everything, I noticed that I held the plug the wrong way. Anyway, I forced it into the socket and it all worked. Then I removed all the wires, turned the plug around, inserted the wires again, and ... a few notes were missing. Obviously, my wires were too thick for the plug, and not all of them connected properly. Finally, I took a ribbon cable from an old hard drive and used that to wire the plug, then I just soldered the ends to the ends of my wires. It look very ugly (will make them nicer), but all the keys worked.

Last night I was able to practice a bit on St. Anne's. It was a great experience, I wouldn't have thought some years ago that I would be able to play the organ in my living room. It gave me an energy burst, so I went down to the garage and cleaned the contact pins on the upper manual. It is almost ready for the diodes now. I also bought some extra cable and 50 more diodes.

Another minor snag: as I have 3 combination pistons per manual, I have 61+3 connections on a manual. It is just the perfect number for an 8x8 scanmatrix. This is the way I wired them up. However, I just read the Hauptwerk User Guide (better late than never) and discovered that this was a bad idea. The transposer in HW acts on the MIDI Input Path level, so it influences the combinations, something you typically don't want to happen. So I just need to rewire things to be with the stop switches on Channel 1, and remove that Input path from the transposer.

Main manual completed

2008-10-27T09:35:00.008-04:00

This past weekend I've finished the work on the lower manual. It went quite well. I noticed one thing, however. When talking about MIDIfying a console, people always think about wiring, diodes, MIDI encoder boards, etc. In fact, CLEANING takes much more time, and it is very boring and repetitive. 61 keys, 122 contact pins, 61 contact frames, 61 keys... Anyway, It is done now, but it was enough for this weekend, so instead of starting to clean the contacts for the second manual, I completed some smaller tasks.

Back to the project. As I have 3 pistons on each manual, I've figured it is best to wire them together with the keys on the manual, thus creating 61+3=64 connections, just enough for a 8x8 scanmartix. I've decided to use two contact pins for the keys, just to have some redundance. For this, I soldered the diodes so that that leads connect the two rows of pins:

I made them alternate by fours, because I used a 4-wire phone cable to connect them, and I wanted to keep the same colours in separate tunnels. This way rows 0-3 of the scanmatrix face one way, rows 4-7 the other way. Here is a picture with half the diodes connected:

I also wired up the 3 pistons as if they were notes c#, d, d# above the highest note of the manual. Hauptwerk will take care of them.

Following that I cut up the common bussbar into pieces of 8 contacts and wired them up to form the columns of the scanmatrix. I made sure that the contacts from neighbouring columns couldn't touch. Finally, I covered the wiring with the original coverings, so you can barely notice them.

This is what the complete assembly looks like (notice the markings on the cables!):

I used two rows of contacts, I had to activate them. These used to be contacts for different stops, one row per stop. Unlike with the pedalboard, here I cleaned only the ones I use. (As a side note, I could have made a velocity sensitive keyboard using two contacts at different positions from the copper frame.)

The rest is easy and quite spectacular. I took the keyboard frame upstairs and installed it in the console:

Then I added the keys, springs and some keyboard hardware. Finally, I had my son check the weight of the keys:

They feel good.

I've also disassembled the upper manual, removed all the old wiring, but haven't started cleaning it yet. On the other hand, I've cleaned the contacts for the stop switches. There are fewer of them, and they are not so close together (my eyes get tired of staring at all those pins for more than 5 minutes at a time). I've also tested the MIDI In port of my M-Audio FastTrack Pro device. It works fine doing MIDI In and SPDIF Audio out at the same time. Following a tip, I have also located a suitable power source for the HWce MIDI encoder board (soon to arrive, I hope): it is the power adaptor of my old US Robotics serial modem!

In the meantime, I ran out of cables! I don't really know why, because I've calculated how much I would need, and added 50% extra. But still, I need almost as much as what I've used so far. Go figure...

50 years ago...

2008-10-23T08:33:00.002-04:00

50 years ago someone in an organ workshop in Kansas marked my pedalboard with the date 10-23-1958. (S)he had probably no idea that in 50 years the console would end up in my living room. In a couple of weeks I hope to have it in working order again, outputting MIDI signals (which hadn't even been introduced until 1983) and driving Hauptwerk. If it lasts another 50 years, I'll be happy.

Pedals complete

2008-10-20T12:44:00.003-04:00

I've finished wiring up the pedals for a contact matrix. The final result is quite nice, although not as nice as it could be.

I made a mistake at the beginning and started to create a common anode configuration instead of the correct common cathode one. I had to reverse all the diodes, which in some cases meant that I had to cut the leads, as unsoldering was not possible. Anyway, the colour-coded phone wires work great, I'm happy it is a solid wire, it keeps its shape. I tested all the contacts with my multimeter. Here's a trick: if your meter has a diode testing setting (most of them do), use that. It will give you a more informative reading than a simple short/open. Also, in case you have some problems with the wiring, you will get a different number, since you'll have more diodes in parallel, etc.

I installed the pedal contact system at the bottom of the console. I've discovered that there are adjusting screws that can control the height of the contact mechanism, thus essentially setting the point of speech. I've adjusted them roughly, will finetune them when there's some sound. I also set the proper weight of speech using a heavy piece of metal (it came from the console itself, it was holding the combination action relay switch). Some finetuning may be needed.

Back in action

2008-10-17T10:55:00.002-04:00

I'm back from the conference. In the meantime I ordered the Hwce encoder from Midiboutique, bought 15 meters of 4-wire phone cables and a MIDI cable. I'll start working on the pedal connections, then continue with the manuals. I hope to complete the soldering by the time the encoder board arrives, so it will not take a long time after that to play the console. I can hardly wait.

The pedals

2008-10-09T13:44:00.006-04:00

Next up is the pedalboard. Not the electrical part yet, just cleaning and taking it upstairs. I would have never thought that 32 can be so many. Wiping them clean one by one was very boring.

I took measurements and concluded that the spacing and width of the pedals conforms to the AGO specifications, but they are slightly shorter than standard. No big deal. After cleaning, I took everything upstairs and assembled them:

The pedals are attached using a very primitive mechanism. No fancy spring, just a piece of steel. I don't expect it to wear out anytime soon. Also, the tighter the screw, the heavier the touch of the pedal. I'll have to adjust them when everything is complete.

Here's the pedalboard together with the console:

Next up is the electrics. I ordered the diodes for the scanmatrices, 200 of them. After they arrived, I started to think: 32 for the pedals, 2x61 for the manuals, 2x3 for the pistons and 41 for the stop switches, that's 201! Well, tough luck.

I was smarter on the cable side. I've measured and computed, and came up with roughly 40 meters. I don't think I have that much at home, so I'll go buy some. But first I'll go to a conference in Washington DC.

It has arrived!

2008-09-30T22:20:00.010-04:00

My organ has finally arrived! It only took a week from Kansas, not too bad. Of course, it had to come when I was teaching, but at least my wife was at home.

After I got home I immediately started to look into it. As the console was built in 1959, I didn't know what to expect. First I put the parts together just to take a look. This is the last time the original 1959 parts are together.

Spiders, dead mice, rotting wood? To my pleasent surprise, the console is in a very good shape inside. After figuring out what's what and trying to understand the workings of the separate parts, I started to remove whatever was not needed any more. I identified the different parts and figured out how the combination action worked. The seller told me it still works, but I wans't brave enough to drive current through anything like this:

It is the big relay switch on the left that stores the combinations. After this, I started to disassemble the console. I had to get rid of stuff that I no longer need, clean every part, and finally, make sure that I can take all the pieces upstairs to our living room. It only took a few hours to reach a state like this:

Actually, the junkyard is not that big, it consists of the old wires and relays, and the wooden boxes that served to hold them. I started with the boring part, cleaning the wooden parts, so that I can assemble it in the living. Also, this leaves more room in the garage to work on the pedal and the manuals. The assembly went quite smoothly, I just wish I had recorded which crew came from where:

After some trial and error it went without a hitch, it was as easy as putting together an IKEA piece.

Before I forget, I also had an apprentice (aka my son):

It's on the way!

2008-09-23T14:30:00.003-04:00

I've just received confirmation that the console has been shipped from Kansas. It should be here by the end of next week the latest. In the meantime, I also discovered that the console came from Zion United church in Kansas City. It was originally installed in 1959 as Reuter Opus 1267.

Introduction

2008-09-22T23:27:00.003-04:00

First, a few things about me. I'm a mathematician, and like many of my colleagues, I also play music. I learned piano through elementary and high school and then had a couple of years on the organ during university.

I've always been in love with the organ, especially Bach, ever since I listened to a tape cassette from my godfather. It was titled "Do you like Bach?" and featured the famous Toccata and Fugue in d minor (BWV 565) played by renowned Hungarian organist Gábor Lehotka, the 3rd Brandenburg Concerto, and excerpts from St John's Passion. I was 10 years old at the time. That's when it all began.