Console Rebuilding

Back ] Home ] Up ] Next ]


When we stored the organ parts in the warehouse, it was only supposed to be for a short time while the new church building was constructed. We did not realize how slow the approval process was for construction permits, nor did we realize how long the actual construction and final approval for a certificate of occupancy would take.

While we waited, we started planning the console conversion. While we intended to hire a professional organbuilder for the installation and voicing work, we needed to save money by doing as much as possible with volunteers. With my experience in building microprocessor-based controllers for radio repeater systems, I was able to evaluate competing commercially-available systems for pipe organ control. I studied product information and visited some completed installations in order to see the actual system installed.

One important factor was the price/performance ratio. Not necessarily the cheapest system, but the most versatile.  The main reason the Artisan Instruments system was picked was that it was the most user-modifiable system with the fewest proprietary parts. This meant that if we wanted to change something, a simple software change (which we could do in minutes on-site, without factory assistance) accomplishes most tasks, and even more extensive changes usually require only connecting extra unused outputs to what you want to control, and adding a couple of lines of source code to the organ definition file, compiling it and downloading the new instructions to the control units (see the hardware and software pages for details).

Along with the purchase of the new control system, we had to replace the drawknobs and tilting tablets with electrically-operated ones. We chose Peterson units because they had a nice look and feel, and fit the console with minimal modification. We also obtained two new Peterson swell shade actuators which have many advantages over the Moller units they replaced.

Moving the console to a more convenient location, all of the internal electropneumatic actions were removed. Also taken out were the pneumatic relays, air regulator, hoses, drawknobs, tilting tablets, indicators, and expression shoe rollers.  All wiring was discarded. Finally, the white paint on the empty shell was stripped as much as possible.

White paint looked fine in Miller Chapel, but did not match the decor of the new church. My original plan was to stain the console wood (mostly birch) but after trying several times I gave up as I was not at all satisfied with the results.  Even with pre-treatment and the use of high quality stains, the results were very blotchy, partly because different types of wood were used and apparently some of the paint (invisible to the eye) remained in the fibers. Perhaps with an extreme amount of sanding it might have been better, but I had already sanded off as much wood as I dared. So I used a dark semi-transparent stain that covered the differences yet still showed the texture, if not the grain, of the wood.

New wires had to be run throughout the console. Each keyboard was removed...
 ...and the existing wires were unbundled from the cable lacing and then unsoldered from the contact pins. Note how many wires there are for just 61 keys!

This is because each key on the manuals and pedalboard has eight or more contacts, since each note has couplers for other manuals and for other pitches within that manual. 

 

With the new control system, the keys only switch logic levels on an input board and therefore need only one contact per key.

For reliability reasons, I paralleled all the existing contacts so that any or all would switch the input.

The common return line connects to the contact bar on every key via flexible coiled wire.

The wires were then laced with flat teflon cable lacing tape to make a neat bundle. The flat tape does not cut into the insulation of the wires over time the way wire ties or round lacing cord can.

The contacts were also cleaned with a special chemical, then coated with an anti-oxidation coating to keep them making good contact. This was important  since the negligible switching current does not provide any self-cleaning action.

 

Moller used spring contacts mounted as part of the keyboard to actuate the combination action. Unfortunately, in order to replace the wires, all the keys first had to be removed in a time-consuming process since the wire connections were between the rail and the keys.
While the keys were off, the contacts for the pistons were deoxidized and cleaned, and new wires were installed and cable-laced.
Here is a view of the contacts after the keyboard has been rewired and reinstalled. The leather nuts by the green felt near the bottom of the picture adjust the key height, while similar nuts on the return springs (not visible in picture) adjust the key tension.
The new Peterson stop action units (tilting tablets for intermanual couplers; drawknobs for stops, intramanual couplers and tremolos) have five wires each: Positive Common, On, Off, Stop Sense and Sense Common. Although the Peterson units are not particularly sensitive to false sense signals from magnet switching transients as are some competing brands, nevertheless I decided best engineering practice required separating the relatively high current magnet wires from the low-level sense logic wires.
By cabling magnet and sense wires at right angles to each other and running the bundles separately, there is no chance of interaction.

Each unit has wires soldered to a plug-in connector, with heat-shrinkable insulating tubing on the pins to prevent shorts.

There is enough slack to easily change a defective drawknob by simply unplugging the wiring connector and removing the two thumbscrews that mount it to its bushing. The units use magnetic reed switches for reliability.

The Moller expression pedals operated a roller with multiple contacts. These were replaced with potentiometers operated by a rack and pinion gearset. The analog voltage output (0-5 VDC) is converted to a digital signal by the controller to operate the shades or crescendo functions.
Here are the Moller toe studs after just a little bit of polishing, along with what they looked like before.
The socket in the upper left is for the music rack light. The homebrew display with all the resistors is a 20-position LED display that indicates the relative position of the expression pedals. Below it is the Artisan LCD 20 character by four line display that can be programmed to display anything you want. We use it to show exact shade and crescendo settings, memory level, transposition value, etc.
Here are some of the input boards. There are seven of them;  one for each division, two for the stop sensing, and one for the pistons and all other control inputs.
In order to provide alignment and support for the wires, a wooden guide with holes matching the spacing of the screw connectors was made for each board, using a marked piece of perfboard as a drilling guide.

Shown here are two of the four output driver boards. These drive the stop action magnets, indicator LED's and any other outputs.
And here is the control computer, that ties everything together. Actually, it is several computers in one (see hardware for technical details)
Here is a rear view of the completed console. Again, for a more detailed look at the rest of the components, see hardware.

 

Renovation


Home ] Where it Was ] The Big Move ] The Big Move Part II ] The Big Move Part III ] The New Church ] Console Rebuilding ] Renovation ] Final Preparation ] Installation ] Credits ]

This page was last updated on 11/11/08.
All content copyright 2006 Brian F. Bailey, W4OLF