Pictures of the innards

close-up photo
Top cover removed
photo
Top switch panel raised
photo
Upper manual raised
photo
Both manuals raised
photo
Settings memory capacitors
photo
Inside, rear view (look, no Leslie)
photo
Power, amplifier, & expression pedal
photo
Back of keyboard, & circuit boards
photo
Reverb springs

Disassembly

Remove the mains power plug from the wall before disassembly.  There are various places inside the organ where you could easily come into contact with mains power.  Also be careful about letting anything conductive come into contact with the insides of the organ while it's powered on, or while any of the capacitors are still holding a charge (even if the organ is unplugged).  It's easy to destroy something by shorting it out.

On the wooden top panel (that has the music rack), remove the top row of screws on the back panel, and carefully push the top panel forward about an inch, then lift upwards.  It may need a bit of a shove, but use something that won't mark the woodwork (like a couple of smacks with the palms of your hands).  It's attached to the two side panels of the organ with keyhole slots.

The roll top can be lifted straight up to remove it, once the top panel has been removed.  I'd carefully vacuum the canvas inside the roll top, as it seems to be the source of most of the dust on the keyboards.  Don't wipe it with anything that'll make it damp, or you're likely to get a mould problem.

The rows of switches above the upper manual are hinged against the side walls of the organ.  There are two screws (accessed from above), one on each side, next to the pivot points, that lock it down.  Don't undo the other two screws that form the pivot point.  This panel can now be lifted up, and none of the cables should need unplugging to do so.

The keyboards are hinged against the side walls of the organ.  They're fastened down by four long bolts, and some shorter ones, that are accessed from under the front of the keyboards.  Once undone, the keyboards can be carefully lifted up.  Again, none of the wiring should need unplugging to raise them.  Once raised, examine the mains power wiring at the right of the upper keyboard, so you know where you'll need to keep your hands well away from, when you're testing things while it's plugged in.  I wrapped mine with insulation tape and left it there.

Remove all the screws around the edge of the perforated back panel, and lift the back panel off (you don't need to do this if you're only accessing the upper section).

The wooden strip between top panel and the back panel can be removed by unscrewing the two screws at each end of it (you don't need to remove this, unless it's in the way).

Most other fasteners should be obvious how to deal with, but be sure that you keep track of which screws self tap, and which go into threaded metal.  And be careful to always put screws back into the same thread of the original holes, else you'll chew through the only thing that will hold them in place:  Place a screw into its hole, slowly turn it backwards (in the unscrewing direction), until it clicks into the original thread chewed into the wood, then change rotation direction, and screw it in, carefully, without over-tightening it.  Treat it like a bolt, not a screw.  You only force a self-tapper into a hole the first time it's fitted into place, and that was when it was built in the factory, not now.  If you chew threw the wood, so the screw has nothing to hold it in place, you'll have to pack the edge of the hole with something like toothpicks or split matchsticks.

Repairs and modifications

Where any hard surfaces rest against each other, they can buzz when the organ is played loudly.  Packing felt between them can eliminate this.  You may have to play with how tightly you screw things back together, as well.  Re-assemble the organ part by part, and test as you go along.  Particularly, test with loud bass tones.  But also test high pitched tones if you'd played with the screws on the tweeters, or if they've slipped out of place by themselves, over time.  You may need to shift the tweeters around, a bit, if you get horrible distortions with very high pitches (the tweeter cones can hit the wooden panel the tweeters are screwed to).  It helps to have someone to assist you when de-buzzing, one of you to play sounds, the other to press against parts to find where the buzzes come from (applying pressure to kill a buzz), fit padding between things, and adjust the tension of any screws.

Intermittent keys

Intermittent keys may be due to dust or other contaminants between the contacts.  Very careful cleaning of keyboard contacts with a small brush, and (perhaps) a very small amount of cleaning alcohol may fix such problems.  I find a toothbrush with half the bristles sliced off, very useful.  Do not bend the contacts under the keys, apply only the lightest of pressure.  Once cleaned, you might want to use some electrical lubricant, to protect the metal.  You can get combined cleaner and lube, use an appropriate product, not WD40 (it's destructive).  Lightly apply it onto a something, then wipe the contacts with it, don't spray into the keyboard.  Be careful not to touch the metal with your bare fingers, it's uncoated copper, and the tarnish will cause a problem.  My problem keys had contaminated copper, I'm guessing that they got touched in the factory (there were visible fingerprints on them).

Be aware that there's a short fine copper wire bonded to one side of the contacts, it's there on purpose to provide a good electrical contact point, don't knock it off when cleaning.

Loss of settings

The memory that holds custom registrations, custom drum rhythms, and the chord computer, is maintained by some large value capacitors that can hold a charge for many days.  Over time, they dry out, and the organ will lose its memory when turned off.  A normal playing session would be long enough to recharge the capacitor to hold settings for several days, perhaps weeks.

These capacitors are mounted on the motherboard (see the above illustration, where I've pointed to them with a screwdriver), and can be replaced with parts bought from any decent electronics supplier.  The organ has three 3.3 farad, 1.8 volt, capacitors wired in series, which can be replaced with a single 1.1 farad, 5.4 volt capacitor.  First, release the cables bound to the wooden strip across the top of the plug in cards (bend the metal tags, leave them screwed in place), unplug the cables going to the daughterboards on the left (mark any cables that are missing labels with texta, so you put them back on the right sockets), unscrew the wooden strip and remove it, lift off the grooved rubber strips across the top of the daughterboards, then carefully lift out each daughterboard (they're all connected with plug and sockets).  You should take anti-static precautions while handling them.  Stack them in order, for ease of replacement; but they are all labeled, and so are the motherboard positions they plug in to.  Do not be tempted to remove the motherboard with all the daughterboards still plugged into it—it won't take the strain, and you'll break things all over the place.  Unplug all the cables to the motherboard, again marking any that aren't labelled.  Carefully squeeze in the centre barbs of the white plastic pegs that fasten the motherboard to the wooden chassis, so that the barb passes through the hole in the motherboard as you carefully lift the board up.  Be careful not to flex the board too much.  Remove the board, make sure the new capacitors are discharged before you solder them in (don't dead short them, put something like a 5kΩ resistor across them, for a while, to gently discharge them), swap the old caps for the new one(s), making sure you get the polarity the right way around, and replace the motherboard.  Replace the daughterboards carefully, try not to bend the motherboard (you may need to slide something plastic or wooden under the motherboard—to give you something firm to press against, that won't damage the motherboard, itself, nor short anything out).  Try to insert the daughterboards vertically and not at any other angle.  Reconnect the cables.  When replacing the grooved rubber strip that keeps all the boards neatly apart, check that no boards touch each other.  I had to place felt between an inductor and the metal shield of another card, to stop them vibrating against each other.  And if the top strip of wood (that fastens them down) doesn't hold the boards snugly, you may need to pack some felt between the rubber strips and the wooden strip.  Don't bend the motherboard down by packing it too much, just fill up the slack so the cards can't wobble.  Originally there was foam, underneath it, as a filler, but it'd squished completely flat over the years.  Anchor the loose wires back down again, check that the hinged panels can move properly without tugging on the wires, and that the wiring doesn't buzz against something when the organ's played loudly.  If they do buzz, reposition them, or wrap felt around the problem areas.

Considering that I may have to replace the capacitor again, in the distant future, and not wanting to go through all that disassembly and reassembly rigamarole, I haven't directly soldered the new capacitor to the motherboard.  The capacitor's been wired-in above the board on flyleads (and glued down so it doesn't vibrate), where it can be replaced with ease.

And, sure enough, about five years later, I had to replace the capacitor, again.  I don't know if I had a bad one, the first time around, or whether there's other faults.  I'll have to see how this one goes.  Especially as it looks identical (same supplier).

An alternative to removing all the boards and the motherboard:

The capacitor is mounted on the motherboard, with the negative end going to a ground, and the positive end connecting to pin 71 (the last pin) on the SAS1 board right next to it.

Although the caps are on the motherboard, they only connect to the SAS1 board.

It is easier to leave the old cap in place, unplug just the SAS1 board, and work on it.  Cut the track leading to pin 71, so the motherboard's cap is isolated.  Wire the the positive end of the replacement cap to the junction of R1 and D1, and the negative end to a suitable ground point (such as pin 8 on IC2).

WARNING:  This means that the SAS1 board will always have some power on it, even when unplugged (from the charge stored in the cap).  So make sure that nothing conductive touches the board, or make your replacement capacitor an easily disconnectable device.

Wire-wrap terminations

Various connections are made with wire-wrap techniques, such as the terminals on the speakers.  When tightly made with a proper tool, this is more effective than soldering.  But if you don't have the right tool, you're better off to solder wires to any replacement speakers, or properly crimp some terminals onto the wires.  Hand-winding wire around terminals doesn't apply it tight enough.  And loose wires can rattle, particularly on the woofer, and cause distortion, or fall off.

Loudspeakers

The organ specification lists the output as 155 watts, but doesn't elaborate how that power is applied between all the speakers.  The STK465 power amplifier modules can supply 30 watts (minimum), per channel (one channel to the woofer, one channel to a full-range and tweeter, one channel to another full-range and tweeter, and one channel to another full-range and tweeter), but remember that an amplifier driven too hard can supply more current than a speaker typically has to handle, and organs supply continuous tones, so speakers can be pushed to their limits for almost all the time the organ is being played.  Replacement speakers need to be chosen that are able to handle the power fed to them.  Remember that it will be driven hard, and for a prolonged periods.  I had the original woofer replaced with 35 watt speaker, but that was inadequate (it clipped, mechanically, and frequently tore part of the cone away from the roll-surround).  Later I replaced it with one that could handle 200 watts, so it ought to survive being driven hard without any problems.  So far, it has held up nicely for several years, and sounded a lot better than the previous woofer.  Virtually all speakers sound different from each other, and you'll just have to get used to any changes.  The alleged “burn in” period people talk about speakers is far more about your brain forgetting how the old speaker sounded, and getting accustomed to the new one, than any significant change in the speaker behaviour.

The organ's audio PA circuitry connects the full output from the driver IC directly to the woofer speaker (without any high-cut filtering), and the tweeter is wired in parallel across the woofer, with just a series capacitor as a low-cut filter.  There's no real cross-over, so adding your own could be an improvement.  I tried fitting a very basic cross-over filter, but can't say it made any significant difference (the organ isn't Hi-Fi, anyway).

There's no DC protection for the woofers, either, and the common failure modes for the STK465 modules sends the full supply voltage through the woofer's voicecoil, and burns it out in seconds.  So a DC protection circuit could be a useful addition, too.  A mere fuse wouldn't be sufficient, it wouldn't blow to protect the speaker, or would nuisance blow during normal operation.  Speaker fuses protect amplifiers from catching fire caused by a shorted speaker, not protecting speakers from the amplifier.

Audio power amplifiers

I've had two of the STK465 power amplifier blocks die over the years.  The last time it happened it was impossible to get a direct replacement part.  They haven't been made for donkey's years, and the knock-offs that you can find on ebay are a huge risk for being old and worn-out, or untrustworthy counterfeit parts.  I found that a couple of LM3875T ICs can be directly wired in to directly replace one STK465 (the STK465 is a two-channel amplifier, the LM3875 is a single channel amplifier, with a higher wattage capability).

I say it can replace it “directly” in the sense that no modifications are required, the on-board negative feedback circuitry is directly compatible, and the zoebel network on the amplifier output will do (you could up the cap value to 100 nF, though I didn't bother).  But since the pinouts are in a different order, flying leads will be required (see the table below for pinouts).  Just unsolder the STK465 and unscrew it, clean off the old heatsink compound from the heatsink, coat the back of the LM3875T ICs with new heatsink compound, and fasten them to the heatsink using the original screws (though add some washers under the screw heads), and wire them up.

NB:  If you buy fully-plastic-encased LM-3875T ICs, they're already insulated.  But if they have an exposed metal back, it's connected to −VCC, and you'll need to use an insulating washer between the IC and the heatsink (with heatsink compound between ICs, washers, and heatsink), as the heatsinks are directly connected to the chassis.

For stability against possible oscillation (especially since I'm using flying leads), I added some bypass caps across the power supply rails on each IC.  I put a 100 nF and 47 μF (in parallel) across the positive rail to ground, and another pair across the negative rail to ground, directly onto the pins of each LM3875.  The bypass caps shunt any high frequency oscillations on the power supply wires, the two values of caps is to deal with how each cap may respond differently to different frequencies.  The caps need to be able to handle at least 40 volts across them.  And I twisted the pair of positive and negative audio input wires around each other for better noise reduction.  You could do the same twisted-pair trick for the positive and negative DC supply wires, but I didn't bother.

STK465 pins Purpose First LM3875T pins (side speakers driver)
1 + audio in (channel 1) 7
2 − audio in (channel 1) 8
3 ground not connected to the IC, use for the supply rails bypass caps
4 gain/bypassing? not connected, not applicable to LM3875T
5 − VCC (channel 1) 4
6 audio out (channel 1) [pins 6 & 7 are joined together by the PCB tracks] 3
7
8 2nd + VCC not connected, not applicable to LM3875T
9 + VCC 1
10 audio out (channel 2) [pins 10 & 11 are joined together by the PCB tracks] not connected, the other LM3875T uses these pins
11
12 − VCC (channel 2)
13 gain/bypassing?
14 ground
15 − audio in (channel 2)
16 + audio in (channel 2)
STK465 pins Purpose Second LM3875T pins (left speakers driver)
1 + audio in (channel 1) not connected, the other LM3875T uses these pins
2 − audio in (channel 1)
3 ground
4 gain/bypassing?
5 − VCC (channel 1)
6 audio out (channel 1) [pins 6 & 7 are joined together by the PCB tracks]
7
8 2nd + VCC not connected, not applicable to LM3875T
9 + VCC 1
10 audio out (channel 2) [pins 10 & 11 are joined together by the PCB tracks] 3
11
12 − VCC (channel 2) 4
13 gain/bypassing? not connected, not applicable to LM3875T
14 ground not connected to IC, use for the supply rails bypass caps
15 − audio in (channel 2) 8
16 + audio in (channel 2) 7
Photo of
Power supply and amplifier module, with replaced mains power filters, electrolytic capacitors, & audio power amp ICs

Since I don't know why the power amp chips failed, I'm considering putting a (silent) cooling fan into the organ, blowing across the heatsinks.  They're cool to the touch when the organ is silent (except that the ceramic resistor anchored to one heatsink at an angle heats up the heatsink), but they get hot when the organ is singing.

Tremolo switch

The tremolo/chorus fake Leslie speed switch is in awkward place for changing its mode in a song.  A remote switch could be wired across it, in parallel, and placed somewhere more convenient (like how the Leslie switches are fastened under the front of Hammond keyboards—a box is screwed under the keyboard, and sits in front; it can be removed without leaving any visible marks to the dress panels of the organ).  Or another kick switch can be fitted to the right of the expression pedal (it already has one on the left, for pitch gliding or stopping and starting the drum unit).  The original speed switch would be left in the slow speed position all the time (which allows the remote switch to work), and the original switch would still work normally if the remote was unplugged, or the remote switch was left in the slow chorale speed position.  I'd use twin core shielded wiring, with the two cores across the SPST switch terminals, and the shield connected to a suitable ground point on the chassis, such as the metalwork next to the original speed switch (to minimise any external noise being picked up and upsetting things).

I installed a kick switch by screwing a garden gate hinge (the type with one short part and one long part) to the organ base right of the pedal, inside of the black plastic shroud around the pedal, with felt wedged through the gaps in the hinge, to stop it rattling; and felt wrapped around the moving section, to make it quieter to operate.  Then a metal L-bracket next to it, on the outside of the plastic shroud, with a push-on/push-off switch mounted in the L-bracket, facing the hinge.  The wiring is cable-tied to the switch and bracket, so if the wiring breaks off the switch, it can't float around and touch anything else (such as the nearby heatsink and chassis (both connected together to the zero volts rail), or (worse), to mains voltage right next to it.  I made a hole through the black plastic shroud around the expression pedal, so the switch pokes through the hole.  While it's tempting to just remove the shroud, it makes it harder for foreign objects to get into the organ, and exposed mains voltages are right next to the expression pedal.  And space is cramped, so it was necessary to have part of the modifications inside it, and other parts outside the shroud.  So it's that, or build a whole new shroud around the expression pedal, with a different shape to it.  The large hinge gives me a wide area of metal to kick against, instead of fumbling around trying to hit a tiny switch.  And using a kick switch means I don't have to take either hand off the keyboard to change (fake) Leslie speeds.  If I ever get around to it, I might modify this by attaching a wide plate to the hinge, so I don't hurt my foot so much kicking at a narrow section of metal.

The photos (above) were taken long before I did this modification, so here's a diagram (below) of how it goes together.  It fits between the expression pedal, and the small amount of space next to the heatsink on the power amplifier module (as seen in the photos above), as close as possible to the front panel.

diagram
Adding a tremolo kick switch beside the expression pedal

Remote switching

The same sort of remote switch modification could be done for the registration memory setting buttons (putting secondary buttons in a place that's easier to reach).  Though I, long ago, gave up on using the registration memories—there's not enough of them, you have to remember what you've programmed into them, the organ gives no visible clues as to what you've registered, and you can't modify the settings (you can only set up the next set of voices that will engage when the registration is released).

If making unpluggable remotes, use a type of plug and socket type that doesn't short any conductors together when plugging and unplugging (e.g. 3- and 5-pin DIN plugs do not short any pins together, but phone jacks do).  Quite apart from the possibility of making annoying noises, you could permanently damage something.


The reverb springs are a bit yucky (in how they sound), yet easy to access.  So I suspect it wouldn't be too hard to tap into their signal lines, and use something else to create the reverb effects.


Other things I'd love to do:  Install longer foot-pedals, so could do proper heel-toe pedalling.  You'd only need to replace the mechanical section with longer struts.  And have a greater-than one-octave pedalboard.  I don't think there's the electronics to directly support it, so it'd be a removal of the originals, make space to put in foreign ones, or sit them just in front of the panel, using new pedals that generate their own bass tones, and patch an audio lead to the organ's auxiliary input.  The original pedal tones don't sound that great, anyway, so it'd be a double-win.


But don't go too far if modifying your organ…

[a faked photo]
A special button

Written by Tim Seifert on 21 August 2007, and last updated 19 September 2023.