Overhauling the AO-29 Amplifier in the Hammond M-100 Series

December 23, 2009

Well built vintage electronic circuitry can last a very long time, as evidenced by the number of Hammond organs that are still working, some from as far back as the 1930s. Hammond’s engineers designed their vacuum tube (valve) amplifiers conservatively, putting minimal stress on the components. All but one of the tubes in my 1962 Hammond M-111 is original and I have no intention of replacing them unless one fails.

However, one type of component that is prone to wearing out, regardless of how little or how much a piece of equipment was used, is the electrolytic capacitor. A capacitor consists of two electrodes separated by an insulating material (dielectric). Capacitors store electric charges and block the flow of direct current through them while allowing alternating current to pass. An electrolytic capacitor is one where the insulating layer is porous and impregnated with an electrolyte solution (somewhat like a battery).

The characteristics of this electrolyte change over time, potentially affecting the capacitance. There is also a risk of short-circuits forming, which can result in the capacitor leaking or exploding. The corrosive electrolyte can damage wiring or the chassis, but more importantly, the sudden failure of the capacitor can cause damage to much more expensive components such as the vacuum tubes that are relying on it to do its job.

Capacitors in the M-100 Series AO-29 Amplifier

In the Hammond model AO-29 amplifier, electrolytic capacitors are only used for filtering. Large capacitors are used to filter the pulsating direct current from the rectifier into something smooth enough to operate the rest of the amplifier. A few other electrolytics are used throughout the amplifier to provide local filtering. Since aging electrolytic capacitors tend to increase in capacitance, they become better filters, so that alone would not justify replacing them. However, minimizing the risk of damage to the tubes, especially the expensive 5U4 rectifier, makes capacitor replacement worthwhile.

There are a total of ten electrolytic capacitors in the AO-29 amplifier. Two of them are on top of the chassis in the large black “can” to the left of the intermediate transformer (when viewed from the back of the organ), four more are in the silver can to the right of transformer, three are on the wiring boards inside the chassis, and one is wired directly to a tube socket.

It is possible to purchase replacement can capacitors from sources such as Tonewheel General Hospital but these are relatively expensive. I opted to replace the cans with separate electrolytic capacitors of the appropriate values instead. The following table summarizes all the capacitors to be replaced:

Designation	Original Location	Original Value	Replacement Value
C14	Silver Can	30µF 25V	33µF 50V
C30	Wiring Board	25µF 25V	27µF 50V
C62	Black Can	50µF 450V	47µF 450V
C62A	Black Can	50µF 450V	47µF 450V
C63	Silver Can	20µF 300V	22µF 450V
C64	Wiring Board	25µF 25V	27µF 50V
C65	Silver Can	40µF 400V	47µF 450V
C66	Silver Can	20µF 400V	22µF 450V
C67	Wiring Board	100µF 3V	100µF 6.3V
C70	Tube Socket	3µF 50V	3.3µF 50V

I ordered the replacement capacitors from Digi-Key (along with new capacitors to replace the aging wax-paper ones in the vibrato line box and tonewheel generator).

Removing the Amplifier

Although it is theoretically possible to replace the capacitors with the amplifier still in the organ, it would have been extremely awkward, so the first step was to remove it from the organ. This required unsoldering a row of connections along the back of the amplifier, including the AC power connections enclosed by a metal cover and some shieled audio signal connections under another cover. To ensure I’d be able to reconnect everything properly afterward, I took several close-up photos of all the connections before starting.

Once all the connections were unsoldered and neatly tucked out of the way, I loosened the screw holding holding the expression pedal lever on the shaft of the expression pedal capacitor (inside the silver box on top of the amplifier) and slid the end of the lever off of the shaft.

Next I undid the four bolts that hold the amplifier chassis to the floor of the organ and carefully lifted the amplifier out. I debated whether I should remove the tubes before or after extricating the amplifier, and decided to wait until the amp was on the workbench. I was afraid that a hard-to-remove tube might let go suddenly and that I would smash it against something in the organ.

After carrying the amplifier to my workbench, I carefully removed each tube, placing them in consecutive compartments of an egg carton. I inspected each tube as I removed it to make sure I could still read its markings in case I managed to mix them up in the carton. Once all the tubes were out, I set them aside, out of harm’s way.

Replacing the Can Capacitors

Before beginning, make sure that any high voltages stored in the capacitors have been discharged. Using a clip lead, connect one end of a 1kΩ resistor to the chassis. Then using another clip lead, touch the other end to each capacitor positive terminal for a few seconds.

The first step in replacing the cans was once again to photograph everything in great detail. In addition to the photos, I drew a diagram indicating all the connections and where they lead to. Finally, I marked each wire to identify the terminal it was connected to. There are a lot of wires connected to the capacitor terminals and many of them are the same colour, so I did everything possible to avoid a later mix-up.

The individual capacitors within each can are identified on the side of the can with a symbol for each one (e.g. triangle, square, semicircle), and the contacts on the bottom of the can are marked with the same symbols. There is one solder lug for the positive terminal of each capacitor, whereas the negative terminals of all the capacitors in the can are connected together and to the can itself. In other words, the four twist-tabs that serve to hold the can to the chassis are also negative terminals for all the capacitors.

The Black Can (C62 and C62A)

The following steps describe the work done to replace the black can (containing two 50µF 450V capacitors, C62 and C62A).

Removing the Can

I unsoldered the wires using a 45W soldering iron (not a soldering gun) and a plunger-type solder pump. As I heated up each solder lug, I used the pump to suck away most of the solder. I used desoldering braid to wick up most of the remaining solder. With the solder gone, it was possible to unwrap the wire from each lug and then heat it up once more to pull the wires out.

Once all the wires were off, I straightened out the twist-tabs holding the can to the chassis and cleanly removed it. After removing the black can, there was still the phenolic socket to remove. I did this by drilling out the rivets with a 1/8″ drill bit from the top side of the chassis.

Mounting the Replacment Capacitors

To mount the individual replacement capacitors, I cut a piece of phenolic perfboard 1.7″ x 1.7″. I drilled out three of the holes to 1/8″ diameter. Two of these corresponded with the existing rivet holes in the chassis and the third was approximately in the centre of the perfboard.

Using 1/8″ steel pop rivets, I riveted the board to the chassis and then riveted a 4-position terminal strip to the underside of the perfboard. I connected the two centre lugs together to act as negative terminal for both capacitors. The two outer lugs were then for the positive terminals of each capacitor. The capacitors were installed from the top side of the board, paying careful attention to polarity and keeping the leads away from the chassis edges, and their leads were passed through the holes in the appropriate lugs underneath.

With the capacitors in place, all the wires were routed to their appropriate lugs. After carefully checking the wiring against the original photos, drawings, and the markings on the wires themselves, everything was resoldered. The end result was quite neat (by 1960s point-to-point wiring standards).

The Silver Can (C14, C63, C65, and C66)

The silver can differs from the black one in two ways: it contains four capacitors instead of only two, and the negative terminal solder lugs are fastened directly to the chassis instead of to an insulated socket. Since there are four capacitors, I needed at least five solder lugs (at least one negative and four positive terminals). Instead of a single 5-position terminal strip, I used two 3-position strips with the mounting lug at one end of each. I fastened one strip with each rivet holding the perfboard in place, which served to connect the end lug to the chassis.

I cut a short length of 14 gauge solid copper wire and passed it through these two end lugs, effectively forming one long negative buss passing diagonally under the perfboard. Three of the four capacitors were then installed on top of the perfboard with their negative leads connected to the copper wire and their positive leads connected to individual lugs. No matter how I arranged things, I was unable to fit all four capacitors above the chassis, so I installed one of the 22µF capacitors (C66) underneath, connecting it directly to the terminal strip lugs.

With the capacitors in place, I reattached the wires to the appropriate solder lugs and then soldered everything.

Replacing the Other Capacitors (C30, C64, C67, and C70)

The two 25µF 25V electrolytic capacitors, C30 and C64, are near the two ends of the wiring board. Before removing the old ones, I marked the bottom of the board with the polarity so I’d be sure to install the replacements the right way around.

Removing the old capacitors was easy, just requiring a few seconds of the soldering iron on the underside of each turret to melt all the solder within, at which point the capacitor’s lead could be pulled out. Installing the new capacitors was the same process in reverse: melt the solder in the turrets and push the leads in.

For some reason, I forgot to replace C67 and C70, which I did not realize until I began to write this article long after reinstalling the amplifier in the organ. Since these capacitors are not riveted in, thus just requiring unsoldering and resoldering, I can probably replace them without removing the amplifier. Fortunately, neither of these capacitors are as critical as the power supply capacitors, and a failure would not likely damage any other components. Both are used as filters in the bass pedal circuit.

Other Maintenance

While the amplifier was readily accessible, I decided it would be a good idea to do some general maintenance:

• The chassis was covered with what looked like dust but which was almost impossible to wipe off. I suspect it was dust, held in place by a thin layer of Hammond tonewheel oil that had accumulated over the decades. I removed it using rubbing alchol, a rag, and a toothbrush.

• 

  Touching up the chassis markings with a marker made them much more legible.

  The chassis has labels stamped into the metal next to each tube socket (identifying the tube type) and external connection (giving the wire colour and designation). These were almost illegible in most lighting conditions, so I darkened them with a black Sharpie® permanent marker, followed by a wiping with an alcohol dampened rag to remove the marker everywhere except within the stamped letters.

• Before embarking on this exercise, I was experiencing a problem with crackling noises emanating from the speaker when the organ was played at high volume. Consulting with the folks on The Organ Forum suggested that it might have been a tube moving in a dirty tube socket, causing an intermittent connection, so I thoroughly cleaned all the sockets using DeOxit D-5 spray.

• I cleaned up all the external connection terminals using the solder pump and desoldering braid to remove all the excess solder and rosin.

• I had briefly looked into the amplifier when I first purchased the organ in 2008 and found one capacitor (C70) that had become disconnected from a tube socket terminal (which is how I know that a failure of C70 won’t cause any damage). With the amplifier on the bench, I gave it a more thorough inspection with a magnifying glass to check for other broken connections.

Reinstalling the Amplifier

With all the capacitors replaced (except C67 and C70), I gave everything one more check, comparing my rewiring against the original photos and drawings, and also the schematics from the service manual for good measure. Everything looked good, so I carried the amplifier back upstairs to reinstall it into the organ.

I first bolted it back in place and then started reconnecting the wires from right to left. After everything was soldered to my satisfaction, I replaced the two metal covers (one for the AC line voltage and one for the shielded audio connections) and then reinstalled the tubes. To keep things neat, I also wrapped the bundle of wires leading down to the amplifier with electrical tape.

One last check and then I started the organ. After letting it warm up for 20 seconds or so, I tentatively pressed a key and was rewared with a note from the speakers. A thorough check showed that everything was (still) working, except the expression pedal. I quickly realized I forgot to reattach the pedal arm to the side of the dog house.

Uh-Oh!

With the expression pedal reconnected, I sat down to play something. Part way through the piece, I heard what sounded like a capacitor going up in smoke (a sort of whooshing sound). Strangely, there was no smoke and no smell, and everything still worked. I looked behind the organ and everything looked fine. I continued playing and the whooshing came and went. Usually, changing anything (like turning the internal rotary speaker on or off) would make the sound disappear for a while, but then it would return.

I would have been concerned that something I did had caused this problem, but it had happened to me once before. I did some more experimentation and discovered that if I pushed the expression pedal all the way, there was a clanking noise and the organ got very quiet. I came to the realization that the plates of the variable capacitor that controls the volume were hitting the wall of the dog house, thus shorting the signal to ground. I took another look at the expression lever and shaft and noticed that the lever end wasn’t seated all the way onto the shaft. I loosened the set screw, pushed the lever further onto the shaft, and retightened the screw. After that, the expression pedal worked properly over its full range and the whooshing was gone. My theory is that the loose attachment of dissimilar metals created what amounted to a diode junction, and that this was producing white noise that was getting into the amplifier.

Conclusion

Did it sound any different when I was done? No, but I didn’t expect it to. All of the old capacitors still measured very close to their stated values, so they were still working as designed. Unlike tonewheel generator and vibrato line box capacitor replacement, this project served primarily to reduce the likelihood of a future catastrophic capacitor failure and the more extensive (and expen$ive) damage that could cause.

                    

Related Articles

If you've found this article useful, you may also be interested in:

• Rebuilding a Hammond AO-29 Amplifier from the Ground Up
• Hammond Organ Tonewheel Generator Capacitor Replacement and Calibration
• Window Seat Bookcase Tone Cabinets for a Hammond Organ
• Retronome – A Versatile Analog Drum Machine for My Hammond Organ
• Overhauling and Improving the Hammond M-100 Series Vibrato System
• Adding a Rotary Speaker to a Hammond M-111 Organ