My Hammond Novachord Restoration Project

I have always been fascinated by the sounds of the Novachord, and I am naturally drawn to odd, complex, and difficult to fix instruments.  Finally, after a long time of searching, and the pain of declining a few ridiculously overpriced opportunities to buy trashed Novachords, I scored this gem from a guy in Indianapolis, Indiana on July 14th, 2005.  According to him, it is in non-working condition, and he has had it for ten years.  I plan on posting as much information and pictures as I can about this process, to all who may be interested.  The serial number is 1536.

July 16th, 2005: Here are some pictures of the precarious moving process, much aided by a forklift.  I got it into my work area with a rented refrigerator cart and two strong helpers.

July 18th, 2005: I replaced some caps and powered the power supply up outside of the Novachord on a variac, slowly bringing the line voltage to 110 volts, and all the voltages are within 10% of their no-load specifications.  I have now begun compiling a parts list and determining my plan of attack for the remainder of the restoration.

July 20th, 2005:  After reattaching all the springs that connect the keyboard and generator, which complete the keying circuit, I also put all the tubes back in.  I slowly powered the entire instrument up on a variac and IT'S ALIVE!  All 12 oscillators are working, and some dividers are as well.  It also sounds like the power amp has some leaky coupling caps, the keying caps are leaking, and the preamp probably has bad caps as well.  Nevertheless, I made a recording of what is currently going on.  Click here to listen!

July 22nd, 2005: I bit the big one and ordered basically an entire new set of resistors and caps for this beast.  I think it is best to get all the old caps out of the Novachord whether they are still working or not to prevent future headaches; I'd prefer to marry a woman rather than be married to this thing. . .

July 26th, 2005:  After replacing some of the capacitors in the preamp, a lot of the motorboat noise is gone, and it appears that the oscillators and more than two octaves of the dividers are working, which is astounding for an unrestored Novachord.  This week I am planning on posting some oscilloscope readings.  I tested the amp by itself by hooking it to my B-3 and there is no noise or hum, so all of my troubles are originating in the preamp and before.

August 1st, 2005: A lot of problems are likely rooted here:

This is an oscilloscope display of the B+!  The peaks are roughly +/- 80 volts, and the waveform at the expression pedal matches this reading in contour.  Theoretically, clean up the power supply, clean up the noise.

August 5th, 2005:  The rest of the signal caps in the preamp have been replaced, and I removed the swell capacitor and cleaned that.  I replaced all the caps in the power supply, and all the resistors except the wire wound ones.  The power supply is very strange.  Essentially, it amplifies the rectified DC, then inverts the phase, and amplifies that and mixes it with the original signal.  The ripple is less than 500mV on a 270 volt supply (much better than 80 volts), with the largest cap being 0.5uF, and the voltages are dead on.  I also replaced the three 5Y3G rectifiers with new old stock RCA 5Y3GTs.

Recapped power supply

Swell capacitor - notice the gears

And the ball bearings

Powered the Novachord up again, and now there are some turn on thumps, but after 30 seconds, it sounds like the circuits stabilize, and it is quiet now.  There is a slight hum still, but I have not yet recapped the amp (and let's not even talk about buying a matched quad of 2A3s).  Nearly all of the notes worked, but were badly out of tune.  Each oscillator circuit has two caps, a large one mounted on the bottom of the generator, and a small one accessible from the top to "tweak" the frequency.  I desoldered the leads to the bottom caps, and removed the top ones.  Then, I used alligator leads and a breadboard to sub in caps until I could get the note in tune.  I found if the value was pretty close, you could try different caps with the same value until you found the right one.  After two hours, all the notes were as close to in tune as I could get them with a small Korg guitar tuner.  Hopefully, I can get a hold of a nice rack tuner this week.

So, for the first time, I actually sat down and somewhat played the thing. You can hear the notes are not decaying right, and some dividers are not dividing, but the thing does indeed play. In the recording you can hear a quick demonstration of the resonant filters. Click here to listen!  It's a very eerie sound, partially because the waveform of the oscillators isn't quite a pure sine wave, but looks like this:

August 7th, 2005: The decay problem was caused by a jammed up sustain mechanism.  That was easy.  I have been letting the Novachord run for 3-4 hours at a time, and will, for the next few days.  More notes are developing problems the more it runs, which is good because I can find the problems now, while its all apart, as opposed to having them come out later.

August 18th, 2005: I have been leaving the Novachord run for hours at a time.  Now, if the attack is set to fast, all the notes play, and if it is slow, none do. In between the two settings, it does play normally.  Fun, fun, fun!  Time to take the control panel apart.

October 25th, 2005: Currently, I have all but the oscillator and first divider module removed.  I also purchased a bunch of Panasonic metallized polypropylene capacitors and a good capacitance meter.  I am replacing all the capacitors, resistors, and wire in each of the modules.  I am using Vishay 2 watt metal film resistors.  Slowly, bur surley, progress is being made.  More pictures to come within a week or so.

October 26th, 2005: After removing some caps and seeing something suspicious, I found this:

Apparently, someone crazier than me removed the caps, carefully cut one end open, and removed the guts. A poly cap was then carefully placed in the can. The can was then filled with candle wax, and carefully closed back up.  Upon closer examination, roughly 30-40% of the caps are like this.

September 1st 2006: After nearly a year away from the Novachord, I am back at it again.  I have removed all the caps and resistors from they keyboard.  I used the existing threaded holes that the capacitor brackets screwed into to mount 5 lug terminal strips.  This was necessary because the old caps had a single lead, with the outside being ground, and new components are much smaller that the originals.  I have replaced the springs between the keyboard and generator with screw terminals.