Novachord #346 Inspection & Initial Wakeup

[HideawayStudio]

0.0937 isn´t 10n instead of 100 btw??

Nope... 0.0937uF rounded up is 0.1uF which is 100N.

AG has mentioned the two field coil speakers in passing - unfortunately they need reconing.

I am probably going to retain the 3 oil filled caps in the Novachord - it's the wax/paper caps in the PSU that must be replaced, along with the resistors as many have overheated and some are many tens of percent out in value!

I take on board your comments about the sonic qualities of oil/wax/paper caps vs modern polyester caps. Talking of retaining original passives, I feel carbon resistors will always be inferior, especially noise and temperature stability wise, vs a metal film resistor. Where I do agree, is that there is a good chance that a modern polyester cap could change the timbre of an analog synth - especially a tube synth where most of the circuits are very high impedence. In such cases, more than just the basic parameters of a cap matter. Both the ESR and inductance of a cap could play a part in the timbre too. Where this is very likely to make a difference is in setting the Q of the LC resonators. I would be very interested to know how Phil Cirocco's filter network sounds compared to mine. I would also like to know what his oscillator and divider waveforms look like.

I wish to leave as many original parts as I can in the main generator chassis but I'm not in the PSU purely for safety reasons. If the PSU does not run to spec there really is no chance of the Novachord being safe, stable or noise free (ish). The PSU in the Novachord is abnormally complex for a tube design and plays an active part in noise suppression (mainly noise generated by mains ripple and the divide down architecture itself) - it's therefore very important that it works correctly.

A Section of Over 800 Non Polarised Paper/Wax Caps!

[HideawayStudio]

This is so interesting and exciting.

Somewhere in the afterlife, there is a cranky old man smiling.

Let's hope so... I do hope he approves To the others, the previous owner was George Hardenbergh a key player in the development of early vacuum tube based computers (ATLAS) at Naval Computing Machine Laboratory (NGML/ERA) at St. Paul. He later worked for Univac.

George later became a Ramsey County commissioner. Very sadly he passed away last year leaving a huge collection of vintage equipment:

http://www.minnpost.com/joekimball/2008 ... ardenbergh

I found this lovely quote from 1949 whilst doing my research:

"Where a calculator like the Eniac today is equipped with 18, 000 vacuum tubes and weighs 30 tons, computers in the future may have only 1, 000 vacuum tubes and perhaps weigh only 1 112 tons. " Popular Mechanics, March 1949

I was thinking a very similar thing but about Laurens Hammond et al. One curious thing, being into design myself, is that, during the necessary reverse engineering process, I've found myself thinking "ok that's logical" and "so that's why they did it that way" and then realising the Hammond and his co-engineers must have been through the very same process too.

The overall design is quite nice to work on and there is nice attention to detail. When you turn the PSU chassis upside down the components are layed out and flow from left to right just like the schematic and quite unlike many cheaper "space wired" tube designs, all of the components are terminated away from the tube sockets in the PSU and wire to them indirectly.

Some parts of the design are eccentric to say the least - the way each individual note trigger signal enters the generator chassis is bizarre. Since the generator chassis can be moved into a "service position" each signal is presented to the chassis via an uninsulated spring! There are therefore 72 springs connected between the back of the keyboard trigger circuits and the rear of the generator chassis - it looks really weird. Again, Phil Cirocco replaced all of these with multicore cable and sockets but I wish to retain this as it's such a unique aspect of the design. In terms of noise/safety I doubt this is much of an issue as the circuits are very high impedance and there is so little current flowing the springs would have to be in a very bad state to not make sufficient contact. That said, I still suspect I'll empty half a can of deoxit on them just to be on the safe side.

[Mooger5]

Back to school again, oops...
Is there a line out on the Nova? Those loudspeakers are still worth repairing.

[HideawayStudio]

Back to school again, oops...
Is there a line out on the Nova? Those loudspeakers are still worth repairing.

Fortunately the whole design is isolated and the preamp channel features a differential output transformer suitable for connection to balanced line output on an XLR connector. I have got an XLR connector ready on a metal bracket - I just need to reconfigure the preamp for non-speaker use (there are configuration terminals that either take negative feedback from the PA chassis or just grounds this connection). I'm not sure exactly what the signal levels will be on the preamp output transformer but at least they are isolated from the HT side of things!

This is supposed to improve the frequency response as the PA used in the Novachord is a very early design and is known to lack mainly treble response. The speakers are the original model A-12 12" energised loudspeakers, made by Jensen.

The Novachord Features Two Jensen A-12 Energised Loudspeakers:

[cornutt]

I was thinking a very similar thing but about Laurens Hammond et al.

I can't help but wonder what Hammond really thought of this thing. No doubt he relished the technical challenge, but I'd be curious as to who talked him into the project in the first place, and how they did it. We all know that Hammond originally got into the organ business so as to produce additional demand for his synchronous motors, and the Novachord doesn't really fill that bill. Was Hammond interested in early electronic music? It's hard to say. Current biographies of him barely even mention the Novachord, but they do state that he had a keen ear for timbres and harmonics (contrary to the popular myth that he had a tin ear) and that he understood the underlying theory well. The company never did anything else like the Novachord during Laurens' tenure, but that could have been for purely business reasons. It makes me wish I could go back in time to the '60s, so I could quiz him about the Novachord and what he thought about the work that Moog and Buchla were doing.

[cornutt]

Back to school again, oops...
This is supposed to improve the frequency response as the PA used in the Novachord is a very early design and is known to lack mainly treble response. The speakers are the original model A12 12" energised loudspeakers, made by Jensen.

Also, I understand the built in speakers fire downwards -- not good if it's in a carpeted room. Were there any provisions for connecting a Hammond tone cabinet? Seems kind of strange that they would have omitted that.

[HideawayStudio]

Back to school again, oops...
This is supposed to improve the frequency response as the PA used in the Novachord is a very early design and is known to lack mainly treble response. The speakers are the original model A12 12" energised loudspeakers, made by Jensen.

Also, I understand the built in speakers fire downwards -- not good if it's in a carpeted room. Were there any provisions for connecting a Hammond tone cabinet? Seems kind of strange that they would have omitted that.

Yes - the Novachord has provision for using standard Hammond tone cabinets - you can daisychain as many as you wish as long as they are powered externally - the internal mains harness is only good for one extra PA chassis. There is a special terminal that has to be set for use with external tone cabinets. Incidentally, it's thanks to the option for a second PA chassis that the Novachord's tube count keeps getting wrongly stated on the net - the Novachord has 163 tubes.

The Power Amplifier Schematic:

[HideawayStudio]

I was thinking a very similar thing but about Laurens Hammond et al.

I can't help but wonder what Hammond really thought of this thing. No doubt he relished the technical challenge, but I'd be curious as to who talked him into the project in the first place, and how they did it. We all know that Hammond originally got into the organ business so as to produce additional demand for his synchronous motors, and the Novachord doesn't really fill that bill. Was Hammond interested in early electronic music? It's hard to say. Current biographies of him barely even mention the Novachord, but they do state that he had a keen ear for timbres and harmonics (contrary to the popular myth that he had a tin ear) and that he understood the underlying theory well. The company never did anything else like the Novachord during Laurens' tenure, but that could have been for purely business reasons. It makes me wish I could go back in time to the '60s, so I could quiz him about the Novachord and what he thought about the work that Moog and Buchla were doing.

You need to read Hammond's patents on the electronic instrument!!! There is no doubting this guy was into the idea of electronic music synthesis - he practically invented it. The idea of convoluting dynamics over time is much earlier than most realise (see page from Patent 2,126,464 below).

What might interest some is that his idea for "tonewheels" came about quite by accident. He was playing with a synchronous motor and realised that if you turned it by hand it made an audio signal. Although I suspect this wasn't the first use of a variable reluctance audio generator, it certainly was for use in a music application. He then had the idea to turn several of these mechanisms with one of his own motors.

When Hammond realised just how stable his tonewheels were vs the instability of his Novachord I strongly suspect he though it wasn't worth the effort in terms of technical support. This is a great shame, and yet more evidence that the Novachord was simply too far ahead of it's time. Life is often unfortunate on those who get there first - it's often the second or third variation on a theme that is a success.

Hammond Patent 2,126,464 "The Electrical Musical Instrument"

[HideawayStudio]

I've spent the evening annotating the original Novachord PSU schematic ready to aid me rebuilding the HT PSU chassis tomorrow. I say HT PSU as the heater transformer is located on the main Generator Chassis and drives all 144 tube heaters!! Here is a copy for anyone who is interested. The two noise cancelling HT rails are amazing for 1939!

The Novachord's HT PSU Schematic:

[Mooger5]

Wow, I just saw the power amp schematics. The output tubes are the 2A3 directly heated triodes. They´re in parallel push-pull aren´t they? First time I see those arranged like that. They must be sending out some crazy 10 Watts! There are still "flea-power" amplifiers made today using just a stereo pair in single-end to drive high-efficiency loudspeakers. These tubes are loved today not by their "crystaline highs" or something too "hifiish", but because they´re so informative in the mid-range, where the music is. The 2A3 is still produced, due to the high-demand and are quite pricey. I´d get a new matched quartet from the Svetlana Sovtek factory in Russia, instead of new old stock.

The output transformer in the preamp section is really there for splitting the signal into an out-of-phase pair. I think if the negative feedback loop is entering before the trafo, there may be a significant voltage gain already. With luck the secondaries decrease the output impedance to acceptable levels too... have you thought about a "direct output" from the generators, right after the volume pot?...

[HideawayStudio]

The PSU rebuild so far......

The final quote is in for the replacement HT transformers and under chassis choke - Majestic Transformers have found sufficiently close laminates to the originals that they've asked me to send them the original end bells to fit to the new transformers. The remains of the transformers will be put in the post tomorrow morning - the new ones should arrive in around 2 weeks.

The components finally turned up and I've started the PSU rebuild. When I stripped the PSU down I found the original label for the oil filled 2 X 4UF cap. It has partially rotted away so I scanned it and, just for fun, I filled in the missing letters in Paint Shop Pro. I then printed out the new label and decided it looked too new so I soaked the label in tea for a couple of hours - the final result was very convincing!

The Oil Filled Cap with it's "New" Label:

I then moved onto to rewiring the on-chassis chokes with modern PTFE high temp wire. Very fortunately these chokes survived the HT overload and test fine.

Rewired Choke: (Terminated within End Bell)

I started to remount all of the tag strips, the oil cap, the chokes and sleeved the grid caps all ready to accept the new wiring and components which I hope to fit this evening and tomorrow. This took a lot longer than I hoped as many of the new brass screws had to be cut down to the correct length.

The Partially Rebuilt PSU With New Paint Job and New Brass Screws:

[cornutt]

Very nice work! Are those the original tube sockets, or did you have to replace them?

[HideawayStudio]

Very nice work! Are those the original tube sockets, or did you have to replace them?

Thanks - it's slowly taking shape now. The sockets are the originals, and very nice quality they are too. They are held in with circlips, which makes them a joy to remove from the chassis - not a rivet in sight. Amazingly, as I mentioned earlier, every tube pin I've seen so far has been shiny without any major corrosion at all. I simply don't know how this is possible after 70 years. If you look at my tube collection in the attic most of them look like they've been sitting on the bottom of the sea - and most are 20 years younger! The beast must have been somewhere pretty darn dry for the majority of it's life. The tubes are mostly the original Sylvania's, which makes this all the more amazing. I read somewhere that, surprisingly, despite the sheer number of them, the Novachord was not known for tube failure.

The Slightly Less Charcoal Grilled PSU Taking Shape Now:

[HideawayStudio]

After much toil I have completed all the new wiring and soldered in the replacement components to the PSU chassis bar the new HT transformers and under chassis choke, which are due to turn up in a couple of weeks time. All of the passives (mostly 400V rated), except the oil filled cap, have been replaced for safety reasons. I will carefully check it over and externally power some parts of it up such as the heater circuits just to make sure.

The Rebuilt PSU Wiring Bar the New Transformers and Under Chassis Choke:

I am currently liasing with another Novachord owner in Italy to determine exactly what the loadings are on each supply rail so I can build a dummy load. This will permit me to power the PSU chassis up, without fear of damaging the Generator Chassis, should there be an issue.

Novachord #346 PSU Chassis: Before & After Rebuild:

Once the transformers are fitted, after a final check, the PSU will be powered up unloaded via a light bulb based current limiter and a variac for safety. Assuming the rails show signs of life, then the load will be applied and the PSU run up to full spec. Assuming the rails come up ok I will check that the layout of the wiring is ok in terms of handling any high temperatures. Fortunately the wire is PTFE and spec'd to 200C, so I think it will be ok.

Before I can apply any power to the Novachord's main Generator Chassis, I have to fix the fault that caused this sorry mess in the first place. I am 80% certain this is just down to the potted decoupling caps in the pre-amp channel going S/C. They have overheated and leaked bitumen all over chassis - I am told this is a very common problem with Novachords. I know of two others that have suffered this failure mode. These will be removed over the weekend and, working outside, they will be heated enough to pour out the bitumen and scoop out the contents of the caps. They will then be stuffed with modern high voltage polyester caps, wire terminated and then potted in black potting compound. The restored mult-caps will be put back in place and the HT rails checked for shorts.

I also hope to power the heater transformer for main tube array this weekend. This should look very pretty and I'm hoping to take a nice piccy of all 144 tubes glowing

[Mooger5]

You collect vintage tubes and here was I with all the audiophile blurb. I´ll stop now
Impeccable arrangement there, with the Teflon wiring and heatshrink. The PSU looks fantastic
The resistors look so tiny compared to the old ones 1 Watt?

Just in case, these A12s look in very good shape.