Building a Hammond Novachord

[boxdoctor]

Hi G.M.B.,
What you are describing sounds like a "Bird" portable organ.
Portable only if you were a body-building fitness fanatic!
I had the misfortune to have a customer who used one of them in a band. It was full of neon frequency dividers, which kept drifting off frequency The associated carbon resistors drifted too, and made for many call-outs, once or twice when the poor chap was actually on a gig, and a divider would not lock.
He eventually scrapped it, and bought a Farfisa Duo, a very much better proposition. Tony

[mole42uk]

Sounds like the Selmer Clavioline to me - that was quite small and suit-case-like.

[GMB]

The Clavioline is monophonic.

[trobbins]

The Schober organs transitioned to using neons in the 1960's.

Selmer made valve based organs in the 1950's - as far as I can tell, I have the only restored or even complete one (and I'm in oz!) - as they have been literally gold mines for valve pullers with 64x Mullard 12AX7 and two KT66.

Whilst perhaps not as complex to restore as a Novachord, or a Miller Mk4, there is plenty of effort required for that type of organ, and they are not easy to relocate!

Bren, if you were keen then you may be able to locate a defunct English valve organ and see how close that gets to your dream.

I could imagine that there are a few 6W7G Noval equivalents that sell very cheaply, and allow a mockup of some of the Novachord circuitry to integrate in to an existing subtractive type valve organ such as a Selmer.

Starting from just a surplus keyboard, and an old amp chassis with room to fit another heater transformer may be a fair way to start. Old Wurl's like the 4100B have ready-made subchassis each with 5 noval sockets for their 12FQ8 oscillator and dividers. You wouldn't be sacrificing a bit of history by reassigning one of those organs.

http://dalmura.com.au/projects/Selmer%20Organ.php

[Boater Sam]

An organ based on EF80s would be great for using up the vast unwanted stocks of these valves, go for it!

[Lucien Nunes]

Old Wurl's like the 4100B have ready-made subchassis each with 5 noval sockets for their 12FQ8 oscillator and dividers. You wouldn't be sacrificing a bit of history by reassigning one of those organs.

Perhaps the situation is different in Aus but I have only seen one of these for sale in the UK in the last decade - the one I bought about 8 years ago. This despite running a permanent online auction search for 'Wurlitzer organ' and scouring the results for any interesting models. The only organs from the 'valve era' that still surface in measurable quantities here today are Hammonds and Comptons, neither of which used valve tone generators.

[jez_145]

As has been mentioned, in previous posts, there have been many "divider" organs and other instruments produced over the years.

Whilst not wanting to put you off, I thought I'd share a couple of photos of my 1966 Lowrey Heritage Deluxe organ which is a valve based divider organ and only contains a mere 73 valves (and a couple of hundred neon lamps).

Take it from me, working on gear like this is not for the faint hearted and can be quite dangerous - with anything up to 425Volts DC being present in this particular instrument.

I wish you well attempting a Novachord !


Jerry

[trobbins]

Originally Posted by Lucien Nunes

Perhaps the situation is different in Aus but I have only seen one of these for sale in the UK in the last decade.

Yes, its rare to see a valve based organ even in Aus, but more so of English origin. Perhaps the decision to 'clean out the back room or shed' a few decades ago meant many an organ got dumped. And there even appears to be a 'guitar amp' inspired effort to chase them down in USA, as long as the drive is not too far - as atested by the many US forum threads on the best equipment to scavenge valves from.

Lucien, given I found the same Williamson amp module you have for the Miller Classic IV, I have thus far searched in vane within Aus for any record of such a beast, although I have come across a complete "The Norwich Organ" Model T.

Bren, if you were still keen, you may benefit from some early protoyping, especially to compare 6W7G performance against any alternatives, as there are a few different circuit sections that I assume just use the 6W7G for its low heater current and availability (then).

[jjl]

It appears that the 6W7G is a lower heater current variant of the 6J7.
The closest European equivalents would be the octal based EF37A or noval based EF86. Both of these are sought after audio types and a project requiring tens of them would be expensive.

John

[boxdoctor]

John (jjl) above indirectly touched on an important point in his post above.
Whichever valve you choose for the dividers, keep the heater consumption in mind. A difference of 100ma may not look much on a data sheet, but consider a divider section of the organ with (say) 80 valves. (this is not really an out of the way number for dividers on a reasonable sized instrument), and you have a difference of 8 amps in the overall heater consumption. This , of course, means a much bigger heater mains transformer.
All adds to the cost! Tony.

[cosmocat]

Why all this pessimism about heater supplies ?
Most valves have the heater isolated form the cathode so just string the heaters in series.
So... 80 valves with 300mA 6V heaters
In 10 parallel strings of 8 valves in series = 48V @ 3Amps no problem.
In 2 parallel strings of 40 valves in series = 240V @ 0.6A and no transformer at all! ( just check the heater to cathode breakdown voltage ! )

Fabulous project, go for it.

[trobbins]

Originally Posted by cosmocat

Why all this pessimism about heater supplies ?
Most valves have the heater isolated form the cathode so just string the heaters in series.
So... 80 valves with 300mA 6V heaters
In 10 parallel strings of 8 valves in series = 48V @ 3Amps no problem.

Cosmocat, can I suggest that an audio project using your recommended heater techniques would be a hum and lifetime pain.

Even with your less esoteric scheme, heater hum would likely become significant when the heater voltage on the 'end' valves is 5-6x the pk voltage swing relative to valves 'in the middle', assuming the best grounding scheme with heater mid-string connected to ground.

And its highly likely that variation of heater voltage along a string would not stay within +/- 10%, even with all NOS and same valves.

[boxdoctor]

Originally Posted by cosmocat

Why all this pessimism about heater supplies ?
Most valves have the heater isolated form the cathode so just string the heaters in series.
So... 80 valves with 300mA 6V heaters
In 10 parallel strings of 8 valves in series = 48V @ 3Amps no problem.
In 2 parallel strings of 40 valves in series = 240V @ 0.6A and no transformer at all! ( just check the heater to cathode breakdown voltage !

You're going into the realms of AC/DC techniques with that configuration!
(I presume you meant putting the two chains across the mains input.)
A heater-cathode short in one of the strings may in those circumstances result in considerable damage to the circuit, and, even of you earth the chassis as a safety measure (it would be foolhardy not to), there is still the potential for a firework display at least.
The 10 x 8 valve configuration still requires a chunky mains transformer, as does any other mains isolated heater scheme, and may still be a source of hum, or A.C. ripple on some of the divider's outputs. (Near the top end of the chain.) It does have, however, the merit of lower current consumption, which means thinner wire can be used for the heater wiring than would be necessary for a simple all-parallel 6.3 volt system..
I really hope the O.P. goes for this project, and keeps us informed! Tony

Why not design for 24V or 48V and use for a start a couple or four of duff* car batteries (can be had for nothing at your local garage). 24/48V SMPS are quite cheap and only need to be got when the whole thing works.

*They are only 'duff' in the sense of cranking current, fine for 10's of amps though.

[Lucien Nunes]

Series heaters and tone generators sound like a disastrous combination. Although not individually as sensitive to hum as say a phono preamp stage, there are 44dB more valves and perhaps >60dB more wiring here than in a hifi system, through which hum from AC supplies at high potential to earth can get into places it shouldn't be. I.e. a beefy power amp often run with the gain at max, and fullrange speakers designed to reproduce tones at full power down to 32Hz. Different stages and valves are constantly being switched in and out as you play so it is tricky to 'null out' residual hums effectively, as the null setting playing one chord may be wrong for another. With 6.3V or 12.6V balanced (stray impedance symmetrical about earth) secondary circuits referenced through a humbucking divider pot, it is possible to get away with twisted pair heater wiring in close proximity to signal for many yards, which you will inevitably have.

DC overcomes the hum problem but the safety / reliability disadvantages remain. With this number of valves, there is often a need to poke around and troubleshoot.e.g. tune oscillators / adjust dividers / substitute valves while in use. This might involve a non-technical musician tweaking presets under the chassis during a gig. Tone generator HT supplies are often very limited in current and fairly safe to handle, compared to which a series heater chain could be a greater risk for shock or valve damage even if not full mains voltage.

Economies that worked for mass-produced domestic radios & TVs are not really appropriate in a project of this size, where the cost of a 250-500VA heater transformer and associated wiring is trivial. The Miller IV generators (174 x 12AU7) consume about 26A of heater current at 12.6V, attached pic shows the setup. Four secondaries, one per rank, each split at a tagboard carrying the humbucking pot. The fanned-out white wires are twisted pairs each serving one octave panel (0.9A)

[Radio Wrangler]

500VA of heater power spread through miles of wiring?

This sounds like a job for <<<<< Transistor Man >>>>>>>

David

[Ed_Dinning]

Hi Gents, don't forget the valves have a max h-k rating, not always easily found in the data sheets. Some twin triodes, designed for cascade use have different values of H-K insulation for each half.

Ed

[Radio Wrangler]

Valve heaters in strings were arranged in a pecking order based on h-k rating, so that less rated ones could be used near chassis potential. Half-wave rectification allowed neutral to connect directly to chassis. This is why SMPS primary sides with bridge-rectified mains are more difficult to work on than old AC/DC sets were.

With an organ having lots of the same types of valves and all being sensitive to AF hum, the pecking-order trick isn't a goer.

David

[M0FYA Andy]

I don't understand all the concern about the need to supply lots of heater current/power.

Surely the whole point of building a valve organ is the fun of doing it with the technology of the past.

If that results in providing lots of amps at 6.3 volts that's just another challenge to be overcome!
You don't tackle this project without expecting to find a number of very hefty transformers in there somewhere.

Go for it!

Andy

[Ed_Dinning]

Hi a heater transformer with multi-filar windings will do the job nicely.

Ed