Hammond organ Model M power supply

[pullshocks]

I snagged the amp out of a dead 1951 Model M.  I found the schematic. 
http://www.captain-foldback.com/Hammond_sub/schematics/M_schematic.gif

I noticed that the PT is a 350-0-350 the recto is a 5U4 but the plate voltage is only 300.  By comparison the Fender SFDR  with a 330-0-330 and a 5u4GB has 400+ on the plates

I can't find my RCA tube manual to look up the differences between 5U4 and 5U4GB  Could that account for the difference?

Or is it that the Hammond power supply circuit is obviously different from typical guitar amp PS circuits in that the PT center tap does not go straight to ground, having about 70K ohms of resistance paralleled with the speaker field coil between the center tap and ground.

My first impulse was to just use the iron with a conventional cathode biased guitar amp circuit, but I started thinking about keeping the Hammond PS and PA circuit with the fixed bias voltages and lower plate voltages.  Since I don't have the speaker or field coil, would the Hammond circuit even work?

I guess I should check that the transformers actually work before I go much farther with this, but I wanted to throw the question out there.

Thanks folks, I haven't posted much here but read the board every day and it is the most interesting forum I have found.

Mark/Pullshocks

[snoof]

I believe you can use a choke or an appropriate resistor in place of the field coil.

[RicharD]

Look at the center tap of the PT.  There is a 1k ohm field coil in parallel with a series of 3 resistors (56k, 2k2, & 12k).  Note that the center tap is also marked -115V.  If you get rid of those resistors and the field coil, you B+ is now a lovely 415VDC.

[pullshocks]

So sort of like reducing voltage with a zener diode?

[PRR]

> So sort of like reducing voltage with a zener diode?

EXactly like reducing voltage with a resistor.

This works when current demand is constant.

While the 6V6 (which draw most of the current) seem to be biased to "-20V", they are actually biased to 14,200/70,200 of the drop in the 1K field coil/resistor. Which works the same as a 202 ohm resistor, as far as the 6V6es care. It is mostly "self-bias", albeit with some non-6V6 current on top (about another 30mA).

If you do not need a field-coil, this particular amp can be re-connected for 415V at 115mA. This suggests 6.6K or 8K loading and 25W-35W output.

> I can't find my RCA tube manual

Frank's: http://www.mif.pg.gda.pl/homepages/frank/vs.html

However the two types are idential enough for our purpose. I can't find a 5U4, the 5U4G is coke-bottle shape, the 5U4GB is straight-side. If your chassis is too cramped or you use specific tube retainers, this may matter mechanically. They are all the old 5Z3 on an octal base.

[HotBluePlates]

One other point to file away in the memory banks:

Notice the output tubes appear to be fixed-bias (cathode grounded, negative voltage applied to grids). They are, but in the old days a reliable rectifier needed a socket, maybe a filament winding, lit up and got hot. An extra negative voltage supply for the fixed bias voltage was often seen as an additional cost that had to be mitigated. This Hammond has all kinds of extra costly circuits to do various things, so maybe they did indeed need to shave a little money.

The scheme used is called "back bias" and is used to create a negative d.c. supply without an extra rectifier if, as PRR said, amp current is fairly constant, and you can afford to waste some power/heat and supply voltage, while still hitting your power output target.

These days, silicon rectifiers are plenty reliable (old-school selenium was, but failed over time) adn cost a couple cents instead of tens of dollars for a tube, socket and transformer winding. If you are designing an amp, you can likely get a transformer that supplies the voltages needed to make a dedicated negative supply rather than have a very warm chassis by heating up resistors. In the old days, the items that were costly at the time dictated what appeared to be a sensible solution.