Hi Guys,

It seems to me that if the plate to cathode DC voltage of the
preceeding grounded cathode triode matches the DC grid voltage of the
CPI then you can directly couple them. True?

Is it possible to raise a 2nd grounded cathode stage high enough above
ground to direct couple two grounded cathode stages together? (anything
to remove a cap and fun learning exercise)

For the willing: I delight in highly educational expounded replies.

Thanks all. I get by with a little help from my RATS

Wessel

Wessel Dirksen wrote:
Hi Guys,

It seems to me that if the plate to cathode DC voltage of the
preceeding grounded cathode triode matches the DC grid voltage of the
CPI then you can directly couple them. True?

True, clever, and a sonic boon. But you don't need "highly educational
expounded replies"; once you grasp the principle Colonel Ohm's
discipline is all you require to work it out. Here's an implemented
example that I used for a couple of years. It is a simplification of a
more highly developed circuit suggested to me by the excellent Anthony
Mills of the Australian Navy. You might want to draw it:

Set up one 417A and one SV572-10. The current requirement is calculated
from the desired operating parameters. Let us say that from the curves,
plus calculation by slew rate method and for Miller, we have decided to
operate the 417A at 160V and 18ma and -2V; the SV572-10 will operate at
Eb 740V Eg -40V and 80mA. We decide arbitrarily on a 20K grid resistor
and with the -40V grid drop that gives us 2mA. Total thus 100ma. The
voltage drop between plate and cathode of the SV572-10 must be the
operating voltage of the 417A plus the negative grid bias of the
SV572-10, that is 160 +40V or 200V. The power supply must thus be high
enough to deliver 940V at 100mA to the plate. (In real life, we usually
start with a fixed voltage from the power supply and then fiddle the
numbers until they match. It is just easier to understand the thought
process if we pretend we run all the numbers regardless and then order
up a suitable power supply for next-day delivery. If only...) Clearly
the power tube must be able to handle the voltage and current; the
SV572-10 is rated 1000V and 140mA, so it will not break sweat.

As I describe what happens, make arrows for the currents to indicate
direction and write down the mA beside the arrows.There will be three
current allocations and they must add up to the correctly.

We require a 2K5 resistor to sink 80mA at 200V (NB!) from the power
tube cathode to ground. This 2K5 cathode resistor may be bypassed in
the normal manner. From the junction of the power tube cathode and the
cathode resistor, a line is taken towards the driver tube, the 417A. It
carries 20mA. First it meets a potentiometer in series, which will be
adjusted so that the voltage drop between the SV572-10 cathode and the
417A plate is 40V, same as the power tube grid negative bias. Then it
meets a suitable audio choke in the region of a 100H or a silicon
pseudochoke (made with say MJE350 and 2N2907; this is a separate
subject we can discuss if you need help) in series and thence to the
plate of the 417A. From the junction of the pot and the choke it helps
to have a decoupling cap to ground (All the decoupling caps for B+ and
cathodes must be adjusted to have the same time constant, okay?) From
the junction of the 417A plate and the choke a line is taken to the
grid of the SV572-10: it is just a piece of wire, zero cap on it, the
end result of our exercise: straight wire with gain! The cathode
resistor of the 417A is sized at 110 ohm to drop the 18mA we already
decided on and the remaining 2mA is conducted to ground via the grid
resistor of 20K between the 417A and the SV572-10. So the current to
ground under the cathode of the power tube, the current led through the
plate and cathode of the driver tube, and the current to ground via the
grid leak add up to the 100mA we put into the plate of the SV572-10.
Similarly, the voltages add up. We dropped 200V between the plate and
cathode of the power tube, of which 40 went to negative bias and 160 to
the plate of the driver.

You might want first to experiment with less rare and less expensive
tubes and (much) lower voltages...

Is it possible to raise a 2nd grounded cathode stage high enough
above
ground to direct couple two grounded cathode stages together?
(anything
to remove a cap and fun learning exercise)

You mean do this trick between cathode and previous plate twice?
Theoretically possible, I suppose, but components wander. I have never
managed for instance to make a White amp work right for long. John
Broskie on his TubeCAD site has a useful article on fixing the classic
direct coupled stages. The one I explain above appears to be too simple
ever to have been much used in classical times!

For the willing: I delight in highly educational expounded replies.

Thanks all. I get by with a little help from my RATS

Wessel

HTH.

Andre Jute