I don't believe there are accurate models of grid current. Certainly
Koren's isn't (although it's a reasonable stab at it). The definitive
text on grid current is Spangenburg, which is hard to find.
There's something in my article at www.john-a-harper.com/tubes201
which is condensed from Spangenburg.

You will LOTS of current - I drive the 6AS7 output stage of my
Atma-Sphere M60s from 6BX7s and they run out of gas to drive
grid current very quickly. You would need another high current
tube (6AS7, 6C33C...!) to get into significant positive grid
voltage. Honestly I would not pursue this idea. Use more tubes...
I can let you have 6C33Cs at a very good price (PLEASE somebody
buy the wretched things!).

John

"David R Brooks" wrote in message
...
Hi RATs,
I am planning a 20W OTL using two 6C33C's.
Several people have published designs using 4 or 6 6C33C's, getting up
to 100W (into 8R). So 20W from two tubes seems achievable.

It is a design aim not to stress the tubes excessively.

Now, to get 20W into 8R, peak current will be about 2.2A. To get this
with a 6C33C requires the grid be driven well positive (Va being about
80V at this point).
This will draw significant grid current (ie Class AB2).

Q1: Is grid current permitted for 6C33Cs? (I have seen some discussion
to the effect that the grid must not be driven positive.)

Q2: Assuming grid current is allowable (on the peaks, anyway), is
there a Spice model published which models the grid current
accurately? Alternatively, has anyone plotted the grid current, and
could share their results?
(I have ordered my tubes, but they haven't yet arrived, so I can't
plot Ig myself).

So far, I have found two Spice models for this tube, due to:
1. Norman Koren
http://www.normankoren.com/Audio
2. Francesco Pintavalle
http://www.geocities.com/CaperCanave...iode-data.html

Both show grid current flowing (as would be expected), but it isn't
clear whether this is accurately modelled.

TIA,

Hi David ,

If you just want to use 2 tubes/channel why not Single Ended Parallel (SEP)
or Push Pull ?
I have plans to build a 6C33 SE (one tube/channel) and I'm aiming for 11W .
(15W is possible) .
But if you still want to go for the OTL with just 2 tubes take a look at
this :
http://laplaza.org/%7Efvuotto/mustage.html
I think a EL83 is a nice penthode for you in this type of driver .

Ronald .


"David R Brooks" schreef in bericht
...
Hi RATs,
I am planning a 20W OTL using two 6C33C's.
Several people have published designs using 4 or 6 6C33C's, getting up
to 100W (into 8R). So 20W from two tubes seems achievable.

It is a design aim not to stress the tubes excessively.

Now, to get 20W into 8R, peak current will be about 2.2A. To get this
with a 6C33C requires the grid be driven well positive (Va being about
80V at this point).
This will draw significant grid current (ie Class AB2).

Q1: Is grid current permitted for 6C33Cs? (I have seen some discussion
to the effect that the grid must not be driven positive.)

Q2: Assuming grid current is allowable (on the peaks, anyway), is
there a Spice model published which models the grid current
accurately? Alternatively, has anyone plotted the grid current, and
could share their results?
(I have ordered my tubes, but they haven't yet arrived, so I can't
plot Ig myself).

So far, I have found two Spice models for this tube, due to:
1. Norman Koren
http://www.normankoren.com/Audio
2. Francesco Pintavalle
http://www.geocities.com/CaperCanave...iode-data.html

Both show grid current flowing (as would be expected), but it isn't
clear whether this is accurately modelled.

TIA,

My vision is : If you build with tubes , go build with tubes .
So I skipped reading the SS part of the story .
Using a penthod as a C.C.S. works great if you use the right penthodes .
Even with an EF184 you can get a Z-out of about 1K .
Imagine what happens with a much beefier EL83 .....

"David R Brooks" schreef in bericht
...
Hi Ronald,
Yes, I have settled on trying an OTL design.
Looking at the page you cite, I see proposed the use of MOSFETs as
constant-current devices. But given they are to be constant-current
(ie high impedance) loads, why use the source-follower as in Fig 5?
Surely a P-channel MOSFET, used as a "traditional" current source at
the B+ rail, would be simpler?

"Ronald" wrote:

:Hi David ,
:
:If you just want to use 2 tubes/channel why not Single Ended Parallel
(SEP)
r Push Pull ?
:I have plans to build a 6C33 SE (one tube/channel) and I'm aiming for 11W
..
15W is possible) .
:But if you still want to go for the OTL with just 2 tubes take a look at
:this :
:http://laplaza.org/%7Efvuotto/mustage.html
:I think a EL83 is a nice penthode for you in this type of driver .
:
:Ronald .
:
:
:"David R Brooks" schreef in bericht
.. .
: Hi RATs,
: I am planning a 20W OTL using two 6C33C's.
: Several people have published designs using 4 or 6 6C33C's, getting up
: to 100W (into 8R). So 20W from two tubes seems achievable.
:
[snip]

Won't that be 20W into 16 or 24R? Only "achievable" if you have 16
or 24R speakers. You can't just scale willy-nilly. 20W into 8R is
a daft idea, I'm afraid. You may have a fundamental problem in your
understanding.

Until you know what your circuit is, you can't know how far +ve you
need to drive the grid.

May I suggest a quick way of estimating drive requirements without a
spice grid model. Find a sensible circuit for a *pair* of valves,
if that is what you are going to use. Find out how far you need to
drive the grids positive. Find the grid *resistance* at that point.
Ensure that the driver circuit has an output resistance of one tenth
of that grid resistance, at whatever amplitude is necessary. The
one-tenth rule-of-thumb seems to me no less applicable just because
the grid is positive.

Neither simulation nor this method will show whether the valves will
survive whatever you do to them.

cheers, Ian

"David R Brooks" wrote in message
...
Hi RATs,
I am planning a 20W OTL using two 6C33C's.
Several people have published designs using 4 or 6 6C33C's,
getting up
to 100W (into 8R). So 20W from two tubes seems achievable.

It is a design aim not to stress the tubes excessively.

Now, to get 20W into 8R, peak current will be about 2.2A. To get
this
with a 6C33C requires the grid be driven well positive (Va being
about
80V at this point).
This will draw significant grid current (ie Class AB2).

Q1: Is grid current permitted for 6C33Cs? (I have seen some
discussion
to the effect that the grid must not be driven positive.)

Q2: Assuming grid current is allowable (on the peaks, anyway), is
there a Spice model published which models the grid current
accurately? Alternatively, has anyone plotted the grid current,
and
could share their results?
(I have ordered my tubes, but they haven't yet arrived, so I
can't
plot Ig myself).

So far, I have found two Spice models for this tube, due to:
1. Norman Koren
http://www.normankoren.com/Audio
2. Francesco Pintavalle
http://www.geocities.com/CaperCanave...iode-data.html

Both show grid current flowing (as would be expected), but it
isn't
clear whether this is accurately modelled.

TIA,

There are several examples of tubed OTL circuits (some with 6C33)
at http://www.tubebuilder.com/schematic3.html

In general, I've found as a rule of thumb tubes that have
a current gain of about 10 when G1 is in conduction.
Done that in both audio amps & DC Regulated power supplies.

Good Luck John Stewart

David R Brooks wrote:

Hi RATs,
I am planning a 20W OTL using two 6C33C's.
Several people have published designs using 4 or 6 6C33C's, getting up
to 100W (into 8R). So 20W from two tubes seems achievable.

It is a design aim not to stress the tubes excessively.

Now, to get 20W into 8R, peak current will be about 2.2A. To get this
with a 6C33C requires the grid be driven well positive (Va being about
80V at this point).
This will draw significant grid current (ie Class AB2).

Q1: Is grid current permitted for 6C33Cs? (I have seen some discussion
to the effect that the grid must not be driven positive.)

Q2: Assuming grid current is allowable (on the peaks, anyway), is
there a Spice model published which models the grid current
accurately? Alternatively, has anyone plotted the grid current, and
could share their results?
(I have ordered my tubes, but they haven't yet arrived, so I can't
plot Ig myself).

So far, I have found two Spice models for this tube, due to:
1. Norman Koren
http://www.normankoren.com/Audio
2. Francesco Pintavalle
http://www.geocities.com/CaperCanave...iode-data.html

Both show grid current flowing (as would be expected), but it isn't
clear whether this is accurately modelled.

TIA,