help with interstage needed

[schiller]

Hello to everyone, I will wind an interstage, which is going to be used to drive a transmitting triode in A2. The core is EI96b, (about 3,8 in wide, 2,9 in tall, 2 in stack). As the late Shishido states, the secondary must be wind in the oposite derection, in order to avoid saturation of the core due to the grid current of the output tube (reaches some 50ma peak).

Unfortunately, this is the smallest problem. I am aware of the difficulties associated with low ratio transformers. I have wind or instruct the "local winder" to wind several output transformers with success and experimented with 1:1 , 2:1 interstage, but with smaller cores.

Ratio will be 3:1 about 2100 turns primary. Source resistance is 1k and load 1,5-2,5k. Some vertical interlaving will be definately needed, I'm thinkig of 4 unequal sections primary all in series and 3 secondaries (series? parallel? i don't know).

This will surely cause unpredictable resonances, because of the capacitance between pri./sec. sections, in series with the inductance of the sec. windings, which cannot be viewed as "earth" as with the few turns of output transformers secondaries.

In order to brake this series resonance, i plan to put screens between the sec. and pri. layers. This will of cource increase the shunt capacitance, but a 1k source can handle 2-3nF whithout much treble attenuation.

Has anybody tried something like this, or knows of a proven good interlave pattern for such a case??

I red a post of the member "i schould coco" who uses the interstage No 10672 from Tango. I would be very interested, if he reads this, to know how the transformer measures (loaded) in order to know what i can expect.
Surely, it is not going to be as good as an output trans, that is not possible, unless i miss something.

I 'll breadboard a 211 with a 6A3 CF as driver (choke loaded), so at the moment i can avoid the interstage. But in the future, with tubes like the 250TH it will be necessary, unless I idle the CF at 120mA, using another tube (e.g. 6AS7), which unforunatelly has a low feedback factor because of the low mu. So i would like to try the option with the transformer.

If nothing of these satisfies me, I' ll use the best solution i have seen on the web: Tubelab's FET driver.
But I really want to keep it "valves only" :-)))

Thanks in advance and
greetings from Greece
Konstantinos