Driver transformer recommendations

[n4joy]

First, I want to thank everyone for their responses and assistance.  I've managed to fix the 130!  It was a combination of problems (some my own doing), but all is well now.  I still need to create a schematic, but that will be for another day.

1) A stock Lettine 130 is simply not capable of 100% modulation due to limitations with the speech amp and some odd component choices.
2) I changed the 6N7 speech amp to a 6SN7.  This required, of course, a complete rebuild of the speech amp section.  The 6N7 was cascaded and used grid leak bias but did not provide enough drive.  My cascaded 6SN7 also uses grid leak bias for both stages and works well; it drives the 807 driver in the modulation section just fine now.  A 6SL7 also works but starts oscillating if I turn the gain up too high.
3) The pair of 807s in the modulation section run class B and had 2.2K resistors tying the screen and grid together; these values were changed to 20K.  Every 807 class B modulator schematic I have seen uses 20K or 22K resistors.
4) The cathode resistor for the 807 driver was lowered from 750 ohms to 470 ohms; I am using a 50uf cathode resistor bypass cap.
4) The grid drive for the pair of 807 finals was way too low, which lowered my positive peaks.  I am now running about 7ma for a pair of 807s.  
5) When I replaced the driver transformer, it was wired for step up and not step down.  Class B typically requires step down.
6) I had to reverse my driver transformer wiring because of phase reversal.
7) I increased the capacitance in the LC circuit of my power supply from 7.3uf (three 450V 22uf in series) to 33uf (three 450v 100uf in series).

It seems to look good on the scope now.  Now for some on air testing!  Thank you again everyone!

Chris N4JOY

[DMOD]

4) The cathode resistor for the 807 driver was lowered from 750 ohms to 470 ohms; I am using a 50uf cathode resistor bypass cap.

What was the DC bias value across the 470 ohm resistor?

4) The grid drive for the pair of 807 finals was way too low, which lowered my positive peaks.  I am now running about 7ma for a pair of 807s.

What was the peak voltage drive to each grid?

Will we ever see a schematic of this mysterious machine?


Phil - AC0OB
 

[Opcom]

from a driver, the 'special class B' circuit evidently there will be a step up transformer because a pair of 807 with 20K grid resistors needs up to 750V grid to grid peak volts. Consider that is only +325 on the screen at peak which is the normal thing.