Hi Eli,
Appreciated your comments, and I will just mention that the article cited was written nearly 20yrs ago..
Now on to the show:
Hi Kff332,
I'm not sure that your choice of output transformer is the right one, Citation II opts ranged from 3.2K pp for the last version to about 3.8K for the early and intermediate versions. The value of both R33 and C12 will have to change to compensate. Until you achieve stability elsewhere this loop should be left open. Ultimately I suspect something like 5.1K - 5.6K will be required to restore proper loop gain margins. C12 will need to be adjusted and depending on the phase margin in the output stage may even need to be removed. The global compensation is specific to the Citation II transformer. I don't believe you will have a particular problem with LF stability in this design with this transformer. The dominant LF poles in this design are the coupling caps and resistors in the output stage, (C10, C11, R24, R27.) and they are considerably more than an order of magnitude higher than the driver stage poles even without the internal loops present.
Additionally as commented elsewhere the "spaghetti" wiring needs to go. The 12BY7 is a video pentode and has lots of bandwidth. The VERY first thing you need to do is install the grid stopper resistors right on the tube socket. They are the 33K resistor R1 in the input stage, and the 10K resistors R10 and R70.. In addition the connections to the screen grids of V2,V3 and R22 should be short and direct.
Again R23 and R28, the 10K grid stoppers on the output tubes must be right at the tube socket.
Note that in the original design most critical connections are a few inches long at most, and I can't recall anything longer than about 6" excluding bias and power supply stuff.
I've never seen a properly working Citation II oscillate into any load so it can obviously be done.
One other thing I need to URGENTLY mention is that you should not run any modern KT88 or 6550 at the design current of 100mA per tube - they just can't take it. Reduce the idle current to about 60mA -70mA depending on your plate voltage. IIRC depending on line voltage this would be around 450V - 475V.
I think that C15 at 40uF is a little small, with the stock Citation II there was a huge improvement in LF performance from just increasing this to 100uF or more. (Tighter, more extended, etc.)
Appreciated your comments, and I will just mention that the article cited was written nearly 20yrs ago..
Now on to the show:
Hi Kff332,
I'm not sure that your choice of output transformer is the right one, Citation II opts ranged from 3.2K pp for the last version to about 3.8K for the early and intermediate versions. The value of both R33 and C12 will have to change to compensate. Until you achieve stability elsewhere this loop should be left open. Ultimately I suspect something like 5.1K - 5.6K will be required to restore proper loop gain margins. C12 will need to be adjusted and depending on the phase margin in the output stage may even need to be removed. The global compensation is specific to the Citation II transformer. I don't believe you will have a particular problem with LF stability in this design with this transformer. The dominant LF poles in this design are the coupling caps and resistors in the output stage, (C10, C11, R24, R27.) and they are considerably more than an order of magnitude higher than the driver stage poles even without the internal loops present.
Additionally as commented elsewhere the "spaghetti" wiring needs to go. The 12BY7 is a video pentode and has lots of bandwidth. The VERY first thing you need to do is install the grid stopper resistors right on the tube socket. They are the 33K resistor R1 in the input stage, and the 10K resistors R10 and R70.. In addition the connections to the screen grids of V2,V3 and R22 should be short and direct.
Again R23 and R28, the 10K grid stoppers on the output tubes must be right at the tube socket.
Note that in the original design most critical connections are a few inches long at most, and I can't recall anything longer than about 6" excluding bias and power supply stuff.
I've never seen a properly working Citation II oscillate into any load so it can obviously be done.
One other thing I need to URGENTLY mention is that you should not run any modern KT88 or 6550 at the design current of 100mA per tube - they just can't take it. Reduce the idle current to about 60mA -70mA depending on your plate voltage. IIRC depending on line voltage this would be around 450V - 475V.
I think that C15 at 40uF is a little small, with the stock Citation II there was a huge improvement in LF performance from just increasing this to 100uF or more. (Tighter, more extended, etc.)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.