I'm documenting this project over in the Instruments forum, but since this is a general-purpose tube design question, I thought it would be appropriate here. I would appreciate any corrections or comments.
Because reasons, one of my gain stages is a cascode using two of those adorable HY615 tubes. This is my first cascode, so I followed Merlin's method. It's in the same general ballpark as his example 12AU7 (same 2.2 gm, slightly higher mu of 22 vs. 17).
Here's what I did:
There's no spec for max heater-cathode voltage, but it's less than 100V here, so I should be safe regardless. Upper tube plate voltage is at 250, safely within the 300V spec. There's no transconductance chart, so I can't calculate the gain with any accuracy, but again eyeballing the curves around my operating point, I estimate gm at 1.07, for a gain of around 72.
Because reasons, one of my gain stages is a cascode using two of those adorable HY615 tubes. This is my first cascode, so I followed Merlin's method. It's in the same general ballpark as his example 12AU7 (same 2.2 gm, slightly higher mu of 22 vs. 17).
Here's what I did:
- B+ is 400V. Had to extend the chart just a touch. I chose 100V for the screen voltage, because at his suggestion of 80V, the operating point ends up pretty far down in the weeds.
- Selected a 68k plate resistor, to get under the "knee".
- 100V screen (red line) and 68k load line (blue line) intersect at 4.4mA, so I set the operating point at 2.2mA (green dot). I eyeballed the position of the operating point at about 2/3 between -1.5V and -3V, so I'm calling it -2.5V bias.
- That implies a 1.14k cathode resistor. I rounded down to 1.1k, bypassed with a 10uF for now; will tweak this for tone shaping if necessary.
- Voltage drop across the plate resistor is 150V (2.2mA across 68k). 100V across the lower tube. That leaves a drop of 150V across the upper tube. To hit 2.2mA at 150V (green dot on yellow line) looks to require about -5.2V bias, or 94.8V on the upper tube grid.
- From a 400V B+, the closest divider I could get to 94.8V with standard resistors was 2.2M/680k, for about 94.4V.
- 2.2M and 680k in parallel are about 520k, which for a 20Hz rolloff implies a 15nF bypass cap on the upper tube grid.
There's no spec for max heater-cathode voltage, but it's less than 100V here, so I should be safe regardless. Upper tube plate voltage is at 250, safely within the 300V spec. There's no transconductance chart, so I can't calculate the gain with any accuracy, but again eyeballing the curves around my operating point, I estimate gm at 1.07, for a gain of around 72.