> crappy g2 B+ max. rating
Speaking generally -- why would you want a higher G2 voltage than needed to pull the maximum current you need? G2 current is just a loss. Use as little as you can.
Also: the plate knee voltage is related to G2 voltage. You know how folks wrongly say a pentode plate must sit higher than G2? Actually, and of-course, it can go lower, but it doesn't work so good when P is below some fraction of Vg2. That fraction is ill-defined, depends on detail design, but for "power" devices may be a number near "5". If you run 450V on G2, the plate won't pull much below 90V; that 90V of B+ is wasted. If Vg2 is 150V, the plate may pull down to 30V; much less waste of B+ power.
Also: screen-tube devices act like triodes except the attracter and collector are separated (G2 and Plate). How do you get more current in a triode (without adding costly cathode area)? Lower the Mu! 12AX7 passes 1-2mA. 12AU7 passes 10-20mA. 2A3 passes 100+mA (albeit with a lot more cathode).
The only annoyance is that odd G2 voltage. If we increase it (ignoring ratings) then the available cathode current is FAR more than we need, and large enough to melt something. Overdrive could smoke. You know why we can't simply resistor drop G2 voltage significantly. Extra windings and rectifiers and caps are a profit-killing expense. So we audio guys tend to run Mu=10 tubes. Then G2 can usually be held near Plate voltage. While we may waste 10s of volts at the knee, another couple dozen volts of main B+ is cheaper than windings rects and caps.
How did this work in a TV? The main power was 117V hot-chassis, 150V DC was cheap. That fed all the small-stuff and G2, and plate to get started. When H-sweeping, energy flowed through the yoke and back through a fat diode. The returned energy was caught on a cap to make B+Boost. It's not free energy, so it does not rise forever, generally levered itself up to 250V-500V. This higher B+ along with the low knee gives a wide swing of plate voltage, good efficiency (however two tubes in the path, sweep and diode), and less leverage needed to make the CRT HV.
> sensitivity is decent
If G2 voltage is "reasonable" (high enough to make your current, but not a lot higher), then grid-swing is nearly Vg2/Mu. 400V/10=40V. 150V/4= 37V. No great difference in drive voltage.
Again, TV is not audio. If we need ~~40V of clean drive, we want over 200V of B+ on the driver. The driver needs more than G2 voltage of a Mu=4 tube, however with Mu=10 the G2 voltage is ample even with added filtering. TV H-sweep drivers didn't follow a guitar string, they slammed rail to rail. Nonlinearity was unavoidable and trimmed-away (for this prescribed wave-shape).
The H-Sweep tubes do make much sense when you get farther above "stage amp" size. Over 100W, the added cost of separate G2 supply looms less large, and the plate supply savings start to mount.
The disadvantage is that H-sweep tubes need huge I and V rating, but ideally don't need any Pdiss. In linear audio, V is limited by economic OT winding, and I is often limited by Pdiss. So we want a bigger plate, not so much cathode.
The advantage is that H-sweep tubes are occasionally a glut on the market.
Agree the 6CD6 and 7867 are the "same" tube aimed for different uses. Note that for audio (where we often know our Pdiss even when distressed) the rating is higher than for TV H-Sweep (where Pdiss is low until stuff goes wrong).
Topic: 7867 PBP (Read 4856 times)