Control Grid Dissipation - Triodes v Pentodes

[eschertron]

Pardon if my searches missed relevant info.

I'm thinking about class AB2, and looking up material on drivers for the output stage. Many designs and discussions out there, but the info I lack pertains specifically to the difference between triode construction and pentode construction. Some tube data sheets list grid current v grid/anode voltage, but others do not. I found at Tung-Sol a graph for the 12AU7 that showed control grid voltages at up to +30 volts, and over 100mA grid current!

http://www.tungsol.com/specs/12au7-tung-sol.pdf

This got me thinking: Do triodes as a family have beefier control grids because there are no screens, or is this an example of a specialty-use tube, or has my ignorance once again trumped my ability to reason this out? The published control grid current in the document above seems pretty whack in the light of the care needed to protect 'those fragile grids' from over-dissipation.

[Old Tele man]

Don't forget that Tungsol data sheet is referring to "pulsed" operation (<15% of scan cycle), not to continuous conduction as AB2 is...and power dissipation is a time-dependent function: shorter time allows higher currents or voltages(sometimes).

[eschertron]

Don't forget that Tungsol data sheet is referring to "pulsed" operation (<15% of scan cycle), not to continuous conduction as AB2 is...and power dissipation is a time-dependent function: shorter time allows higher currents or voltages(sometimes).

Don't forget? How can I forget something I never knew?
Is that spec something I missed on the sheet, or something a RF engineer knows instinctively?

edit: I see the answer is "both". Would a sine wave qualify if the grid drive is positive WRT the cathode 15% or less of each cycle?