Not necessarily disputing because I am not familiar with that alloy(just to clarify), but considering that a Q & T material generally relies upon the formation of martensite in the quench process (in other words extremely rapid cooling) which then is tempered to form some carbides, removing carbon from solid solution and relieving some stresses(and therby improving toughness), I don't see right away how slowing down the cooling rate with an increase in preheat(possibly changing the transformation from austenite to martensite/bainite to one of austentie to ferrite), can be beneficial to toughness measured as impact strength. I could see it with lateral expansion but not impact strength.
The other thing is I didn't notice whether we were talking about weld metal charpies or HAZ charpies. The Q & T microsturcture of the base metal will be destroyed by the welding process in HAZ, rendering it essentially a normalized microstructure. And the weld metal response would be entirely independent.
I think Giovanni's thinking is sound. I believe that in most as welded Q & T applications some reduction in toughness from that which the base metal is capable is expected(in the HAZ). Weld metal will be based upon its chemsitry. The question is how much?
The 200C range is certainly higher than the Q & T preheat examples provided in FJ Winsors article in ASM Volume 6.
The general idea is to have preheat high enough to prevent cracking but low enough to maintain as rapid a cooling rate as possible or you lose your martensite, and may even overtemper portions of the HAZ.
And a 10deg interpass range. My advice would that anyu range that narrow chould remain an ideal target and not written into the procedures because its all but impossible to apply on the shop floor.