I've been thinking about how this is all going to work without heat storage. Despite our reputation here it gets cold at night and the unconditioned uninsulated part of the house has already gotten down to 40F degrees in the middle of the night, while the outside temperature is 30F. And that part of the house rarely gets to 60F at any time. So what seems to be happening is that there is about a 20 degree variation in the uninsulated part of the house. On warmer days it goes from about 45F to a high of 65F.
I'm going to be insulating it with dense packed cellulose due to SF and Daox's good advice. It will be equivalent to about 2x6 framing, but without most of the thermal bridging that goes on there. I put 2x2 staggered against the existing 2x4s. If I can get that temperature variation down to 6 degrees F from warm to cold without any heating input during an 18 hour period then this will work. I really do not have the experience to know if the level of insulation I'm putting in will accomplish that. In this climate my house will be hugely better insulated than my neighbors, but nowhere near to Passivhaus standards. If anyone has direct experience with temperature variations in an unconditioned house with my level of insulation (R60 in ceiling, R30 in floor, and spray foam to R10 on the rim joists) I'd be thankful.
In the meantime I'm planning on a 6 degree variation. Here's a picture of the hopeful, but not neccessarily accurate predicted outcome.
Right now on average to good solar days the roof temperature reaches 75F at noon and does not dip below it until 6PM. On those days it is over 90F in the attic for 2 or 3 hours. On cold days it doesn't do as well. For instance today it only reached 84F and the outside high temp was 57F. The coldest day so far. (I know, its not cold) So that's what I'm basing it on. Obviously this is a crude estimate but for now I think it's in the ballpark. I feel very fortunate that I have the attic temperature that I have to work with.