Great, thanks for all the thoughts and I'll definitely decrease the amount of mix I work with in any one batch... maybe I'll get the the carapils down into the 5-8% range of the total grist.
On a slightly related note (and maybe I should go look for threads on this), how much attenuation is possible with malts like Carapils or CaraMunich when mashed with a base malt? I was always under the impression they were completely unfermentable, but learned yesterday that only applies when they are steeped or mashed alone, and that the enzymes from a base malt will allow some level of conversion even in crystal malts. Is this calculable?
I don't think that is quite right. Cara/crystal malts are more or less fully converted. The process of making cara/crystal malt is to essentially mash the whole kernel and then kiln to the desired color. The sugar is not convertible because it is already converted, otherwise they would just add starch which is not sweet at all. I have read experiments that seemed to indicate that crystal malt in the lighter color range are actually fairly fermentable while the darker ones are less so. This also has a lot to do with your choice of yeast and that yeasts ability to effectively metabolize those longer chain more complex sugars.
so in short, I don't think there is a simple rule you can follow or calculation you can perform easily.
You could build out some charts for your favorite yeast through empiric method. Mini mash a bunch of 1 quart batches with a handful of pilsner malt and varying amounts of carapils, or mix up the color crystals and do one with carapils, one with c10, c20, c40... etc.
mash all at the same temp and time and pitch the same amount of the same yeast. you get the idea