Dual Extruder HIPS Density Experiment

I’ve been trying to use HIPS as a dissolvable support material in my prints, but I’m not having a lot of good luck. Even at its best, the whole process takes so long an SLA printer could do a better job in less time and with less elbow grease in the post-print realm.
At first I was having Cura lay down the support material itself, but it still made for some really ugly overhangs and the rest of the time the filament (PETG) wouldn’t even stick to the supports, warping over on itself like a shriveling flower. The next step was Meshmixer with its symbiote-like organic stilts, but that had the same problems (plus a few new ones). Finally, I tried the positive/negative method:

I made the part, a small insertable key for turning a potentiometer that had to be under a panel and accessible only through a small hole in the casing.
I’m sorry I can’t show photos to go along with this since it’s work-related and the company’s pretty protective of some of its prototypes.
I made a cradle or casket for the part. I took the key, set it inside a primitive block, subtracted the key, and then combined both parts in Cura.

It seems to have worked well enough, but there’s a problem. The block turned out to be way too big and since the key had to be nearly solid in order to handle the stress of turning, the block was the same density. An entire weekend was spent soaking in a jar of limonene, and it’s still not totally free. That’s when I started thinking about a workaround.

The question I keep coming back to is: if it’s two extruders with two different materials making two distinctly different shapes, then why do they have to be the same infill density? Cura and Slic3r both let me set different extruders for different parts of the model, but there’s still only one setting for infill. Is there a reason for this, some logistical problem I’m missing?