We have found that glue strips (mouse glue boards cut into small pieces) installed above the opening is a very effective, although not perfect, way to monitor use of den boxes by fishers and martens as they will pretty much always leave a few guard hairs on the strips when entering or leaving the box. We have not seen a lot of hairs from squirrels, and squirrels or other critters occasionally pull them out. But they work very well and require very minimal time spent at boxes to install and then check in early summer or fall once the denning season is well past. I would be happy to mail you a handful of them -- the glue boards you can bet in Walmart are impregnated with peanut scent and squirrels always chew those off. We have others that seem to work better. If you use these and get hair samples, I would be happy to analyze the samples to tell you what species they belong to.
One other option that folks in the UK have used for well over a decade is to simply visit the box once a year (late spring, early summer, or fall) to look for scats in/on the box and to inspect for a depression in the bedding material (note: we use untreated chipped bark or wood chips for bedding, but pet bedding can work, too). We have started to see obvious depressions when used by flying squirrels, martens, or fishers, and the diameter of the depressions indicate who was using it. Upon completing an annual visit, one can either add a bit more bedding or fluff up the existing bedding to remove the depressions so that you can observe future use at the next annual check.
Indeed, both options can be used together.
We tend to point our boxes south to avoid getting back-lit pictures that are hard to analyze. You wouldn't have to point them south, but I would recommend pointing them any direction other than west/NW so that the prevailing wind doesn't cause the boxes to be too cold. A south/southeasterly orientation could help with this. But I will note that fishers, which do not create the cavities they use, have been documented using cavities pointing pretty much any direction. I am guessing most are south/southeastern but we didn't actually measure this on the natural cavities we found fishers using, so I can't say for sure.
This is something I don't have great data on at the moment, other than the fact that we did occasionally find natural fisher dens in cavities that were in the same stand and somewhat close. I don't think there is a minimum spacing, and it may be that having more cavities in a stand made the stand higher quality to fishers. But it also depends on what caused the fishers to switch dens when they were rearing kits (note: fishers use on average 3 dens to raise a single litter; they don't typically use the same den in subsequent years, but have been observed doing so). If they flee a cavity after a predator shows up, they will not likely want to use a different cavity in the same stand that year, but if the stand is well-positioned in their territory, they may use the same or a different cavity in that stand in subsequent years. We also think one reason they use more than one cavity is because of kits defecating in the cavity and nest parasites (beetles) getting at high density if they always use the same cavity. In this case, having another den opportunity nearby would be great so that they only have to move the kits 10s of meters rather than 100s or 1000s of meters.
In the longer-term, I would be interested in setting up a phase 2 of this study to work with citizens to help determine how fishers respond to different box/cavity densities.
Cedar would be a good material, and cedar is pretty resistant to rot and decay. We used plywood to mimic the boxes used in the British Columbia project and due to price and dimensions, but solid wood would certainly withstand the weather much better in Minnesota. In fact, we have some boxes at the Cloquet Forestry Center that high school students at Carlton HS made that have plywood interior, foam insulation, and solid-wood exteriors. All this to say, I think the cedar 2x6's could work just fine. I would encourage you to maintain an internal dimension of about 11-12" square length/width and a height of about 32-36 inches as this appears to be the dimensions of a typical fisher cavity in aspen. If it gets smaller, they may not have room. If the box were larger, it might be harder for them to keep warm with their body heat.
We have done some modeling that suggests the insulation is important early in the denning season as the temperature within the box more closely approximates that within natural cavities we have monitored. But we also know that fisher kits have been documented in wood duck nest boxes in Minnesota and elsewhere so it is not absolutely necessary. If you have the materials, two-layers of 1.5" thick boards would get pretty close to the insulation provided by natural cavities. We used exterior construction adhesive to seal joints/cracks, but we have also seen plenty of natural cavities with cracks, so the box being airtight is not an absolute necessity.