Hey lpfink and Bob,
I'm jumping in here because I am chasing down answers for similar questions, but I can offer a bit of experience I have dealing with product design using non-arcing/non-sparking components - these are non-arcing/non-sparking during normal operation, but they would pose a problem in the event of a failure in a Division 1 location. From my experience, this is OKAY in Division 2 because you aren't assumed to have an explosive atmosphere at all times, whereas in Division 1 you are assumed to always have said atmosphere. In the Division 1 case, the "first failure" is of the non-arcing component and causes a hazard because the explosive atmosphere is assumed. In the Division 2 case, the "first failure" would be the presence of the explosive atmosphere to begin with, hence the general use of non-arcing/non-sparking stuff being sufficient as a protection method. This would be different if the component were say Intrinsically Safe, because in that case even a component failure would not cause a hazard.
So for your C1/D1 scenario, you would have to put the non-arcing/non-sparking bits into a suitable enclosure for the Division 1 space, and then meet the Division 1 sealing requirements based on your specific installation (trade size, distances, etc.).
Your reference to the line in 501.15(B)(2) would apply if you were Class 1 Division 2. From what I know, there are specific pack & pour style sealing fittings for entering boxes in Division 1 locations and/or leaving Division 1 locations with a conduit to an unclassified space. My read of (B)(2) would be that you don't need these (large, bulky) specifically listed explosion proof sealing fittings for leaving a Division 2 location and instead you can use some kind of fitting normally used to prevent the transmission of vapor - the point there is to prevent your unclassified location from becoming classified (because in the event of a failure, aka flammable atmosphere, in the Division 2 location it would just migrate therefore making your unclassified location actually Division 2). This is how we handled it on many power gen systems we had running off natural gas, propane, hydrogen, etc. The idea is to create some kind of vapor boundary to keep the unclassified locations as such, but not to the level of needing to contain an explosion as in the case of a Division 1 space.
This is from my experience - I've been lurking on these forums for a while and never had occasion to post, but from my reading over some time I would say if Bob says different then I would go with his advice and not mine. I'm still a little wet behind the ears at 10 years in my electromechanical design seat
I have some additional questions about the requirements of the explosion proof sealing fittings in C1/D1 spaces, for which I will post another thread (when my newbie permissions go through...)