I find it peculiar that people use bulbs for heating in this application. I'm not aware of any particular effect visible light has on wax, but it is certainly difficult to achieve even temperatures across high aspect ratio items using a few point sources of substantially radiant heat. For one thing, the center of the
Minimizing any thermal gradients or local hot spotting is difficult enough with forced convection. I've built a couple boxes in the last year and I wouldn't trust valuable skis to one without investing in several temperature measuring devices. A half dozen thermocouples and a noncontact thermometer is cheap compared to a fast pair of DH skis.
I think the appropriate use of lightbulbs in hotboxes is for admiring your work and seeing what you are doing when you load and unload them.
For the original question: Unless you need some localized source of warm air in your shop, I can't see any benefit in putting a hole in your oven. The energy you put into the box isn't wasted during winter in that it ends up as heat in the surrounding environment in any case. In summer it is waste. If you do need local warm air, you could build such a dual use device, but it will make the design more complex and expensive. You will need a lot
more heat with any useful flow rate out of the box.
My opinions, still in development as I build, test, use these devices:
-Too much fan never hurt anyone. Too little fan is easy to achieve. Keep pressure ratio in mind. There really isn't anything at Home Depot designed to do this job...but don't let that stop you from trying.
-Any internal electrical/electronic devices including fans need to be rated for the conditions they will be subject to. I experimentally determined that electronic fluorescent lamp ballasts last for approximately 15 minutes at sixty degrees Celsius.
-Windows are nice.
-Controlling temperature accurately is easy. Controlling ramp rate is hard, especially if you want to be able to load varied numbers and sizes (masses) of skis. The thin layer measurement techniques needed to validate the importance of ramp rates (and ultimately the entire endeavor) are way out of my wheelhouse...can anyone help there?