I am creating a custom box that will contain 3 * 400 watt metal halide bulbs (1200 watt total). The box will remain closed for 60 minutes, open for the next 60 minutes, then closed, then open and so on.. for a period of 10 hours. The box will open/close from the the front panel (the "door") using Arduino and Servo motors.

Since the bulbs will be generating a lot of heat I will attach a Duct Fan at 240 CFM on top of the box (is the top the best place to put it?). Something like this: http://www.amazon.com/dp/B005KMTYFK/ref=wl_it_dp_o_pC_nS_ttl?_encoding=UTF8&colid=V2E1GXE4DR4J&coliid=I3B1JSME5M15UN

The box itself will be made out of wood so I will coat it with a heat resistant spray from the inside so it doesn't catch on fire. Something like this: http://www.amazon.com/dp/B000CPJLV2/ref=wl_it_dp_o_pC_nS_ttl?_encoding=UTF8&colid=V2E1GXE4DR4J&coliid=I122VPINWI1GBY&psc=1

When the box is open I will need the box to emit as much light as possible. Therefore, I will coat the inside with a reflective substance (something like this: http://www.amazon.com/Rust-Oleum-1915830-Metallic-Silver-11-Ounce/dp/B000PQGE2M/ref=pd_sim_hi_3?ie=UTF8&refRID=0DKBGQGVVCT0YSWCW1WB) so all light rays are dispersed from the box..

What do you think?


  • 1
    \$\begingroup\$ Heat-resistant paint simply means that the paint itself won't be affected by high temperatures. It doesn't confer any protection to whatever it's painted on. \$\endgroup\$
    – Dave Tweed
    Commented Mar 13, 2015 at 11:05
  • \$\begingroup\$ I was always bad with multi-medium heat transfer math but this fixture needs to be made of sheet metal. That I'm pretty sure of. \$\endgroup\$
    – user39962
    Commented Mar 13, 2015 at 17:39

3 Answers 3


1 - Location of the fan is irrelevant, since the airflow is driven, rather than occurring due to convection. Actually, putting the fan under the box rather than on top is best, so that the fan never gets heated by the hot air it is exhausting. At any rate, you want free air flow through both top and bottom. Note that this directly competes with your desire to reflect as much light as possible. Whether 240 CFM is adequate I don't know, but I'd suspect you'll be OK. No guarantees, though.

2 - Depending on a spray-on paint to fireproof the box is a really bad idea.

3 - Putting a probably-flammable layer of reflecting paint on top of the first layer is an invitation for the reflecting layer to catch fire. Your problem is that there is no obvious way to determine the reflectivity of the reflective paint in the infrared, and that is where most of the energy is emitted by the bulbs. Metallic silver by itself has excellent reflectivity into the IR, but that does not necessarily imply that your paint does as well.

4 - If you really think you'll be present for 10 hours continuously, well, all I can say is that I admire your bladder control. Not to mention your willingness to go hungry.

5 - As I recall, you're trying to illuminate a 3 x 5 foot area. If you want a fairly uniform illumination level over the entire area, you will need to make a tradeoff on box size. The closer you put your light source to your test area, the bigger the box (or at least the transmitting aperture) has to be. With a larger separation you can use a smaller box, but then you lose intensity. Also (and this is important), the bigger the box the harder it is to control exactly where the air goes, and the harder it is to guarantee that there are no dead zones/hot spots.

6 - In line with 1 and 5, just exactly how big a box were you figuring to make?

  • \$\begingroup\$ The box will be 18 inches by 12 inches by 12 inches. You mentioned 240 CFM is "OK" - if I use a 400 CFM does this minimize (or eliminate risk)? About item 4 - if I go hungry (& thirsty) there's nothing to admire about bladder control anymore.. \$\endgroup\$ Commented Apr 13, 2015 at 21:49

I'd be concerned about the wooden box cathing fire (irrespective of the coating you put on the wood) so maybe I'd add a protection circuit based around a thermistor that tripped a relay thus cutting energy to the lights should things get too hot.

Basically, safety is my main concern.

  • \$\begingroup\$ Why are you worried about fire with a Duct Fan at 240 CFM and coating? Can you be more specific? I mean what is the reason based on? \$\endgroup\$ Commented Mar 13, 2015 at 9:15
  • \$\begingroup\$ Why do people put fuses in houses when the cable is perfectly fine for taking the load of a kettle and oven? Simple answer is removal of excessive risk. What if the duct fan stopped working i.e. it failed to spin because of some electrical fault in its hardware. C'mon dude think about risk - it's important. \$\endgroup\$
    – Andy aka
    Commented Mar 13, 2015 at 9:20
  • \$\begingroup\$ @ what temperature should I trip the circuit? In other words, how "hot" is too hot? \$\endgroup\$ Commented Apr 13, 2015 at 21:46
  • \$\begingroup\$ I really don't know. You need a materials engineer to guide you on this. \$\endgroup\$
    – Andy aka
    Commented Apr 13, 2015 at 21:52

If all of the heat of the lamps was being transferred directly to the air, moving 240 CFM through the box would result in approximately a 9°C temperature rise.

The problem is that the air itself will absorb very little of the lamp power directly; instead, most of it will be absorbed first by the box, and then subsequently transferred to the air. The temperature rise of the box will be dependent on the thermal resistance between it and the air, which will be fairly high unless you take special measures (think heat-sink fins) to control it.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.