Not my brightest moment to be sure…
Start with a cabinet full of electronic equipment. Each piece of equipment puts out a certain amount of heat. When the heat production balances the heat loss of the container (in this case, the cabinet) you have the operating temperature of the cabinet. It’s a lot like a falling object; think of acceleration due to gravity as analogous to heat production, and air resistance as to heat loss; eventually they balance and you have terminal velocity, or operating temperature respectively.
If the operating temperature of the cabinet is higher than, say, eighty degrees
, you get increased equipment failures. Manufacturers allow higher temperatures than that in their manuals but in my experience that’s about where you start seeing increased costs. Since we have 28 of these cabinets, such costs can mount up. Now imagine you are me, and a little bit slow on the draw at times…
The rate of heat loss in the cabinet is a function of airflow and cabinet wall conduction. You can’t do much about the latter but you can cut holes and install fans to improve the former. So you take baseline measurements, and build a prototyping vent panel. Experiment with different arrangements of fans and grilles to suppliment the cabinet’s built-in exhaust fan. Record your results and when you hit the target temperature, develop your plan and order parts.
But it turns out you have trouble obtaining (at an acceptable price) the Nidec fans you prototyped with, so you try a different fan, one with slightly higher noise rating but easily available. Too noisy. You pore over fan specifications, trying to find the right combination of power, flow, noise level, price, and availability.
In all of this you’ve learned more about fans and airflow than you ever thought possible. You’ve read white papers from fan manufacturers and struggled with the math of obstruction and pressure.
Then you think; “Doh! Static pressure!” The exhaust fans you’ve tested compete with the one already in the cabinet, creating a negative static pressure, or partial vacuum. After all the reading you’ve done, you now realize this has a huge effect on the CFM, or ‘cubic feet per minute’, that a fan can move.
The original fan couldn’t do its job because of static pressure, and the prototype arrangement you rigged up only worked because it was powerful enough to be a brute force solution. You’ve overlooked the obvious. Could delivering free airflow in the right place – where the components themselves exhaust their heat – allow the existing cabinet fan to move more air and hit the target temperature?
You remove your prototyping fans and reinstall the panel with only the fan grilles in place. In essence, you’ve invented “the hole”.
It works. The resulting operating temperature was 1.5 degrees cooler than any other arrangement you’d tested.
Sometimes finding the right solution can make you feel like a complete idiot.