Yeah it's pretty much always an unmanned environment. No women either. When you're cooling a house, the surface area facing the outside and the number of door & window penetrations is a big deal, but POP sites are usually in a space the size of a walk-in closet with one door. So I'm pretty sure that the heat generated in the room is what I need to worry about, but I'm not sure how well a resistance heater translates to electronics. I'm also fuzzy on power consumed by fans. Kinetic energy and heat are really the same thing, right? Right?
Maybe when I'm rich, bored, and retired I'll put servers inside of a water jacket and measure their actual heat output. -Adam -----Original Message----- From: AF <[email protected]> On Behalf Of D. Bernardi Sent: Monday, June 26, 2023 5:03 PM To: AnimalFarm Microwave Users Group <[email protected]> Subject: Re: [AFMUG] Heat estimate Not directly related but when I researched cooling for a new datacenter the key was efficient airflow (or even hot isle/cold isle), and setting the thermost higher not lower. Although that saves engery and $$$ you have a much shorter time to recover from a cooling failure as system temperature will raise rapidly with little margin. There's an interesting study where Intel ran tests for nearly a year with servers in an air cooler environment in New Mexico with outside temperatures approaching 100F. I'll see if I can find it. From there, finding the right size cooling requirements is still a challenge (easy to over do it) but it helped. Cooling for resistive heat is different than for human comfort so if it is an unmanned environment, there's that to consider as well. At 02:50 PM 6/26/2023, you wrote: >Content-Type: multipart/alternative; > boundary="----=_NextPart_000_0262_01D9A83D.942822B0" >Content-Language: en-us > >I've been assuming that whatever energy you put into an electronic >system is coming out as heat -there might be multiple conversions >before it becomes heat, but it must become heat eventually. > >A resistive heater is said to convert 1 Watt to >3.41 BTU/hr. So that's the conversion I'm using. > >I'm largely ignoring other factors because I'm assuming the >overwhelming majority of heat in the room is the heat from my >equipment, and the goal is to pump that heat outside. > > > >I don't think I've undersized an air conditioner >yet using that methodology, but is it overkill? >-- >AF mailing list >[email protected] >http://af.afmug.com/mailman/listinfo/af_af.afmug.com -- AF mailing list [email protected] http://af.afmug.com/mailman/listinfo/af_af.afmug.com -- AF mailing list [email protected] http://af.afmug.com/mailman/listinfo/af_af.afmug.com
