Hot Temperature and High Stakes Performance [pdf] | Hacker News Reader

Hot Temperature and High Stakes Performance [pdf]

www3.nd.edu

56 pointsrobg10 comments

Something I’ve noted about working from home is that I can simply turn up the air conditioning if needed to keep myself at what feels like an optimum temperature. I’ve heard of this effect of heat on decreased performance and thought it a sign of bad management leadership when I work at a customer site and it feels like a sauna. Especially in the IT field, they might be throwing away millions of dollars worth of lost productivity annually because they can’t be bothered adjusting a thermostat.

Where are you that has office buildings that are too warm?

Everywhere I've ever worked I bring a sweatshirt to the office because it's always freezing. And usually summer is even colder than winter!

I often see offices get over-filled with staff and the A/C is not upgraded to match. E.g.: a shift from cubicles to hot-desking with an open floor plan can increase the density of people in the space substantially.

>a shift from cubicles to hot-desking

quite literally

Have you ever looked at how A/C systems are designed? The engineers determine the parameters on a “design day” by calculating the “sensible load” (heat entering the space from outside + sunlight + people + equipment, etc) and the “latent load” (rate at which water vapor enters the air from ventilation, leakage, people, kitchens, showers, etc, times the heat of vaporization of water to get it in units of heat), and then design an air conditioner that will remove heat and vapor, in the corresponding amounts and ratio, at steady state at the indoor temperature and humidity they choose, under the design conditions. Then they throw a thermostat in so the temperature is under closed-loop control. Of course, this means that the humidity is under open-loop control based on the wild guess they made — YOLO! And there is absolutely no effort made, even if the calculation is right, to get the system to converge to design conditions quickly.

Add people to an office with lots of windows and sunlight coming through those windows, and the sensible to latent load ratio changes, and there goes any semblance of humidity control. If you’re in a hot, dry climate, it may be impossible to get the parameters right under any conditions with ordinary air conditioners.

Never mind that the installer probably messes up, resulting in leaky ducts, and the sign of the resulting massive error isn’t even well determined if you can’t predict which duct leaks the most.

I’m aware of exactly one vendor of A/C equipment that will attempt to adjust the coil temperature to control humidity based on actual measurements. Mitsubishi has a “dry” mode that is very much not closed-loop and is barely documented. Good luck.

At least in humid climates, it’s possible to install separate systems for cooling and dehumidification.

Congratulate your management. I once worked in a newly-built, multi-million dollar facility with inadequate and faulty air conditioning that would go from barely below sweating temperature to physically dangerous when it failed for at least a week every summer.

I worked in an office building where the AC was on one circuit, but somehow the difference between the warm side of the building and the freezing side of the building was over 10F, which was about 30 feet away.

I don’t think it’s the case here since the subjects would have been unaware of the observation but temperature to human performance metrics usually leads to a classic example of the Hawthorne effect where you’ll see an improvement in performance no matter which way you change the temperature simply because the participants are aware of the experiment.

https://en.m.wikipedia.org/wiki/Hawthorne_effect

I think the US Air Force has some interesting studies on the effect of temperature on performance of pilots.

Could you link to some references if you're aware of them?

Reply on news.ycombinator.com