My tentative conclusion from looking at these results is that they’re inconclusive because neither metric captures what consumers care about? As others have pointed out, the ability to cool down a room more than you need isn’t necessarily impressive.
More relevant may be something like
days in year in which the one-hose model isn’t good enough
amount of time you have to leave it on
level of noise it makes
energy consumption
(Note that I’m not saying there was an easy way to measure any of these.)
You could read the post’s results as suggesting that a two-hose model performs better according to those metrics, but it’s not trivially true.
I propose a follow-up experiment to measure daily energy consumption alternating hose configuration with the same set temperature. The previous experiment tried to answer “how much does maximum cooling power change between configurations,” while here we would answer “how much does efficiency changes between configurations.”
Potential issues:
If this causes your unit to run at different power levels, you would also capture any efficiency change based on the power level, but I would guess your unit regulates simply as on/off (check the (instantaneous) power consumption with the Kill-a-Watt to be sure).
If one configuration works faster, it may not do as much work on the other side of the room before the unit senses “cool,” and turns off. Fans in the room increasing circulation will mitigate this.
I don’t know how many days you would need for good statistics to smooth over all the day-to-day environmental changes.
You should also analyze weekend vs. weekday, and possibly exclude one.
You will need to monitor that both configurations actually are up to the task.
I’m not sure I want to register an advance prediction, but if OP agrees to do this, I will at least put some thought in towards one.
My tentative conclusion from looking at these results is that they’re inconclusive because neither metric captures what consumers care about? As others have pointed out, the ability to cool down a room more than you need isn’t necessarily impressive.
More relevant may be something like
days in year in which the one-hose model isn’t good enough
amount of time you have to leave it on
level of noise it makes
energy consumption
(Note that I’m not saying there was an easy way to measure any of these.)
You could read the post’s results as suggesting that a two-hose model performs better according to those metrics, but it’s not trivially true.
Measuring energy consumption is cheap and easy with a $30 Kill-a-Watt: https://www.amazon.com/P3-P4400-Electricity-Usage-Monitor/dp/B00009MDBU/ref=sr_1_3?crid=2F9E51CGW3ZDS&keywords=kill+a+watt&qid=1656014636&sprefix=kill+a+watt%2Caps%2C107&sr=8-3
I propose a follow-up experiment to measure daily energy consumption alternating hose configuration with the same set temperature. The previous experiment tried to answer “how much does maximum cooling power change between configurations,” while here we would answer “how much does efficiency changes between configurations.”
Potential issues:
If this causes your unit to run at different power levels, you would also capture any efficiency change based on the power level, but I would guess your unit regulates simply as on/off (check the (instantaneous) power consumption with the Kill-a-Watt to be sure).
If one configuration works faster, it may not do as much work on the other side of the room before the unit senses “cool,” and turns off. Fans in the room increasing circulation will mitigate this.
I don’t know how many days you would need for good statistics to smooth over all the day-to-day environmental changes.
You should also analyze weekend vs. weekday, and possibly exclude one.
You will need to monitor that both configurations actually are up to the task.
I’m not sure I want to register an advance prediction, but if OP agrees to do this, I will at least put some thought in towards one.