Ideally should a heat pump run continuously?
![](https://secure.gravatar.com/avatar/a74f30dfb98458ee9da05eab97910d60?s=52&d=https%3A%2F%2Fimages.greenbuildingadvisor.com%2Fapp%2Favatars%2F200x200%2Fr.png&r=g)
I have seen this alluded to a few times on GBA, but never straight out stated. If cycling is the enemy of modern inverter heat pumps (mini split or ducted), does that mean in an ideal world we want our heat pumps to run continuously?
In other words if a heat pump has an output range of say 35% – 100%, and the heating load is always within that range, should it exactly match the load and just never switch off?
The “old school” rule of thumb was that you always want a certain number of cycles per hour, and if you aren’t achieving this then there is something wrong with the system (airflow issues, undersized, oversized).
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb1.png)
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb2.png)
Replies
Short cycling is what you want to avoid. In theory every cycle start uses extra energy, but beyond a certain cycle length it starts to become insignificant.
I haven't done the math, but I think it's very unlikely that you can find a heat pump that will throttle back enough to cover the entire heating load. Will the minimum demand in April be at least 35% of the high demand in January? Definitely not where I live.
Re: first question: At some point, cycling becomes more energy efficient and provides better dehumidification than operating continuously.
>"...if a heat pump has an output range of say 35% – 100%, and the heating load is always within that range..."
>"At some point, cycling becomes more energy efficient and provides better dehumidification than operating continuously."
In heating mode a heat pump doesn't provide ANY dehumidification.
A 3:1 turn down ratio would be on the smaller-ratio end for single zone mini-splits, but pretty common for multi-splits or modulating full-size air handler heat pumps. With higher static pressure higher volume air handlers the air handler powers starts to eat up the efficiency when modulating to ultra-low levels. There are several single zone mini-splits with 10:1 turn down ratios that are still pretty efficient across that range though, and can run nearly continuously for most of the heating season.
Would you mind elaborating on this? Since you are talking about dehumidification I assume you mean cooling rather than heating. Why would cycling be more efficient and better at dehumidifying than continuous operation?
Are you assuming that a variable speed system operating at low capacities will not do a good job dehumidifying b/c at lower operating capacities the coil does not get cold enough to condense moisture from the air?
That's certainly true of my Mitsubishi FH09 mini split. But my Trane XV18 (traditional ducted unit) removes a good amount of moisture even when operating at 40% capacity (I think the SHR is something like 0.8 even in low capacity stages).
As I said, "at some point". I usually see fan CFM that doesn't decrease in proportion to a wide range compressor. For example, a 4:1 range compressor with a 2:1 fan. So too much airflow and poor dehumidification if it runs continuously at low load. Specs on this issue seems to be rare (anyone have a SHR vs output graph?). If fan CFM did decrease all the way down, then there would be a throw/mixing/stratification problem.
Re thermal efficiency, degradation coefficient varies, but might be 15% (which isn't bad).
There is a crossover point where the output of the unit drops to a level where it isn't justified to keep it running. For example there has been some discussion here on whether the LG cold climate wall mounts might be more efficient if they didn't modulate so low.
Ideally the space has enough thermal and latent inertia/mass to keep it from short cycling at minimum capacity. When the unit shuts off, the space will take a while to drift off setpoint enough to justify a restart. The inertia is different than heat loss/gain.
Different units behave differently when starting, some load up substantially and then slowly unload, while others are slower to assume the load. In a tight well insulated space with adequate thermal mass I think slower loading is ideal. (I think Fujitsu has a setting for highly insulated spaces to slow it's loading ramp).
There is probably a max cycles per hour that would be ideal (I'm guessing 2-3 max).
>"For example there has been some discussion here on whether the LG cold climate wall mounts might be more efficient if they didn't modulate so low."
Actually the COLD climate LGs retain high efficiency at low modulation- it's their nearly identical part number cousins that fall flat. eg:
https://ashp.neep.org/#!/product/25816 (COP=5 @ +47F, 1023 BTU/hr)
https://ashp.neep.org/#!/product/25836 (COP=1.5 @ +47F, 1023 BTU/hr)
(deleted b/c responded to wrong message)
At what time does a cycle become too short? Is 3 cycles/hr OK? Does equipment longevity depend on number of cycles, or just run time?
At some complexity level of HVAC, they solve this by putting in a buffer tank; Instead of spending 6 minutes out of every 30 minutes running the compressor at full-blast for a given set of conditions, you spend 60 minutes out of every 300 minutes running the compressor full-blast.
Extra credit to anyone who demonstrates a bedroom design that has acceptable efficiency and temperature/humidity fluctuation with a mini-split that can only modulate down to 3K btu/hr in a room with a 1K btu/hr design load.