I am working my way through ACCA's latest manual LLH (Low Load Homes). The postulate of continuous fan increases indoor humidity in the cooling season arises in Section 8. This principle has been shouted from the roof tops of many building science venues for years. And up until this point, I have embraced that Kool-Aid with the masses. 

The subject matter is discussed in an Allison Bailes article found here from 2012. 

Apparently, I am a slow learner. Because, as I think about the physics, it isn't intuitive to me. Perhaps someone can clarify it, possibly Dr Bailes.

Assume a heat pump air handler with duct system that is entirely in the conditioned space. Cooling is energized. Coil gets cold and falls below dew point of the return inlet air. Humidity in the air changes from vapor to liquid. Condensation flows down the fins of the coil into the drain pan and hopefully is drained out of the unit to the exterior. Compressor shuts off, leaving X amount of liquid water on the coil. 

This is where I stumble. If the blower shuts off with the compressor, the condensation resident on the coil just sits there and eventually evaporates or dries to the ambient air in the air handler. And since the duct work is connected to the conditioned space (technically just another unspecified room of the house), that evaporated moisture eventually becomes part of the air in the living space. Hence it will elevate the indoor RH. 

Second scenario is the blower runs continuously, which does have a lot of desirable benefits. The air dries the coil quicker. Consequently, you might see a spike in the indoor humidity levels. 

But in the end, I don't see the difference. One introduces additional humidity over time. The other in less time. 

One possible difference is the unit energizes before the coil gets completely dry. That would tend to imply the coil stay predominately wet and there is a consistent reservoir of condensation that simply stays put much like a hydrophilic buffer.  Even if this is so, it needs to stop to inhibit microbiological and biofilm production. 

Any comments to straighten me out????

You need to be a member of Building Performance Community to add comments!

Join Building Performance Community

Email me when people reply –


  • Hi folks

    Well this conversation is well above my knowledge base but just a couple of observations. Danny started this out as a discussion regarding low load homes and so any conversation about what might occur with duct leakage most likely wouldn't apply as ducts would certainly be inside. Also, my understanding is that modern high performance heat pumps (which you would see in a low load home) do keep the air handler running for a brief period after the compressor cycles out so as to harvest the last amount of heating or cooling that remains in the coil. I am not sure what mini-splits do in respect to this but assume that it would be the same.

  • There are two issues that need to be considered in your "thought experiement".

    Evaporator coils use gravity to drain the condensate off the coil (duh). Many if not most AC systems blower oppose the gravity force and try to "blow" the condensate up the coil. When the blower is on but the compressor is off, I suspect much of the water that remains on the coil will be held in place on the coil until it evaporates or the compressor cycles on again. Only when the compressor is running does the condensate droplets get large enough to run down the coil to the drain pan. I read somewhere for a typical residential system it takes about 15 minutes of compressor operation before the first drop of condensate actually leaves the coil and drain pan. If the compressor cycles off in the first 15 minutes there is effectively zero water removal. I suspect that when the blower cycles off with the compressor, some water drips off the coil and makes its way out of the house/plenum/ducts at the end of a less than 15 minute run time.

    The other point is the *ALL* duct leak. With leak testing, we might get that leakage as low as 5 to 10%. (My experience is that most tested and 'approved' ducts leak more like 20 to 25% of the air flow.) Unless your ducts are in conditioned space or you are in a climate where humidity control is not a concern, all of the duct leakage is bringing outdoor humidity into the house. If the blower runs continuously that water is introduced 24/7. If the blower cycled with the compressor, there are fewer hours of water introduction.

  • Danny - you are correct. An evaporator with no airflow over it will dry out slowly. Keeping the airmover on and some air velocity over the evaporator increases the liquid water evaporation rate exponentially.

    The inside of the evaporator plenum becomes cool and very humid with the airmover turned off and the diffusion of the water vapor out away from (and heat energy into) that plenum is pretty slow so the rate of evaporation from the wet coil is very slow. Turning the airmover on and moving air over the wet coil moves the evaporated water vapor out of the coil plenum space rapidly providing a constant supply of "fresh" drier air eager to evaporate more liquid water from the coil. That supply of "fresh" air also brings energy in the form of heat which is transferred to the wet coil to provide more energy for evaporation.

    I believe most A/C system evaporators running at part load do not dry out between many of the cooling cycles if the system fan is shut off with the compressor (or does not overrun too long after) while systems with the fan left running after the compressor shuts off will completely dry the evaporator on many more off cycles. So there is some nuance here - a long enough off cycle will dry the evaporator either way, but the fan left on will dry the evaporator in a matter of minutes and the compressor will need to "rewet" the evaporator every cooling cycle before any liquid water drips into the drain pan and is removed from the space greatly increasing the effective sensible heat ratio of a system in cyclic operation.

    • As I have read through everyone's excelent responses, it has caused me to reassess my assumptions. Tim O'Brien made a observation that is rightly relevant when he provided,

      "I believe most A/C system evaporators running at part load do not dry out between many of the cooling cycles if the system fan is shut off with the compressor (or does not overrun too long after) while systems with the fan left running after the compressor shuts off will completely dry the evaporator on many more off cycles " 

      This is an excellent reminder and may be a bigger issue than I thought. Absent an initiator to increase molecular activity, H20 wil be completely content to reside on the coil between cycles. So my premise that the coil will eventually dry isn't as cut and "dried" as it seems. 

      However, when the blower stays on, the return air increases the water's activity, which is no longer content to stay in liquid form. 

      Another factor that came to me from some of David B's comments is the blower speed has a huge influence on drying after an off cycle. If the mfr allows high speed air on fan only, the coil warms faster and dumps more water vapor into the air from high bypass factor. If the mfr programs the blower to run at 50% or 30%, the coil stays cool longer and less vaporization takes place from low bypass factor.

      Now the question is, if the compressor must cycle, is it better to dry the coil or leave it wet during off cycles. It would appear benefical for moisture drainage to allow the coil to retain this buffer to enhance water drain from the fins. BUT there is a risk of creating biofilm and microbial activity. Might even see some Stachy in that pan. 

  • All cooling coils which remove moisture (which is most of them!) require a drain pan below them to capture this. So much is known and ideally this would be either stainless steel or an approved thermoplastic or ABS for corrosion protection. 

    The missing link in your post is how that moisture is drained away.  It cannot, as you suggest, simply re-evaporate; that would defeat the purpose. 

    Enter the drain line- not just a line leaving the unit and hoping for the best, but one with a P-trap seal. 

     Consider that the suction side of the fan, where most cooling coil drain pans would be, is under a negative pressure. That fan is sucking air in and with sufficient pressure to keep water in place.  Only when the fan is off (assuming you have this drain connection piped at all), will the water drain. 

    So what does this P-trap do?  

    The P-trap has a certain depth and allows water to fill it. This forms a seal to atmospheric pressure. 

    The difference in elevation between the drain pan low-point and the out-flowing drain pipe on the other side of the trap, is deeper than the static pressure at the fan suction.  In my institutional work, this can be four to six inches. In a house, two inches is usually sufficient but it depends on the system pressure above all.

    It seems counter-intuitive, but this trap seal helps the water drain.  It does this by allowing the water enough height to stack up and develop sufficient "head" to force/allow the water to drain out of the downstream port. 


    • @Bradford, Danny is referring to residual moisture that remains adhered to the coil after the compressor cycles off. Due to adhesion, a meaningful amount of bulk water will remain on the coil rather than dripping off into the catch pan. Eventually, that moisture will evaporate back into the air.

      • Bradford wrote:

        >I do not see that amount of moisture on that size of coil as a problem, let alone a 3 SF residential coil with limited outside air.

        Probably not. But to Danny's point, a lot of folks in residential building science & home performance warn that continuous fan can contribute to indoor humidity problems, especially in the extra-humid Southeast. The tiny unitary mechanical systems (often mini-splits, which are already latent constrained) being installed in super low load homes sometimes have trouble keeping up with latent loads, typically associated with ventilation (or over-ventilation, in my opinion). Also, the latent loads associated with occupancy begin to loom large when you ratchet down the sensible load. You can manage all of that with a built-up system but we're talking about homes with sensible design loads as low as 3 BTUH/ft2!

        • That is why I like VRF systems, David.  We agree especially in the humid south and of course with cycling in general of on-off single-capacity systems.

          Inverter drives are the current ideal and I use them whenever I can on larger systems (which are lagging in that technology vs. residences).

          Any system that can use dehumidification priority, ramping down fan speed and modulating refrigerant, as has been done for 20 years or more, is essential. 

          I do not see anything wrong with continuous fan use though. Absent fan heat added, it is an enthalpy-neutral process. It is the refrigerant side that has the work to do.

          But out of this discussion, I think we can agree that the off-cycling of the compressor after the sensible load is satisfied, is much more of an issue than a residual film of moisture on an evaporator whether or not the fan continues to run.



          • Bradford - I have seen single family homes with ducted central AC or heat pump hold enough water upon shutdown that the living space relative humidity increased >10% when the fan was left on. I'm sure there is some diversity among the equipment and installations out there, but some residential systems do hold back enough water to make a more than measurable impact on the conditioned space relative humidity.

            I have not seen similar behavior on the few mini-split systems I have monitored. They seem to operate with a high SHR when cooling and do not remove much water vapor. I assume they are trying to maximize sensible cooling efficiency by avoiding latent removal. So if you have a system which runs with a high SHR cycling with the fan on is probably not a big deal, but if you have a system set up to run cold and lower SHR they can hold enough water into an off cycle to cause humidity problems if the fan is left on.

      • Got it and understood thank you. I was figuring if there was a problem, the water was not leaving the unit.

        That amount of moisture though is negligible though, at least in large, high-latent air handling units.

        For example, a 10,000 cfm 100% OA air handler taking air from 91/74 down to 52 saturated (19.4 delta enthalpy and 451.8 MBH latent heat) will remove about 430 lbs of moisture per hour at that rate. 

        That is 7.2 lbs. per minute or a bit under a gallon per minute.  0.12 lbs. moisture removal per second.

        The residence time on the coil at 500 fpm and a 10-row coil (12 inches deep) and 20 SF of coil face area is less than 1.5 seconds. 

        The total amount of moisture on that surface area at any given time is about 0.18 lbs. 0.022 gallons.  About a third of an ounce. Spilling the residue of your last ice cube. 

        I do not see that amount of moisture on that size of coil as a problem, let alone a 3 SF residential coil with limited outside air.

This reply was deleted.