There was a story that came out recently based on an ASHRAE study performed by David Yuill from University of Nebraska that appeared to indicate that cleaning condenser coils makes no difference on system performance and efficiency.
Those of us who have worked in the field know that coil cleaning matters because most of us have had a system that wasn’t working well, or possibly even cutting out on high head pressure. We cleaned the coil and the system started working properly… over and over again
But as an exercise… a thought experiment… let’s work through this and see some possible reasons why this conclusion may have been reached.
The job of an air cooled condensing coil is to reject heat from the refrigerant to the air. The rate at which it does this is a function of contact time, temperature differential, the thermal conductivity of the material through which heat is being transferred and turbulence of both fluids (refrigerant on the inside of the tubing and the air on the outside).
You may have noticed that modern condenser coils are larger than they used to be, the reason for this is simple, the larger the surface area of the coil, the more heat can be transferred from the refrigerant to the air resulting in a lower required condensing temperature and lower head pressure. In other words, by increasing the contact time we don’t need as great of temperature difference between the the refrigerant in the tubing and air passing over it to accomplish the same amount of heat transfer.
Engineers have also learned that by changing the design of coils we can get greater contact surface area with less refrigerant with coils such as micro-channel or they can get greater internal turbulence by adding grooves or rifling in the tubes of better external turbulence by adding little kinks to the fins of the coil. They do all of this to attempt and move heat from the refrigerant in the most efficient way possible and I applaud them for their efforts.
So how could a “dirtier” coil ever be more efficient? it is at least theoretically possible that certain types of surface fouling might act to create more air turbulence and actually increase heat transfer… and if you tested 100 systems in field conditions you may find a few that exhibit this undesigned behavior depending on the type of coil and the type of soil.
In the field we know this isn’t normal…
How many of us who do small kitchen refrigeration have gone out to a freezer not keeping temp or an ice machine making ice like it once did, only to clean the condenser and everything starts working properly again?
In our minds we imagine that the dirt or grease is acting like an insulating “blanket” preventing heat transfer, and that is certainly one factor, but it isn’t the only thing going on.
Condenser Fan Efficacy
Condenser fans are prop fans, more technically known as “axial” fans as opposed to blower wheels which are known as radial or centrifugal. Axial fans are good at moving a lot of air against very low pressure, but as soon as the pressure starts to build their performance drops off REALLY quick. We have all walked up to a condenser fan where the air was just sort of beating out of the side instead of really pushing out the top like it’s supposed to. Once you clean the coil it starts moving air again and you can really tell the difference.
So much of the decreased heat transfer comes from the fact that dirt blocks the airflow causing less air to move over the coils which drives up the condensing temperature and head pressure.
As the head pressure and condensing temperature increase the compression ratio increases (absolute head divided by absolute suction) which causes the amount of refrigerant the compressor moves to decrease resulting in both higher compressor amperage and lower system capacity. This effect is greater with TXV/EEV systems because the valve will tend to throttle down as the head pressure increases to maintain superheat further increasing the compression ratio.
On fixed metering device systems higher head pressure will also drive up suction pressure which will tend to keep the compression ratio slightly lower but will result in higher coil temperature and poor latent (humidity) control.
So to put my money where my mouth is we picked a nice dirty coil and ran a full, white paper style test. For the sake of complete disclosure we used the fan curve charts to come up with evaporator air flow, which is fine because it was a before / after test. I used MeasureQuick for the calculations and my phone was giving me trouble and kept losing my manually entered data so I realized later that in my AFTER report (that some of you may have seen in my group) the airflow was set to 750 and before was set to 700 so I went back in and changed the math so everything was apples to apples. Either way… the results are pretty self evident. You will notice that the “official” results below are slightly different than those in the screenshots at the top, and that math change is the reason.
2-ton 1999 Trane R22 10 SEER “Spine Fin” Heat Pump Split system with a direct return operating and 0.4” WC total external static pressure on a PSC blower and a fixed piston type metering device.
I Allowed the system to run 20 minutes continuously and took detailed measurements sufficient to compare wattage, total BTU/H removal and therefore the EER of the system using wireless connected digital instruments and the MeasureQuick app.
We cleaned the condenser coil only while performing this test no other cleaning or servicing and making no adjustments to refrigerant charge.
We then allowed the system to run continuously for another 20 minutes to ensure the coil is completely dry while confirming by measuring condenser air dew point entering and leaving. Retake the same measurements and compare the results.
The before results showed clearly that the head pressure and liquid line temperature were both high with a low subcooling and superheat. The measured system performance was poor even though the evaporator coil, air filter and blower wheel were quite clean considering the age of the system.
After cleaning the head pressure and suction pressure dropped, the subcooling and superheat increased and the compressor amperage dropped. It became clear after the cleaning that the system was slightly low on refrigerant because it maintained a stable 31° superheat.
The system performed significantly better in terms of decreased wattage and increased BTU removal after the cleaning.
|Suction Pressure / Evaporator Temp||75.9||67.9|
|Liquid Pressure / Condensing Temp||278.6||216.5|
|Outdoor Air DB||89.0||91.0|
|Total BTU Capacity||19,372||20,992|
After this test was complete we added 9 oz of R-22 to achieve the factory required superheat. Following the adjustment the EER and total system capacity improved even further.
This illustrates that cleaning this condenser indisputably improved –
I wrote to David asking him to come on the podcast and explain his findings a few moths ago and he responded to that via email with this –
“At some point I’d like to set everybody straight in one fell swoop, and maybe your HVACrSchool is the venue for that, but I haven’t decided yet.”
I don’t think David’s research is “wrong”, I’m sure they got the results they said they got, the issue must be a disconnect in the way the tests were performed and the way many systems perform in the field. I do think the conclusion the article came to was incorrect…
no… I KNOW IT IS INCORRECT
The point of the study was all about heat transfer and in real life if we control for ambient conditions all we would need to do it measure head pressure, clean the coil, let it dry and measure head pressure again. If it goes down then more heat transfer is occurring (again, controlling for changes in ambient conditions and indoor load).
For fun, I would encourage you to try the same tests and let me know your findings. I used MeasureQuick, a Redfish meter and Fieldpiece Joblink probes to collect the data but you could do it with any accurate modern digital instruments. Just make 100% sure the coil is dry after cleaning or you will get false measurements. If you find a system that doesn’t improve, or gets worse it would be great to know the “why” behind that example by reviewing the application and data.
If you want to come to your own conclusions as to why the research came to the findings it did… the test apparatus is shown below.
This is the peer reviewed article
Image shown under Creative Commons from –
Mehdi Mehrabi & David Yuill (2019) Fouling and Its Effects on Air-cooled Condensers in Split System Air Conditioners (RP-1705), Science and Technology for the Built Environment, 25:6, 784-793, DOI: 10.1080/23744731.2019.1605197
P.S. – Out full report on coil cleaning will be available on Speedclean.com in a few weeks so keep your eyes peeled.