Skip to main content

Low refrigerant charge and the search for real data


By Richard Hawkins MACS contributor


For many years, I had read about how low refrigerant charges could increase compressor operating temperatures and decrease the amount of oil flowing to compressors, but I had never been able to find results from any tests.  After discussing this with MACS, we decided to do some testing to get some real data.  The event was a Temperature Testing Symposium held at the MACS headquarters. 

Objectives

In addition to obtaining some definitive information about compressor operating temperatures and oil flow to the compressor, there was other information of prime interest. The following is a list of the test objectives:

  • Compare the system high side and low side pressures running in a full charge condition verses an undercharged condition.
  • Monitor the ambient temperature (temperature inside the shop in this case) and humidity (inside the shop in this case).
  • Compare center vent temperatures running in a full charge condition verses an undercharged condition.
  • Compare evaporator inlet and outlet temperatures running in a full charge condition verses an undercharged condition.
  • Compare condenser inlet and outlet temperatures running in a full charge condition verses an undercharged condition.
  • Compare accumulator inlet and outlet temperature running in a full charge condition verses an undercharged condition.
  • Compare compressor operating temperatures running in a full charge condition verses an undercharged condition.
  • Compare the amount of oil in the compressor after being run in a full charge condition verses an undercharged condition.

    Equipment

To accomplish these test objectives, a manifold gauge set, and four contact type temperature testing thermometers were utilized.  The contact type temperature testing thermometers were connected at the following points:

  • Inlet and outlet of the evaporator.
  • Inlet and outlet of the condenser.
  • Inlet and outlet of the accumulator.
  • Compressor case

In addition, one of the temperature testing thermometers had a probe which was inserted into the center vent inside the vehicle and another in front of the vehicle.  This unit had the capability of providing the vent temperature, ambient temperature (temperature inside the shop) and humidity (inside the shop), by just pushing a button.

Here is the test equipment set up.  The unit labeled “accumulator” also had the capability of providing vent temperature, ambient temperature (temperature inside the shop) and humidity (humidity inside the shop).

The subject vehicle was a 1998 Jeep Cherokee with a 4.0 6cylinder engine and an orifice tube system.  It had a 20-ounce refrigerant capacity and 8.1-ounce system oil capacity.  It had been used for some other testing which resulted in the condenser being replaced.  Also, the orifice tube and accumulator were replaced, the system was flushed, and all of the oil was carefully drained from the compressor.  Then 8.1 ounces of oil were added to the system and after being re-assembled and having a vacuum pulled, it was charged with 20 ounces of refrigerant utilizing a J-2788 R/R/R machine which of course has an accuracy of .5 ounces.

The next order of business was to run the A/C for an extended period of time (50 minutes) to ensure the oil was distributed throughout the system and get the compressor to maximum operating temperature, then take readings and record them.

The following system settings were used:  Doors closed and windows up.  High blower speed.  System was run in the recirculation mode.  The temperature knob was turned up slightly to provide additional heat load to prevent system cycling.

The readings were as follows: (See pictures 2 through 5)

Ambient temperature (temperature inside the shop).  89.0°  

Vent temperature.  47.4°

Humidity (inside the shop).  22.4 %

The ambient temperature (temperature inside the shop) and center vent temperature are displayed on the screen on the left side.  The humidity is displayed on the screen on the right side.

System pressures.  High Side: 187 PSI   Low Side: 23 PSI

Evaporator inlet and outlet temperatures.  Inlet: 37.0°    Outlet:  36.8° (.2° differential)

The system pressures are displayed on the left side.  The evaporator temperature readings are displayed on the right side.  

Condenser inlet and outlet temperatures.  Inlet: 146.8 degrees     Outlet:  108.6 ° (38.2° differential)

Accumulator inlet and outlet temperatures.  Inlet: 39.3°     Outlet: 39.3° (0° differential)

The condenser inlet and outlet temperatures are displayed on the left side.  The accumulator inlet and outlet temperatures are displayed on the right side.

Compressor case temperature.    125°

The refrigerant was then recovered from the system, the compressor was removed, and the oil was drained from the compressor and measured.  Quantity of oil drained from the compressor:  1.75 ounces.

The compressor case temperature is displayed on the left side.  The quaintly of oil drained from the compressor (1.75 ounces) is displayed on the right side.

Next week we will cover the results of the same tests running the system with a 25% undercharge.

What do you think of this test? Have you seen similar results? Let us know.

2 responses to “Low refrigerant charge and the search for real data”

  1. William Arneson says:

    At no time was it discussed that the ambient temperature must be at least 75degrees. The molecule structure of the refrigerant is changed during lower ambient temperatures. This causes major reading changes and overcharging when trying to determine why your gauges are reading low.

  2. Thomas Lech says:

    @ Williams Arnedon you are correct trying to run test at low ambient temperatures on refrigerant system you will get drastically different results and depending on what kind of fan control system constant or cycling over the condenser airflow is king and will also greatly skew the results and the data.

    I myself live in San Francisco where is the majority of the year below 70° you can basically throw anything you read in books into the garbage can when it comes to trying to determine anything by pressure.

    As for compressor case and compressor discharge line temperatures having that high superheat and lack of oil return these were all things that were talked back in high school auto class at least by my instructor. As your compression ratio gets more extreme the compressor works harder and becomes less efficient.

    As a kid before high school my father used to have me take apart neighbors and fellow church petitioners members air conditioning systems apart before recharging as a teaching tool for me as a child.
    This was my father’s message to teach me to grasp the concept of oil flow and where it was located and how much in each component when it would come in very low on charge.
    My father would have me hook up a four channel Simpson test meter with thermal couplers . Attached thermal couplers to different lines on the compressor and lines to take the Delta across each component.
    Then filling the systems completely up with a proper charge by weight running the system and then having me vary slowly bleed off the R 12 to the atmosphere, slowly was the secret to not cooling refrigerant flow backwards and flash the refrigerant with oil moving it out of the components in lines that it was sitting in for this test. (you can tell how long ago this was. R12 was only $.39 a can)

    Then my father would have me take everything apart again and pour out oil and Measure to see the before and after difference.

    In back in those days with fixed displacement compressor on a hot day you can actually come very close to setting a charge using sub cooling on expansion valve systems and later when office tube systems were introduced superheat.

    Something that I would never even attempt to try now with a variable displacement Compressor, Computer controlled vehicles and variable fan speed or cycling fans over the condenser especially on cool ambient temperature days.

    Things were so simple back then especially before variable displacement compressors were introduced.

    But you get a really good insight on Automotive when you take four years of my classes in residential air-conditioning and commercial refrigeration/. And then throw another major on top of that building automation which on compared to vehicles would be considered your body control module integrated with your power control module integrated with your HVAC control model and with your battery and charging and load shedding.
    Automotive and commercial and industrial HVAC have a lot of things in common

    The one thing Automotive doesn’t have yet but getting better is the training and education

Leave a Reply

Your email address will not be published. Required fields are marked *