CS650 30 cm Soil Water Content Reflectometer
More accurate in soils with high bulk electrical conductivity
weather applications water applications energy applications gas flux & turbulence applications infrastructure applications soil applications


The CS650 is a multiparameter smart sensor that uses innovative techniques to monitor soil volumetric water content, bulk electrical conductivity, and temperature. It outputs an SDI-12 signal that many of our dataloggers can measure.

Note: The cable termination options for this sensor are not suitable for use with an ET107 station. For this type of station, use the CS650-LC sensor instead, which has a suitable cable connector.

Read More

Benefits and Features

  • More accurate water content measurements in soils with bulk EC up to 3 dS m-1 without performing a soil-specific calibration
  • Larger sample volume reduces error
  • Measurement corrected for effects of soil texture and electrical conductivity
  • Estimates soil-water content for a wide range of mineral soils
  • Versatile sensor—measures dielectric permittivity, bulk electrical conductivity (EC), and soil temperature


Detailed Description

The CS650 consists of two 30-cm-long stainless steel rods connected to a printed circuit board. The circuit board is encapsulated in epoxy and a shielded cable is attached to the circuit board for datalogger connection.

The CS650 measures propagation time, signal attenuation, and temperature. Dielectric permittivity, volumetric water content, and bulk electrical conductivity are then derived from these raw values.

Measured signal attenuation is used to correct for the loss effect on reflection detection and thus propagation time measurement. This loss-effect correction allows accurate water content measurements in soils with bulk EC ≤3 dS m-1 without performing a soil specific calibration.

Soil bulk electrical conductivity is also calculated from the attenuation measurement. A thermistor in thermal contact with a probe rod near the epoxy surface measures temperature. Horizontal installation of the sensor provides accurate soil temperature measurement at the same depth as the water content. Temperature measurement in other orientations will be that of the region near the rod entrance into the epoxy body.



Measurements Made Soil electrical conductivity (EC), relative dielectric permittivity, volumetric water content, soil temperature
Required Equipment Measurement system
Soil Suitability Long rods with large sensing volume (> 6 L) are suitable for soils with low to moderate electrical conductivity.
Rods Not replaceable
Sensors Not interchangeable
Sensing Volume 7800 cm3 (~7.5 cm radius around each probe rod and 4.5 cm beyond the end of the rods)
Electromagnetic CE compliant
Meets EN61326 requirements for protection against electrostatic discharge and surge.
Operating Temperature Range -50° to +70°C
Sensor Output SDI-12; serial RS-232
Warm-up Time 3 s
Measurement Time 3 ms to measure; 600 ms to complete SDI-12 command
Power Supply Requirements 6 to 18 Vdc (Must be able to supply 45 mA @ 12 Vdc.)
Maximum Cable Length 610 m (2000 ft) combined length for up to 25 sensors connected to the same datalogger control port
Rod Spacing 32 mm (1.3 in.)
Ingress Protection Rating IP68
Rod Diameter 3.2 mm (0.13 in.)
Rod Length 300 mm (11.8 in.)
Probe Head Dimensions 85 x 63 x 18 mm (3.3 x 2.5 x 0.7 in.)
Cable Weight 35 g per m (0.38 oz per ft)
Probe Weight 280 g (9.9 oz) without cable

Current Drain

Active (3 ms)
  • 45 mA typical (@ 12 Vdc)
  • 80 mA (@ 6 Vdc)
  • 35 mA (@ 18 Vdc)
Quiescent 135 µA typical (@ 12 Vdc)

Electrical Conductivity

Range for Solution EC 0 to 3 dS/m
Range for Bulk EC 0 to 3 dS/m
Accuracy ±(5% of reading + 0.05 dS/m)
Precision 0.5% of BEC

Relative Dielectric Permittivity

Range 1 to 81
  • ±(2% of reading + 0.6) from 1 to 40 for solution EC ≤ 3 dS/m
  • ±1.4 (from 40 to 81 for solution EC ≤1 dS/m)
Precision < 0.02

Volumetric Water Content

Range 0 to 100% (with M4 command)
Water Content Accuracy
  • ±1% (with soil-specific calibration)
  • ±3% (typical with factory VWC model) where solution EC < 3 dS/m
Precision < 0.05%

Soil Temperature

Range -50° to +70°C
Resolution 0.001°C
  • ±0.1°C (for typical soil temperatures [0 to 40°C] when probe body is buried in soil)
  • ±0.5°C (for full temperature range)
Precision ±0.02°C


Note: The following shows notable compatibility information. It is not a comprehensive list of all compatible or incompatible products.


Product Compatible Note
21X (retired)
CR10 (retired)
CR10X (retired)
CR200X (retired)
CR206X (retired)
CR211X (retired)
CR216X (retired)
CR23X (retired)
CR295X (retired)
CR500 (retired)
CR5000 (retired)
CR510 (retired)
CR9000 (retired)

Additional Compatibility Information

RF Considerations

External RF Sources

External RF sources can affect the probe’s operation. Therefore, the probe should be located away from significant sources of RF such as ac power lines and motors.

Interprobe Interference

Multiple CS650 sensors can be installed within 4 inches of each other when using the standard datalogger SDI-12 “M” command. The SDI-12 “M” command allows only one probe to be enabled at a time.

Installation Tool

The CS650G makes inserting soil-water sensors easier in dense or rocky soils. This tool can be hammered into the soil with force that might damage the sensor if the CS650G were not used. It makes pilot holes into which the rods of the sensors can then be inserted.


CS650 / CS655 Firmware v.2 (429 KB) 12-02-2015

Current CS650 and CS655 firmware. 

Note:  The Device Configuration Utility and A200 Sensor-to-PC Interface are required to upload the included firmware to the sensor.

View Update History

Frequently Asked Questions

Number of FAQs related to CS650: 54

Expand AllCollapse All

  1. The permittivity of saturated sediments in a stream bed is expected to read somewhere between 25 and 42, while the permittivity of water is close to 80. A CS650 or CS655 installed in saturated sediments could be used to monitor sediment erosion. If the permittivity continuously increases beyond the initial saturated reading, this is an indication that sediment around the sensor rods has eroded and been replaced with water. A calibration could be performed that relates permittivity to the depth of the rods still in the sediment. 

  2. No. The principle that makes these sensors work is that liquid water has a dielectric permittivity of close to 80, while soil solid particles have a dielectric permittivity of approximately 3 to 6. Gasoline and other hydrocarbons have dielectric permittivities in the same range as soil particles, which essentially make them invisible to the CS650 and the CS655. 

  3. No. The equation used to determine volumetric water content in the firmware for the CS650 and the CS655 is the Topp et al. (1980) equation, which works for a wide range of mineral soils but not for organic soils. In organic soils, the standard equations in the firmware will overestimate water content.

    When using a CS650 or a CS655 in organic soil, it is best to perform a soil-specific calibration. For details on performing a soil-specific calibration, refer to “The Water Content Reflectometer Method for Measuring Volumetric Water Content” section in the CS650/CS655 manual. A linear or quadratic equation that relates period average to volumetric water content will work well.

  4. The bulk electrical conductivity (EC) measurement is made along the sensor rods, and it is an average reading of EC over that distance at whatever depth the rods are placed.

  5. The electrical conductivity (EC) of sea water is approximately 48 dS/m. The CS650 can measure permittivity in water with EC between 0 and 3 dS/m. EC readings become extremely unstable at conductivities higher than 3 dS/m and are reported as NAN or 9999999. Because EC is part of the permittivity equation, an EC reading of NAN leads to a permittivity reading of NAN as well. Thus, the CS650 cannot provide good readings in sea water.

    With regard to sea ice, the electrical conductivity drops significantly when sea water freezes and the permittivity changes from approximately 88 down to approximately 4, as the water changes from a liquid to a solid state. With both EC and permittivity falling to levels that are within the CS650 measurement range, the sensor is expected to give valid readings in sea ice. The sensor is rugged and can withstand the cold temperatures. However, as the ice melts, there will be a point at which the electrical conductivity becomes too high to acquire a valid reading for either permittivity or electrical conductivity.

  6. No. The abrupt permittivity change at the interface of air and saturated soil causes a different period average response than would occur with the more gradual permittivity change found when the sensor rods are completely inserted in the soil. 

    For example, if a CS650 or a CS655 was inserted halfway into a saturated soil with a volumetric water content of 0.4, the sensor would provide a different period average and permittivity reading than if the probe was fully inserted into the same soil when it had a volumetric water content of 0.2.

  7. Damage to the CS650 or the CS655 electronics or rods cannot be repaired because these components are potted in epoxy. Cable damage, on the other hand, may possibly be repaired. For more information, refer to the Repair and Calibration page.

  8. The CS650-series sensors have several logical tests built into their firmware to ensure that the sensors do not report a number that is known to be erroneous. Erroneous readings are either outside the sensor’s operational limits or outside of published accuracy specifications.

    A reported value of NAN or 9999999 does not necessarily mean that there is a problem with the sensor hardware. The conditions outlined below can lead to a value of NAN or 9999999 for permittivity and volumetric water content.

    SDI-12 communications issue

    If all of the following are true, there is likely an issue with the SDI-12 communications between the sensor and the datalogger: the sensor is being polled with an M1! SDI-12 command, the permittivity value reported is NAN, and subsequent values are all zeroes or never change. Possible causes include the following:

    • The sensor is not powered. Check to make sure that the red wire is well connected to a 12 Vdc source and that the black wire is connected to datalogger ground.
    • The sensor signal wire is not connected properly. Ensure that the green wire is well connected to the control port specified in the datalogger program.
    • The sensor has an SDI-12 address that does not match the datalogger program.
    • The sensor has cable or lightning damage.

    Calculated permittivity is less than 0 or greater than 88

    The equation used to convert period average and electrical conductivity values to permittivity is a three-dimensional surface with two independent variables and eleven coefficients, plus an offset. Some rare combinations of period and electrical conductivity result in a permittivity calculation that is less than air (1) or greater than water at 0°C (88). These rare combinations are not expected when the sensor is in soil. 

    Bulk electrical conductivity (EC) is greater than 1.14 dS/m

    When bulk electrical conductivity is greater than 1.14 dS/m, the solution EC is greater than 3 dS/m, which is the upper limit for accurate readings with the CS650. When this occurs, the soil is considered out-of-bounds and will report a value of NAN or 9999999 for both permittivity and volumetric water content.

    Calculated permittivity is less than 80% of the permittivity limit

    A permittivity limit based on the bulk electrical conductivity (EC) reading is used to determine whether the bulk EC at saturation exceeds the sensor’s operational limit. That permittivity limit is calculated and compared to the permittivity reading. If the measured permittivity is more than 20% beyond the permittivity limit, both permittivity and volumetric water content are reported as NAN or 9999999. This is the most common cause of NAN values with the CS650-series sensors, and it occurs because of the soil properties and not because of a sensor malfunction.

  9. No. It is not possible to disable the logical tests in the firmware. If soil conditions cause frequent NAN values, it may be possible to perform a soil-specific calibration that will provide good results.

    If permittivity is reported but the volumetric water content value is NAN, Campbell Scientific recommends a soil-specific calibration that converts permittivity to water content. This will take advantage of the bulk electrical conductivity correction that occurs in the firmware.

    If both permittivity and volumetric water content have NAN values, it may be possible to perform a calibration that converts period average directly to volumetric water content.

    For details on performing a soil-specific calibration, refer to “The Water Content Reflectometer Method for Measuring Volumetric Water Content” section in the CS650/CS655 manual. After a soil-specific equation is determined, it may be programmed into the datalogger program or used in a spreadsheet to calculate the soil water content.

  10. If a system has multiple CS650 or CS655 sensors, it will be necessary to connect many wires to a 12 V supply and many wires to ground. The DIN-Rail Mounting Kit is useful for attaching many wires to the same source in a clean and organized way. For more details, see the 5458 DIN-Rail Terminal Kit instruction manual

    Other methods of connecting several wires together, such as terminal strips or wire nuts, would also work.

Articles and Press Releases

Privacy Policy Update

We've updated our privacy policy.  Learn More