SBE 3S Oceanographic Temperature Sensor

SBE 3S Oceanographic Temperature Sensor

Slow-response, frequency-output temperature sensor, for high-accuracy industrial and environmental monitoring at fixed sites.

Intended primarily for use on moored and fixed-site temperature measuring systems, the SBE 3S can also be used as a component in custom systems or for high-accuracy industrial and environmental temperature monitoring applications. The low noise characteristics of the SBE 3S allow the use of hybrid frequency measuring techniques to obtain rapid sampling with very high resolution; 40 μ°C resolution can be readily obtained at a 6 Hz sampling rate.

The superior performance of the SBE 3S results from its optimized electronic design, superior calibration, and quality testing program. The SBE 3S has a time response of approximately 0.6 sec and an initial accuracy of 0.001 °C, and is typically stable to 0.002 °C/year.

Note: See the SBE 38 for a digital oceanographic thermometer with RS-232 or RS-485 output.


  • Glass-coated thermistor bead, pressure-protected in 2.1 mm diameter thin-walled stainless steel tube. Exponentially related to temperature, the thermistor resistance is the controlling element in an optimized Wien Bridge oscillator circuit. Resulting sensor frequency is inversely proportional to the square root of the thermistor resistance and ranges from approximately 2 to 6 kHz, corresponding to -5 to +35 °C.
  • Built-in acquisition circuits and frequency outputs; allows for calibration as separate modules.
  • Individually calibrated in Sea-Bird’s computer-controlled, super-low-gradient calibration baths.
  • Overall system accuracy limited only by the accuracy of the logger’s master clock. A typically small clock error of 1 ppm affords a temperature error of less than 50 μ°C.
  • 3400 m aluminum or 10,500 m titanium housing.
  • Five-year limited warranty.


SBE 3S sensors are calibrated to ITS-90 temperature using Sea-Bird’s computer-controlled calibration baths. Extremely well insulated, the baths provide a uniform toroidal circulation, yielding an overall transfer accuracy against an SPRT within 0.0002 °C. Repeatability at each of twelve individually mapped sensor positions is better than 0.0001 °C.

Sea-Bird’s metrology lab underpins the temperature calibration baths. Following consultation with the U.S. National Institute of Standards and Technology, the lab was configured to achieve temperature precision of 50 µK and accuracy of0.0005 °C. To obtain this performance, premium primary references, including four Jarrett water triple-point cells (with maintenance bath) and an Isotech gallium melt cell, are operated in conjunction with two YSI 8163 standards-grade platinum resistance thermometers and an ASL F18 Automatic Temperature Bridge.


  • Aluminum (3400 m) or titanium (10,500 m) housing.
  • XSG or wet-pluggable MCBH connector.


The calibration yields four coefficients (g, h, i, j) that are used in the following equation (Bennett, 1972):

T [°C] = [1 / (g + hln(fo/f) + iln²(fo/f) + jln³(fo/f) )] - 273.15

where T is temperature [°C], ln is natural log function, and f is SBE 3S output frequency [Hz]. Note that fo, an arbitrary scaling term used for purposes of computational efficiency, was historically chosen as the lowest sensor frequency generated during calibration. For calibration results expressed in terms of ITS-90 temperatures, fo is set to 1000. Calibration fit residuals are typically less than 0.0001 °C.

Example Calibration Data (sensor serial number 2213, 30 May 1996):
g = 4.28793855e-03     h = 6.25807786e-04     i = 2.19368239e-05     j = 1.84262924e-06     f0 = 1000.000

Temperature [°C]
Frequency [Hz]
Temperature [°C]
(Instrument - Bath) [°C]
-1.4262 2727.631 -1.4262 -0.00001
1.0833 2888.869 1.0834 0.00003
4.5745 3124.656 4.5745 0.00000
8.1730  3382.083 8.1730 -0.00006
11.6052 3641.635 11.6052 -0.00001
15.1623 3925.500 15.1623 0.00003
18.6658 4220.277 18.6659 0.00006
22.1644 4530.069 22.1644 0.00001
25.7234 4861.419 25.7234 -0.00004
29.1380 5195.062 29.1380 -0.00007
 32.6711 5556.867 32.6711 0.00005


Measurement Range -5.0 to +35 °C
Initial Accuracy 1 ± 0.001 °C
Stability 0.002 °C per year typical
Response Time 2 0.580 sec ± 0.010 sec (1.0 m/s water velocity)
0.690 sec ± 0.010 sec (0.5 m/s water velocity)
Self-heating Error < 0.0001 °C in still water
Settling Time < 0.5 sec to within 0.001 °C

1 NIST-traceable calibration applying over entire oceanographic range.
2 Time to reach 63% of final value following a step change in temperature.


Input Power 11 - 16 VDC, 25 mA
Output Signal ± 0.5V square wave



6061 Aluminum housing Depth rating: 3400 m; Weight: 0.6 kg in air; 0.3 kg in water
6Al-4V Titanium housing Depth rating: 10,500 m; Weight: 0.9 kg in air; 0.6 kg in water


The list below includes (as applicable) the current product brochure, manual, and quick guide; software manual(s); and application notes.

Title Type Publication Date PDF File
SBE 3S Brochure Product Brochure Monday, May 18, 2015 03sbrochureMay15.pdf
AN42: ITS-90 Temperature Scale Application Notes Wednesday, May 18, 2016 appnote42May16.pdf
AN57: Connector Care and Cable Installation Application Notes Tuesday, May 13, 2014 appnote57Jan14.pdf

What are the recommended practices for connectors - mating and unmating, cleaning corrosion, and replacing?

Mating and Unmating Connectors:

It is important to prepare and mate connectors correctly, both in terms of the costs to repair them and to preserve data quality. Leaking connectors cause noisy data and even potential system shutdowns. Application Note 57: Connector Care and Cable Installation describes the proper care and installation of connectors for Sea-Bird instruments. The Application Note covers connector cleaning and cable or dummy plug installation, locking sleeve installation, and cold weather tips.

Checking for Leakage and Cleaning Corrosion on Connectors:

If there has been leakage, it will show up as green-colored corrosion product. Performing the following steps can usually reverse the effect of the leak:

  1. Thoroughly clean the connector with water, followed by alcohol.
  2. Give the connector surfaces a light coating of silicon grease.

Re-mate the connectors properly — see Application Note 57: Connector Care and Cable Installation and 9-minute video covering O-ring, connector, and cable maintenance.

Replacing Connectors:

  • The main concern when replacing a bulkhead connector is that the o-rings on the connector and end cap must be prepared and installed correctly; if they are not, the instrument will flood. See the question below for general procedure on handling o-rings.
  • Use a thread-locking compound on the connector threads to prevent the new connector from loosening, which could also lead to flooding.
  • If the cell guard must be removed to open the instrument, take extra care not to break the glass conductivity cell.

What are the recommended practices for storing sensors at low temperatures, and deploying at low temperatures or in frazil or pancake ice?


Large numbers of Sea-Bird conductivity instruments have been used in Arctic and Antarctic programs.

Special accommodation to keep temperature, conductivity, oxygen, and optical sensors at or above 0 C is advised. Often, the CTD is brought inside protective doors between casts to achieve this.

Conductivity Cell

When freezing is possible, we recommend that the conductivity sensor be stored dry. Remove larger droplets of water by blowing through the cell. Do not use compressed air, which typically contains oil vapor. Attach a length of Tygon tubing to each end of the conductivity cell to close the cell ends. See Application Note 2D: Instructions for Care and Cleaning of Conductivity Cells for details.

There are several considerations to weigh when contemplating deployments at low temperatures in general, and in frazil or pancake ice:

  • Ensure that the instrument is at or above water temperature before it is deployed. If the cell gets colder than 0 to -2 ºC while on deck, when it enters the water a layer of ice forms inside the cell as the cell warms to ocean temperature. If ice forms inside the conductivity cell, measurements will be low of correct until the ice layer melts and disappears. Thin layers of ice will not hurt the conductivity cell, but repeated ice formation on the electrodes will degrade the conductivity calibration (at levels of 0.001 to 0.020 psu) and thicker layers of ice can lead to glass fracture and permanent damage of the cell.
  • For accurate measurements, keep ice out of the sensing region of the conductivity cell. The conductivity measurement involves determining the electrical resistance of the water inside the sensor. Ice is essentially a non-conductor. To the extent that ice displaces the water, the conductivity will register (very) misleadingly low. Some type of screening is necessary to keep ice out of the cell. This is relatively easy to arrange for the Sea-Bird conductivity cell, which is an electrode-type cell, because its sensing region is totally inside a long tube; plastic mesh could be positioned at each end and would have zero effect on accuracy and stability.

The above considerations apply to all known conductivity sensor types, whether electrode or inductive types. 

If deploying at low temperatures but no surface frazil or pancake ice is present, rinse the conductivity cell in one of the following salty solutions (salty water depresses the freezing point) to prevent freezing during deployment. But this does not mean you can store the cell in one of these solutions outside . . . it will freeze.

  • Solution of 1% Triton in sterile seawater (use 0.5-micron filtered seawater or boiled seawater),   or
  • Brine solution (distilled seawater or homemade salt solution that is higher than 35 psu in salinity).

Note that there is still a risk of forming ice inside the conductivity cell if deploying through frazil or pancake ice on the surface, if the freezing point of the salt water is the same as the water temperature. Therefore, we recommend that you deploy the conductivity cell in a dry state for these deployments.

Commercially available alcohol or glycol antifreezes contain trace amounts of oils that will coat the conductivity cell and the electrodes, causing a calibration shift, and consequently result in errors in the data. Do not use alcohol or glycol in the conductivity cell.

Temperature Sensor

In general, neither the accuracy of the temperature measurement nor the survival of the temperature sensor will be affected by ice.

Oxygen Sensor

For the SBE 43 and SBE 63 Dissolved Oxygen sensor, avoid prolonged exposure to freezing temperature, including during shipment. Do not store the with water (fresh or seawater), Triton solution, alcohol, or glycol in the plenum. The best precaution is to keep the sensor indoors or in some shelter out of the cold weather.

What is the function of the zinc anode on some instruments?

A zinc anode attracts corrosion and prevents aluminum from corroding until all the zinc is eaten up. Sea-Bird uses zinc anodes on an instrument if it has an aluminum housing and/or end cap. Instruments with titanium or plastic housings and end caps (for example, SBE 37 MicroCAT) do not require an anode.

Check the anode(s) periodically to verify that it is securely fastened and has not been eaten away.

How often do I need to have my instrument and/or auxiliary sensors recalibrated? Can I recalibrate them myself?

General recommendations:

  • Profiling CTD — recalibrate once/year, but possibly less often if used only occasionally. We recommend that you return the CTD to Sea-Bird for recalibration. (In principle, it is possible for calibration to be performed elsewhere, if the calibration facility has the appropriate equipment andtraining. However, the necessary equipment is quite expensive to buy and maintain.) In between laboratory calibrations, take field salinity samples to document conductivity cell drift.
  • Thermosalinograph — recalibrate at least once/year, but possibly more often depending on the degree of bio-fouling in the water.
  • DO sensor —
    — SBE 43 — recalibrate once/year, but possibly less often if used only occasionally and stored correctly (see Application Note 64), and also depending on the amount of fouling and your ability to do some simple validations (see Application Note 64-2)
    — SBE 63 — recalibrate once/year, but possibly less often if used only occasionally and stored correctly and also depending on the amount of fouling and your ability to do some simple validations (see SBE 63 manual)
  • pH sensor —
    — SBE 18 pH sensor or SBE 27 pH/ORP sensor — recalibrate at the start of every cruise, and then at least once/month, depending on use and storage
    — Satlantic SeaFET pH sensor — recalibrate at least once/year. See FAQ tab on Satlantic's SeaFET page for details (How often does the SeaFET need to be calibrated?).
  • Transmissometer — usually do not require recalibration for several years. Recalibration at the manufacturer’s factory is the most practical method.

Profiling CTDs:

We often have requests from customers to have some way to know if the CTD is out of calibration. The general character of sensor drift in Sea-Bird conductivity, temperature, and pressure measurements is well known and predictable. However, it is very difficult to know precisely how far a CTD calibration has drifted over time unless you have access to a very sophisticated calibration lab. In our experience, an annual calibration schedule will usually maintain the CTD accuracy to within 0.01 psu in Salinity.

Conductivity drifts as a change in slope as a result of accumulated fouling that coats the inside of the conductivity cell, reducing the area of the cell and causing an under-reporting of conductivity. Fouling consists of both biological growth and accumulated oils and inorganic material (sediment). Approximately 95% of fouling occurs as the cell passes through oil and other contaminants floating on the sea surface. Most conductivity fouling is episodic, as opposed to gradual and steady drift. Most fouling events are small and mostly transitory, but they have a cumulative affect over time. A severe fouling event, such as deployment through an oil spill, could have a dramatic but only partially recoverable effect, causing an immediate jump shift toward lower salinity. As fouling becomes more severe, the fit becomes increasingly non-linear and offsets and slopes no longer produce adequate correction, and return to Sea-Bird for factory calibration is required. Frequently checking conductivity drift is likely to be the most productive data assurance measure you can take. Comparing conductivity from profile to profile (as a routine check) will allow you to detect sudden changes that may indicate a fouling event and the need for cleaning and/or re-calibration.

Temperature generally drifts slowly, at a steady rate and predictably as a simple offset at the rate of about 1-2 millidegrees per year. This is approximately equal to 1-2 parts per million in Salinity error (very small).

Pressure sensor drift is also an offset, and annual comparisons to an accurate barometer to determine offset will generally keep the sensor within specification for several years, particularly as the sensors age over time.

Do I need to clean the exterior of my instrument before shipping it to Sea-Bird for calibration?

Remove as much biological material and/or anti-foul coatings as possible before shipping. Sea-Bird cannot place an instrument with a large amount of biological material or anti-foul coating on the housing in our calibration bath; if we need to clean the exterior before calibration, we will charge you for this service.

  • To remove barnacles, plug the ends of the conductivity cell to prevent the cleaning solution from getting into the cell. Then soak the entire instrument in white vinegar for a few minutes. After scraping off the barnacles and marine growth, rinse the instrument well with fresh water.
  • To remove anti-foul paint, use a Heavy Duty Scotch-Brite pad ( or similar scrubbing device.

Family Model . Housing Connector Miscellaneous (factory use)
03 S – Slow . 1 – 3400 m (aluminum) 1 – XSG x
      3 – 10,500 m (titanium) 2 – MCBH  

Example: 03S.11x is an SBE 3S with 3400 m housing and XSG connector. See table below for description of each selection:

3S OCEANOGRAPHIC TEMPERATURE SENSOR - 500 ms (Slow) time response, modular sensor (square wave output). Includes complete documentation. SBE 3S is a slow response temperature sensor, intended for moored applications.

For faster response sensors for profiling applications, see SBE 3plus for premium temperature sensor supplied with standard SBE 9plus CTD, and see SBE 3F for standard temperature sensor supplied with SBE 25 and 25plus CTD.

SBE 3S Housing (depth rating) Selections — MUST SELECT ONE
03S.1xx Aluminum housing, 3400 meter depth rating  
03S.3xx Titanium housing, 10,500 meter depth rating  
SBE 3S Connector Selections — MUST SELECT ONE
03S.x1x XSG connector

Wet-pluggable connectors may be mated in wet conditions. Their pins do not need to be dried before mating. By design, water on connector pins is forced out as connector is mated. However, they must not be mated or un-mated while submerged. Wet-pluggable connectors have a non-conducting guide pin to assist pin alignment & require less force to mate, making them easier to mate reliably under dark or cold conditions, compared to XSG/AG connectors. Like XSG/AG connectors, wet-pluggables need proper lubrication & require care during use to avoid trapping water in sockets.


03S.x2x Wet-pluggable (MCBH) connector



Many cables, mount kits, and spare parts can be ordered online.


  • 17029 3-pin pigtail cable (RMG-3FS with lock sleeve), 0.5 m, DN 30580
  • 17030 3-pin pigtail cable (RMG-3FS with lock sleeve), 2.4 m, DN 30580
  • 171638 3-pin pigtail cable, Wet-pluggable (MCIL-3FS with MCDSL-F), 2.4 m, DN 32646

Mount Kits

(seawater intake pipe leading to) SBE 21 
80403 Thermosalinograph Stainless Remote Temperature Sensor Mount Kit (document 67040)
Note: SBE 21 with serial number > 3110 are not compatible with SBE 3S; use SBE 38 instead.

Spare Parts

  • 23041 Zinc anode ring for end cap