Log In   |   Sign up

New User Registration

Article / Abstract Submission
Register here
Press Release Submission
Register here
coolingZONE Supplier
Register here

Existing User

            Forgot your password
December 2005
library  >  Application Notes  >  General Articles

Air Velocity Measurement Using Thermistors

traditional methods of air velocity measurement use either a hot-wire anemometer or vane anemometer. both are single-point measurement devices that force the user to take one measurement point at a time.


this article introduces multipoint air velocity measurement using thermistor-based technology and discusses using embedded single-point thermistor sensors for continuous monitoring of air velocity. advantages over traditional methods include durability, reliability, accuracy, and shortened testing times. plus, the small sensor adds flexibility to where measurements can be taken.


when thermal airflow sensors were developed originally, the test measurement industry chose the hot-wire as the principal measurement medium. hot-wires provide a fairly linear signal with reasonable repeatability and fast response time.

however, hot-wires have several drawbacks. fragility causes loss of calibration or physical damage if a unit is dropped accidentally. also, the hot-wire may oxidize over time, changing calibration. further, because the hot-wire signal is weak, a small drift over time can cause loss of calibration or inconsistent readings.


the key factor is that if any of these problems occur, the entire unit must be sent back to the factory for repair and/or recalibration, leaving the certifier or tester without equipment to meet customer demand for several days. what's more, even positioning the hot-wire can become a problem. a turn of one's hand or not being in the exact sash plane will produce different readings.

glass-encapsulated thermistors have been in use for the past several years in such areas as medical applications. the original drawbacks of the thermistor caused the industry to use hot-wire when creating the first air velocity instrument. every thermistor has a slightly different shape, which made interchangeability difficult. further, the raw signal from the thermistor is extremely non-linear. however, with the advent of inexpensive microprocessors, interchangeability and linearity were achievable in the calibration process.

note that the multipoint thermistor-based system is different than the multipoint pressure pickup device used to obtain an average velocity. the pitot-based pressure system gives only one reading of the flow. it does not provide individual readings for each sensor location. hence, it is not an appropriate tool for profiling laminar flow fluctuations over an area. also, using a pressure-based system to measure airflow can be problematic at 100 fpm. the equivalent differential pressure at 100 fpm is approximately 0.0007 inches of water.


finding an accurate pressure sensor for these low air velocities is very difficult and extremely expensive. clearly, thermistors provide the best solution for measuring low air velocity.


thermistor air velocity sensors, when they are designed in conjunction with the latest technologies, can have several unique features when compared with other types of airflow sensors.

  • interchangeability - inexpensive microprocessors allow for each thermistor to provide linearity and repeatability. each thermistor sensor has its own unique calibration intrinsic to itself, not dependent on the accompanying multiplexer instrument. linked to the instrument via a connector, the sensors are completely interchangeable with other such sensors. because the calibration is resident in each sensor, the use can plug any sensor into any location without having to tell the instrument. different calibrations can be used at the same time with no field calibration. in addition, if a sensor must be returned to the factory (such as for recalibration), the user can simply plug in another sensor to take its place and continue working without any disruption. in other words, the user will not have to be without the instrument for several days, or longer, while sensors are being recalibrated.

  • strong signal - the thermistor signal may be as much as 1000 times larger than that of a hot-wire. the sensor has minimal drift, which will not change calibration significantly.

  • durability - thermistors are resistant to shock and vibration. thermistors can be dropped without losing their calibration or sustaining physical damage. the glass-encapsulated thermistor is also durable if exposed to a hazardous material, and it can be cleaned safely using an ultrasonic bath.

  • repeatability - every thermistor is mathematically modeled, giving performance within 3% of each other.

  • small size - the sensors have an extremely low profile (approximately .190 inches wide by .028 inches deep), with a thin teflon cable, providing minimal obstruction of the airflow and flexibility of location.

  • measurement range - the sensors can accurately measure as low as 30 fpm to as high as 5,000 fpm. 


compensation for temperature

thermal airflow sensors are all affected by temperature. the signal for the sensor is a function of the temperature difference between the thermistor (or hot wire) and the ambient airflow. it follows that to obtain a true reading, the signal has to be corrected for ambient temperature effects. so, the temperature of the airflow must be measured so that effective compensation can be developed.

there are 2 ways this is done. the first method uses a second thermistor to measure temperature, positioned very close to the airflow thermistor, providing a continuous temperature reading for compensation.

the second method uses the same thermistor to carry out both airflow and ambient measurements. what is done is that the thermistor is heated up for the airflow measurement, and then allowed to cool to ambient for the temperature measurement. this on-off cycle operates continuously. as usual, there are trade-offs with each method.

the single thermistor method, because of the need to heat up and cool down, is limited by the thermal time constant of the thermistor. in practice, such a cycle cannot be reliably completed in much under 1 second, unless a smaller device is used (which reduces the time constant). a smaller thermistor increases the fragility of the sensor, thereby reducing ruggedness. furthermore, continuous and severe thermal cycling reduces mtbf, due to associated stresses in the glass-to-metal seal and resulting contamination of the device. these limitations probably offset any perceived marginal advantages this method might offer.

the dual thermistor method also has its trade-offs. there is a small positional (perhaps 0.100" or so) difference between the 2 thermistors, which marginally increases air resistance and size. in the vast majority of cases this is not an issue. it has also been suggested that the small positional difference can cause the 2 thermistors to operate in slightly displaced adjacent layers of airflow. in practice, airflow in cooling systems, and in most other applications, is turbulent enough to render the difference meaningless and immaterial.


applications-simultaneous multipoint measurement

multipoint air velocity measurement has applications for thermal management, fume hood and biosafety cabinet testing and certification, and for cleanroom certification and monitoring. up to 24 sensors can be used simultaneously to run "what-if" scenarios quickly while providing valuable information regarding potential flow disruptions.

thermal management

as critical components, such as microprocessors, become smaller and more powerful, there is an increased demand for better cooling methods. yet, increasing power densities conflict with the demand for smaller product size, creating significant thermal design issues for engineers.

traditional methods:

using a hand-held probe from location to location is time-consuming, inexact, and tedious. holes are drilled into the system and the hand-held is moved from location to location. the hand-held anemometers are large, difficult to maneuver, and do not give repeatable readings. furthermore, the tester must take a measurement, write it down, take another measurement, etc., then run the tests all over again. this method can take several days when testing a large electronics enclosure.

multipoint air velocity measurements:

there are several important advantages to using multipoint measurements. first, thermistors allow for the creation of extremely small sensors extended on thin, flexible cable as long as 10 meters. the user can access tight and difficult to reach locations that were not previously possible with a traditional anemometer. now, air velocity rates can be determined in areas that were previously unreachable. in addition, the small size does not obstruct the airflow to the same level seen with the large anemometers, providing more accurate and repeatable readings.

multipoint measurements also determine the flow pattern throughout the system quickly. scenarios can be changed with immediate results known. ease of data collection and analysis provides a tremendous advantage over a single-point anemometer. the data is transmitted via an rs-232 to a pc program which collects the data in a spreadsheet and creates graphs. calculations such as average, minimum, maximum and standard deviation are automatically calculated. no data entry is necessary by the tester.

a variety of thermal management problems have been solved using multipoint air velocity measurement. some obvious examples include the evaluation of fans and heat sink performance (i.e. do i need more or larger fans; are the heatsinks appropriate?). another application is verification of cfd software modeling. it is an accepted industry practice that software models are verified with test experiments and mock-ups. similarly, multipoint measurements have been used to profile printed circuit bards to determine critical component placement and to monitor quality. one application had a large computer enclosure with a complex baffling system to move air throughout the system.

multipoint measurements were used to achieve a uniform airflow by adjusting the baffles. the multipoint system allowed the user to adjust the baffles and see the velocity effects immediately for the entire computer. one cannot even imagine trying to adjust baffles using single point measurements. lastly, the multipoint air velocity system can help answer the overall question of "why is my enclosure burning up?"

fume hood face velocity measurement and biosafety cabinet downflow testing

standards for both fume hoods and biosafety cabinets (bscs) require an imaginary grid at the face for fume hoods or within the bsc where multiple measurements are taken and averaged. facilities can have several hundred fume hoods or bscs in one location. simultaneous, multipoint measurements can reduce testing time while running "what-if" scenarios quickly and easily. conversely, single-point measurements can be tedious in computing such averages and individual readings.


repeatability and consistency of results are also important factors in certification and testing. because of its ruggedness and minimal drift over time, thermistor technology provides much better repeatability than hot-wires. due to the modeling of the thermistor behavior in calibration, different thermistor sensors will yield results within 3% of each other.

positioning the single-point instruments in the imaginary grid can also be a problem. turning one's hand when fixing the location of the hot-wire in the sash plane can lead to very different results. with fume hood face velocity measurement, in particular, it is difficult to fix the location of the hot-wire on the sash plane. however, because the multipoint system has an actual, fixed grid that sits along the sash plane, repeatability in sensor placement is achieved easily.

cleanroom testing

typical cleanroom testing dictates measuring air velocity throughout the cleanroom, and particularly in the ceiling hepa filters. in large cleanrooms, there can be hundreds of areas that must be tested. typically, the volumetric flow or single air velocity is taken using a multipoint pitot pickup into a pressure sensor. however, like safety enclosures, laminar flows at the hepa filters are extremely important.


a change in the laminar flow could have critical consequences on the manufacturing process within the cleanroom. therefore, it is important to understand the profile of the air coming out of the filter, not simply a single-point velocity or volumetric flow. taking a multipoint air velocity measurement will not only calculate the volumetric flow (cfm), but will also give a profile of the air velocities within the hepa. therefore, the balance/tester can have a better understanding of occurrences surrounding critical areas of hepas.

applications-continuous monitoring of airflow

typically, continuous measurement of airflow has been done using either a ceramic tipped or hot-wire anemometer. rtd's are also used differential pressure is sometimes used to calculate air velocity. we have already noted the contra-indications for using differential pressure and some of the problems with the hot-wire. if one wants to continuously monitor air velocity by permanently embedding a hot-wire or ceramic sensor, then the issues of fragility and loss of calibration are much more pronounced than when they are used in the traditional role as testing devices.


ceramic tip sensors are relatively fragile, have an expensive per unit cost, and their recalibration and repair costs are high. the thermistor-based airflow sensors give a linear output with either a 4-20ma or voltage output at approximately half the cost of the ceramic sensor.

thermistor-based sensors have been used successfully in the following applications:

continuous monitoring of airflow in cleanrooms by hepas and in critical areas

the thermistor-based sensors provide air velocity measurements on a full-time basis. critical areas such as minienvironments have used them to monitor the flow of air out of a hepa filter and to measure the airflow in minienvironments. bi-directional versions allow the user to determine which direction the airflow is flowing, which can be crucial in cleanroom applications. the sensors can be linked to a facility monitoring system for long-distance data collection.

variable air volume control of fume hoods and safety enclosures

the linear signal from the thermistor sensor enables the user to quickly adjust the vav control of the fume hood. the sensor's fast response provides quick detection of sash openings and closings, saving on energy costs while obtaining maximum protection for the user. operating rooms and other isolation rooms operating rooms must have the air flowing in the proper direction to protect patients and the external environment from infection. the ability to read low air velocities combined with bi-directional sensing gives designers affordable options for room containment applications.

as a final note, the thermistor-based sensors meet nsf accuracy requirements for the downflow tests of a biosafety cabinet.

about raouf esmail:

with a b.a. & m.a. from cambridge university, raouf came to the us and developed an expertise in using thermistors to measure airflow while getting his mba from harvard.

he founded cambridge aero instruments, a leader in sailplane instrumentation and then went on to create cambridge aeroflo, a developer of fan controllers for electronics cooling. in 1995, raouf then founded cambridge accusense to focus on air velocity sensors and multipoint instrumentation. major computer and telecommunications companies around the world use accusense's products.

if you have questions about this article, please email our editor jperry@coolingzone.com.

Choose category and click GO to search for thermal solutions


Subscribe to Qpedia

a subscription to qpedia monthly thermal magazine from the media partner advanced thermal solutions, inc. (ats)  will give you the most comprehensive and up-to-date source of information about the thermal management of electronics


Submit Article

if you have a technical article, and would like it to be published on coolingzone
please send your article in word format to articles@coolingzone.com or upload it here

Subscribe to coolingZONE

Submit Press Release

if you have a press release and would like it to be published on coolingzone please upload your pr  here

Member Login

Supplier's Directory

Search coolingZONE's Supplier Directory
become a coolingzone supplier

list your company in the coolingzone supplier directory

suppliers log in

Media Partner, Qpedia


Heat Transfer Calculators