Effects of temperatureResistance measurements are dependent on temperature. If the original data was read at one temperature but later tests are conducted at other temperatures, this temperature data is required to determine the suitability of the measurements. All materials do not react to temperature to the same degree. Aluminum, steel, copper and graphite have specific temperature coefficients that will affect the degree of changes that can take place with temperature at the site of the measurement.Low resistance measurements rely on the user conducting the tests within the operating temperature range of the instrument (the user must be aware of field conditions). When the user sees out-of-tolerance measurements, one of the first steps is to check the instrument’s reading with a suitable calibration shunt.As mentioned previously, resistance measurements are dependent on temperature. The resistance of all pure metals increases with rising temperature. The proportional change in resistance for a specific material with a unit change in temperature is called the temperature coefficient of resistance for that material. Temperature coefficients are expressed as the relative increase in resistance for a one degree increase in temperature. While most materials have positive temperature coefficients (resistance increases as temperature rises), carbon graphite materials have negative temperature coefficients (resistance decreases as temperature rises). Table 6 shows the temperature coefficients of resistance for selected materialsvi:Table 6: Temperature coefficients of resistance MaterialPer ºCPer ºFAluminum0.00380.0021Carbon (0 - 1850 ºC)-0.00025-0.00014Constantan (0 - 100 ºC)NegligibleNegligibleCopper (@ 20 ºC)0.003930.00218Iron0.00500.0028Lead0.00430.0024Manganin (0-100 ºC)NegligibleNegligibleMercury0.000900.00050Platinum0.00380.0021Silver0.00400.0021Tungsten0.00450.0025Zinc0.00370.0021Fig 27 shows the temperature resistance curves for some of these materials (based on a baseline reading of 1000 microhms at 20 ºC (68 ºF).vi Electrical Metermen’s Handbook; Third Edition; 1965; page 480When making a measurement on a specific material, the user can calculate the change in resistance due to a change in temperature by multiplying the resistance at the reference temperature by the temperature coefficient of resistance and by the change in temperature:R2-R1 = (R1)(a)(T2 – T1)R1 = resistance of the conductor at the reference temperatureR2 = resistance of the conductor when the measurement is madeT1 = reference temperatureT2 = temperature at which the measurement is madea = temperature coefficient of resistance for the material being testedThe user should also be aware of operating and storage temperature specifications of the instrument they are using to ensure that it is suitable for the environment in which it will be used.Fig 27: Temperature resistance curves for iron, copper and carbonEffects of humidityThe relative humidity of the test specimen should not affect the resistance reading unless the material is hygroscopic, in which case more moisture will be absorbed into the sample at higher humidities. This will change the measurement conditions and will affect the achieved result. However, most conductors are non-hygroscopic. Therefore, since instruments are typically designed with an operating range of from 0 to 95% RH, providing that moisture is not actually condensing on the instrument then a correct reading will be obtained.Background noise, current and voltage Resistance measurements can be degraded by static voltages and ripple currents (electrical noise) impressed on the test specimen. The user should be aware of the level of noise rejection in the instrument being used. Changing to a different model can help the user make a measurement at a difficult test site.The magnitude of the test current used by the instrument will affect the noise rejection capability of that instrument. A 10 A test current will provide much better noise rejection than a 0.1 A test current. Beware of excessive test currents which can change or damage the test sample due to heating (W = I2R). If 100 A is used in place of 10 A, the sample will experience 100 times the heat of the lower test current. With that said, use appropriate test current based on the nominal current rating.The open circuit voltage on most low resistance ohmmeters is low. When making measurements on transformer windings, additional power is required to saturate the winding and allow the meter to stabilize more rapidly. Instruments designed for this type of application have a higher open circuit voltage (in the 50 V d.c. range) to deliver the energy needed to saturate the windings. Then a constant current mode of operation is used to do the resistance measurement.Thermal emf / Seebeck voltage compensationThermal EMF / Seebeck voltage is generated when different conducting materials are part of the same circuit or at different temperatures. The effects of this can be overcome by increasing the current used for the test. Increasing the current will reduce the error, but ensure that it is not too high (heating), see tables below:Table 7: Current error percentage Current VoltageErrorCu-NiCu-AlCu-Ag1 A50 μV400%200%20%10 A500 μV40%20%2%100 A5 mV4%2%0.2%600 A30 mV0.7%0.3%0.03%Table 8: Conducting materials temperatureJunctionμV/ºCCopper - Copper<0.3Copper - Gold0.5Copper - Silver0.5Copper - Brass3Copper - Nickel10Copper - Lead - Tin Solder1 - 3Cooper - Aluminum5Copper - Kovar40Copper - Copper Oxide>500Contact resistance contaminationContact resistance is the resistance to current flow through a closed pair of contacts. Sometimes it takes a high current to break through, melt or soften the contact point and its surrounding area, which increases the contact area and, as such, reduce the resistance.Example: A circuit breaker is tested and its main contact shows a resistance of 300 microhms using a 100 A test current. The test is repeated using a 600 A test current and a resistance of 80 microhms shows, the test is again repeated using a 100 A test current, again the result is 80 microhms.Fig 28: Circuit breaker corrosionNoise ratio and induced currentsIt's common to have noise in a power environment, so to establish an accurate result the measurement signal needs to be greater than the noise generated: ■Low resistance measurement of 50 Ω ■1 A => measurement signal 50 μV ■10 A => measurement signal 500 μV ■100 A => measurement signal 5 mV ■600 A => measurement signal 30 mVFig 29: Noisewww.megger.com 2726 A guide to low resistance testing www.megger.com
