Tom Bishop, P.E.
EASA Senior Technical Support Specialist

Note: This "encore" technical article first appeared in the September 2004 issue of Currents.

ll of us in the electrical appara­tus service industry test the winding ground insulation resistance of ma­chines such as motors and generators. A frequent question is: What is the minimum acceptable megohm (M.) value for this winding? The good news is that there is a standard that identi­fies minimum values for insulation resistance of rotating machines. 

That standard is the “IEEE Recom­mended Practice for Testing Insulation Resistance of Rotating Machinery,” IEEE Std 43-2000. The EASA “Recom­mended Practice For The Repair Of Rotating Electrical Apparatus,” ANSI/EASA AR 100-2010, uses IEEE 43 for its insulation resistance test references. Note that IEEE 43 only applies to rotat­ing machinery. There is no equivalent standard for non-rotating electrical machinery such as transformers. In this article we will delve into determining minimum insulation resistance for rotating electrical machinery. 

Purpose of testing
Before going into the main topic of mini­mum insulation resis­tance, let’s go through a brief refresher on the in­sulation resistance (IR) test. The primary pur­poses of an insulation resistance test are to determine if a machine can be placed back into service, or if high poten­tial winding tests such as hipot or surge can prudently be per­formed. Another important reason for the IR test is to establish a baseline value for the winding insulation condition of an installed machine. If the winding IR value is below the acceptable minimum, the machine should not be energized, and high potential tests should not be performed until the IR value meets or exceeds the prescribed minimum. (Note: See “A Closer Look at High Potential Testing of Rotating Electrical Machine Windings” in the August 2003 issue of Currents for information on performing a hipot test.)

Good winding ground insulation behaves like a capacitor, becoming charged when voltage is applied across it. The IR value per IEEE 43 should be taken after applying and holding test voltage on the winding for 1 minute. That allows some of the capacitive ef­fect to stabilize, making the readings more consistent. (The theory of insula­tion capacitance and other components is very complex and beyond our scope in this article.)

The most common instrument used for the test is the battery powered 500 volt (DC) megohm-meter. One lead is applied to the winding leads (typically all tied to the instrument lead) and the other instrument lead to the frame (ground) of the machine. The meter, which may be digital or analog, will display the winding megohm value. The meter actually applies a voltage and measures the current, then dis­plays the value of voltage divided by current, i.e., resistance, in megohms. The megohmmeter voltage should be applied and held for 1 minute, and the reading at 1 minute is recorded as the IR value for the test. The recommended test voltage to apply with the megohm­meter increases with machine voltage rating as illustrated in Table 1.

Winding temperature affects the megohm value result. As temperature increases, insulation resistance decreases. The cause is not insulation degradation with temperature, but is a physical prop­erty of the insulation materials. The IR reading must therefore be corrected for temperature. The temperature correction per IEEE 43 should be to 40º C; and the correction factor for temperature is such that the minimum insulation resistance value is doubled for every 10º C decrease in winding temperature.  

Note:  Windings that are very hot, e.g., over 100º C due to oven baking, may result in relatively low megohm values. Allow the wind­ing to cool to 60º C or lower and then perform a temperature corrected IR test.
Expressed as a formula the tem­perature correction factor is: 
K_t = (0.5) ^(40-T)/10 
K_t     Factor to multiply T by to obtain insulation resistance corrected to 40°C
T     Temperature in °C at which insulation resistance was measured 
Here is an example for a winding being tested is at 35°C: 
K_t  = (0.5) ^(40-35)/10 = (0.5) ^5/10  = (0.5) ^1/2  = 0.707 
If the measured IR was 100 MΩ. 
at 35º C, the corrected megohms would be: 
100 x 0.707 = 70.7 or about 71 MΩ. at 40°C. 

Humidity can affect IR readings. However, there are no formulas or “rules of thumb” for the effect of humidity with respect to rotating electrical machine windings. None­theless, it is a good practice to record the humidity reading as well as winding temperature and insulation resistance for each IR test. Knowing these parameters, insulation resis­tance readings can be evaluated for the possible effect of humidity, rec­ognizing that at higher humidity the IR reading may be lower for a specific temperature. 

Interpreting the results
Having considered how to per­form the insulation resistance test, we will now address how to interpret the results. For many years the IEEE 43 standard used the rule of “kV + 1” (1 kV = 1000 volts) to determine the minimum megohms for rotating electrical machinery. After nearly 3 decades of being reaffirmed, from 1974 to 2000, the standard was sig­nificantly revised for the 2000 edition. IEEE 43 now states that “for most DC armature and AC windings built after about 1970 (form-wound coils)” the minimum IR is 100 MΩ. That standard also states that “for most machines with random wound stator coils and form-wound coils rated below 1 kV” the minimum IR is 5 MΩ. Captur­ing the remaining windings IEEE 43 states that “for most windings made before about 1970, all field windings, and others not described” above, the minimum IR is kV+1. For a 240-volt field winding the minimum, using the “kV+1” rule, would be .24 + 1 = 1.24 MΩ.   

Note that all of these mini­mum values are based on a winding temperature of 40º C. The winding rated voltage can be either AC or DC, and the standard states that line-to-line voltage be used for three phase windings, line-to-ground volt­age for single phase windings, and (nameplate) rated volt­age for DC machines or field windings. If the IR is below the minimums as described above, the machine should not be started or high potential tested. Corrective measures, such as cleaning and baking, should be taken to increase the IR value to an acceptable level.

Minimum IR rules
We will use some examples to demonstrate the application of the minimum IR rules. Our first example is a form-wound stator winding rated 4000 volts, with an IR of 60 MΩ. at 20º C. Recalling that every 10º C decrease results in a halving of the minimum IR value, the 20º C difference means we halve the value twice (1/2 x 1/2 = 1/4), i.e., it is reduced by a factor of 4. That means that the IR value will be 15 MΩ. (60 ÷ 4) at 40º C, a value less than the minimum IR of 100 MΩ. The insulation resistance of this winding is not acceptable and corrective measures should be taken.  

The second example is a random wound armature rated 500 volts DC with an IR of 52 MΩ. at 30º C. Correct­ing the megohm value from 30º C to 40º C, a 10º C difference, the result is 26 MΩ. (52 ÷ 2). The 26 MΩ. value exceeds the 5 MΩ. minimum value; therefore the armature could be energized (e.g., put back in service) or high potential tested if desired.  

The field windings of the random wound armature machine will be used for the third example. The field voltage rating is 150/300 volts and the IR value at 25º C is 12 MΩ. The 15º C (40º C – 25º C) difference dictates that we use the K_t correction factor formula given above. For the sake of brevity we will skip displaying the math and use the resulting factor of 0.354. The corrected megohms are 4.25 MΩ. (12 x 0.354), and the minimum IR is (kV + 1) 1.30 MΩ. Notice that we used the higher voltage rating of the winding to determine the minimum acceptable megohm value. Since 4.25 MΩ. is greater than the 1.30 MΩ. minimum, the fields can be energized or high potential tested if desired.

Polarization index test
Another test, the polarization index (PI) test, is an expansion of the IR test. The PI test extends the time of the IR test, lasting for 10 minutes with constant test voltage applied. The PI is the ratio of the IR at 10 minutes to the IR at 1 minute. For example, if the IR at 1 minute was 270 MΩ. and the IR at 10 minutes was 590 MΩ, the PI would be 2.2 (590 ÷ 270). In general, the PI of a winding should be 2.0 or greater. The PI test uses the capacitive behav­ior of winding insulation to check its integrity. As the capacitance effect stabilizes, leakage current decreases and megohms increase.  

If a DC hipot is used to perform the PI test, the leakage current in mi­croamps is typically measured. Cal­culating the PI from current readings can be done by dividing the 1-minute microamp value by the 10- minute microamp value. For example, if the 1-minute value was 0.28 microamps and the 10-minute value was 0.15 microamps, the PI would be 1.9 (0.28 ÷ 0.15). In this case the PI is less than 2; therefore the insulation system condition does not meet the IEEE 43 standard minimum requirement. Fur­ther baking and/or cleaning would be in order. 

The megohm value of the insula­tion resistance can be obtained from the DC hipot readings by dividing the applied voltage by the microamps of leakage current. If the PI test voltage in our example was 1000 volts, the 1-minute IR would be 3571 MΩ. (1000 ÷ 0.28), and the 10-minute value would be 6667 MΩ. (1000 ÷ 0.15). Also of note, according to IEEE 43, if the 1-minute IR is greater than 5000 megohms, the PI may not be meaningful and the the PI may be disregarded as a measure of winding condition.

Further testing
One other potentially useful test is to apply the IR test at the lower recom­mended test voltage, then discharge the winding, and repeat the IR test at the higher recommended test voltage. For example, apply 1000 volts to a wind­ing rated 4000 volts and obtain the IR, discharge the winding and repeat the IR test at 2500 volts. The winding can be discharged after the IR test by connect­ing the winding leads together and to ground, e.g., the frame of the electrical machine. (Caution:  See “A Closer Look at High Potential Testing of Ro­tating Electrical Machine Windings” in the August 2003 issue of Currents for information on discharging after a hipot test.)

If the IR values are about the same for both voltages, and exceed the minimum IR recommendations, the winding ground insulation is probably in satisfactory condition. Conversely, if the IR value decreases with applied voltage there is a weak­ness in the insulation system that may necessitate rewinding.

In addition to temperature and humidity, other factors that can affect megohmmeter readings are the instru­ments and the method of application of the test. The implied accuracy of an instrument may change when the megohm scale is changed. By “im­plied” we mean the number of signifi­cant digits. For example, a reading of 190 megohms on the 1000-megohm scale of a meter may be indicated more accurately as 185 MΩ. on the 200-megohm scale. The 1000-megohm scale has 2-place accuracy in this case when reading below 200 MΩ., and the meter has 3-place accuracy on the 200-megohm scale.  

Results when voltage reapplied
Another factor is the charging cur­rent decay and change in reading as voltage is reapplied. The insulation is a capacitor and the current across it be­comes lower with time. Applying the megohmmeter and maintaining the voltage causes the capacitive current to decrease, thus increasing the megohm value (megohms = volts divided by microamps). If the megohm reading at one minute is obtained from one lead of a winding and the megohmmeter reapplied to another set of leads, the second megohm test value will be higher than the first. 

The reason is that some of the winding capacitance has been dis­charged by the first test, thus the sec­ond test begins at a megohm value greater than that of an uncharged winding. This case applies when the leads of the winding being tested are interconnected internally, i.e., there is continuity among all the leads. When there are separate windings, e.g., a 6-lead wye-delta motor with separate circuits of leads 1-4, 2-5 and 3-6, each circuit should be tested individually. The separate circuits should be tested to ground, with all other circuits intentionally connected to ground.

A final note of caution:  Make certain that the machine to be tested has been ventilated prior to the IR test, particularly if a winding fault is sus­pected. Gases from the fault (or from a volatile substance such as a solvent or paint) may be ignited by the IR test if an arc occurs during the test.

Categories: Testing, Motor testing

Tags: Insulation resistance (IR) test

Rate this article:

3.0

Print