Electrical safety testers are instruments designed to produce high voltage, in excess of 5000VAC and 6000VDC, and monitor the leakage current while the voltage is applied. Electrical safety testers often include ground continuity or high current ground bond testing. This article examines the differences between routine operational checks done on a regular basis on the production line and verification/calibration. We also address the techniques associated with the verification of electrical safety testers including the measurement of voltage, current and ground bond resistance.

Calibration, Verification and Routine Checks

It is important to distinguish between instrument calibration and verification. Calibration is a comparison of an unknown with a known standard with a report of the results. Calibration can include adjustment of the instrument to correct for any deviation between the unknown and known standard. In instrumentation, this means comparing the displayed value on the instrument with a known value from a calibrated meter or standard, and then reporting the results.

The U.S. national standard ANSI/NCSL-Z540 defines “verification” as the “calibration and evaluation of conformity against a specification”. Verification differs from calibration in that the reported results are compared with the manufacturer’s specifications, and that adjustment of the instrument is not included.

A routine operation check is not to be confused with verification. A routine operation check does not verify that the instrument meets manufacturer’s specifications. It is simply designed to quickly check that the instrument is still operating (i.e. outputting voltage).

Routine Operational Checks

A routine operational check, for hipot testing, should determine if voltage is being applied, the tester will fail bad products and the cables out to the device under test are not broken. There is nothing worse than believing the correct voltage is being applied to the DUT, only to find out a cable is broken and the only thing that has been tested is the hipot tester itself and part of the cable. This is especially true if this fault was not caught in time and product has already been shipped to the customer.

Frequently, the equipment used for routine operational checks is a simple load box. The load box consists of one or more high voltage resistors in the range of 100kW to 10MW. The resistors in the load box do not have to be precision resistors or have an accurate calibration. They are only intended to check, not verify or calibrate. Most National Recognized Test Laboratories (NRTLs) such as Underwriters Laboratories will recommend use of a load box on a routine basis. This becomes an important part of a manufacturer’s quality management system.

Figure 1: Example of Resistive Load Boxes

Choosing the correct resistor for a high voltage application requires a little thought. It is necessary to consider the maximum voltage rating, power rating, and voltage coefficients when specifying resistors for these types of applications.

The voltage rating is the maximum voltage that can be applied to the resistor without causing damage to the resistor due to arcing or breakdown.

The power rating defines how much power the resistor can dissipate without damaging (overheating) the resistor. Calculate the power being dissipated at the intended operating voltage. Do not assume that just because the resistor is being used under the maximum voltage rating that there will not be a problem. It is always advisable to calculate the power using P=V²/R, where V = test voltage and R = resistance value of the resistor, or P = I²*R, where I = current through the resistor.

The voltage coefficient expresses the change in resistance value due to a change in the amount of applied voltage. The voltage coefficient for a resistor is normally expressed in ppm/Volt and is always negative. This means that, the higher the applied voltage, the lower the resistance value.

Example Load Box Application

AC hipot tester is setup as follows:

Test Voltage = 1250VAC

High Limit = 5mA

Low Limit = Disabled

Ramp Time = 1s

Test Time = 1s

The example load box for this application should have two resistor values, one that will cause a PASS indication and the other a FAIL indication. The pass indication allows the operator to look for a known leakage current reading when the resistor is connected. If the current reading changes, then this could indicate that the programmed voltage has changed or the incorrect setup is being used. The pass indication also verifies that the meter is reading correctly.

Using Ohms Law, V=I*R, then 1250V/5mA = 250kohms. The resistor value for a pass indication should be slightly higher than 250kohms and the resistor for a fail indication should be slightly lower than 250kohms. Values of 200kohms for fail and 300kohms for pass in this case were chosen.

The required power dissipation = (1250V) 2/200,000ohms =  7.8W (Fail Condition)

The required power dissipation = (1250V) 2/300,000ohms =  5.2W (Pass Condition)

The recommended resistor for this application was the Vishay RS10, as the power rating is 10W. However, the voltage rating is 1000VAC and needs to be taken into consideration. To increase the voltage rating, the use of two 100kohm resistor in series would mean that 1/2 of the voltage is applied to each resistor. This configuration reduces the power requirements as well.

Figure 2: Series Configuration

When the hipot tester is connected between GND terminal and the PASS terminal, the hipot tester should indicate a leakage current of 1250V/300000ohms = 4.16mA. The operator would then look for this value of leakage current each time the routine operational check is performed. This check is normally performed at the beginning of every shift, or at another convenient time interval.

Calibration and Verification

Most modern digital electrical safety testers have a built-in calibration routine. Performing the calibration routine does not guarantee that the electrical safety tester meets specification. The calibration routine is simply designed to adjust calibration points within the tester to account for any differences between the measured value and the known standard.

The remainder of this section primarily discusses the verification of an electrical safety tester. Note that calibration is generally performed only if the unit is found to be out of specification.

Verification involves comparing the difference between displayed values with values from a known meter or standard, and manufacturer’s specification. The first step in verification is to understand the terminology and specifications of the electrical safety tester.

The reading is the measured and displayed value on the instrument. The accuracy is typically stated as a percentage of the reading shown on the meter’s display.

A count is the least significant digit that can be displayed. For example, if the accuracy specification is ± (1.5% of reading + 3counts) and the display resolution is 0.001mW, then the accuracy specification becomes ± (1.5% of reading + (3 * 0.001mW)) or ± (1.5% of reading + (0.003mW)). Manufacturers specify ‘counts’ or ‘% of range’ to account for noise and resolution at the low end of the instrument’s measurement range.

It is important to determine how many measurement ranges the instrument has, as each range may have a different display resolution. This affects the meaning of a ‘count’. In this example the hipot tester has two ranges. One range display resolution is 0.000mA and another 0.00mA. In the first case, 3 counts would be equal to 0.003mA; in the second case, it would equal 0.03mA.

Application of Verification

The specifications for a QuadTech Guardian 6000 Plus Electrical Safety Analyzer are used herein for the example tester. Although this particular analyzer has additional test modes, this example is only concerned with the AC hipot and ground bond specifications. In addition to verifying the AC voltage, current and ground bond specifications, the interlock function must be verified. A safety feature for the protection of the operator, the interlock function prohibits the output of high voltage when the interlock connector is not in place.

AC Specifications

AC Output Voltage Range: 50V to 5000V AC

Voltage Display Accuracy: ± (1% of reading + 5V)

Regulation: ± (1% of setting + 5V)

Current Display Accuracy: ± (1% of reading + 5 counts)

Ground Bond Specifications

Output Current: ± (1% of setting + 0.3A)

Resistance Accuracy: ± (1% of reading + 3 counts)

AC Voltage Verification

The output voltage range for AC hipot is from 50V to 5000V with a display accuracy of ± (1% of reading + 5V). In this example, the tester is programmed for various voltages and the output voltage is measured and compared with the voltage display accuracy specification. A high voltage voltmeter that has been calibrated traceable to NIST is used to measure the output voltage. The accuracy of the high voltage meter must also be determined. The calibration uncertainty (in this case, accuracy) of the high voltage meter should be 4 times better than the specification of the electrical safety tester. As an example, if the specification on the electrical safety tester is 1%, then the accuracy specification for the high voltage meter at the voltage being measured should be better than 0.25%.

Table 1 details the voltage setting, measured voltage from a high voltage calibrated voltmeter, displayed voltage or reading from the electrical safety tester, column for error in percent, and the calculated accuracy specification.

Voltage Setting (V)	Measured Voltage(V)	Displayed Voltage(V)	Error (1-(d/m))*100%	Specification
50	50.0	51	2.0%	10.8%
500	500.5	501	0.1%	2.0%
1000	1001.5	1000	-0.15%	1.5%
2000	2008	1999	-0.45%	1.25%
5000	5010	5000	-0.2%	1.1%

Table 1: AC Voltage Verification

Figure 3: Setup for Voltage Verification

The Specification column is obtained by taking the accuracy specification of ± (1% of reading + 5V) and determining the specification for each displayed voltage. The error for each measurement is shown in the Error column. The Error column is determined by dividing the displayed voltage by measured voltage, subtracting that number from 1 and displaying the result as a percentage. The Error column can then be compared with the Specification column. The error should always be less than the specification.

AC Current Verification

Electrical safety testers also monitor the leakage current through the device under test. During verification, voltage as well as current need to be measured and compared with readings from the hipot tester to determine if the unit meets specifications. The arrangement of this setup is illustrated in Figure 4. The electrical safety tester is programmed for 1200VAC and various resistive loads are attached in series with a calibrated current meter. The accuracy of the current meter should again be 4 times better than the specification of the electrical safety tester. The series resistor is a load resistor and does not need to be calibrated, but it does need to have sufficient power and voltage ratings.

Table 2, the current verification table, is completed and the error compared with specification for each load resistor. Note that the table has a column for high limit setting. The high limit setting changes the range in this electrical safety tester. When the range changes it changes the display resolution and effects the specification by changing counts. This phenomenon is illustrated in the displayed current column of Table 2. Note that, with the high limit set to 20mA, the displayed current resolution is 0.01mA; when the high limit is set to 3mA, the display current resolution is 0.001mA.

Load	High Limit Setting	Measured Current (mA)	Displayed Current (mA)	Error (1-(d/m))*100%	Specification
80kW/15mA	20mA	15.001	15.09	0.59%	1.3%
120kW/10mA	20mA	10.062	10.06	0.02%	1.5%
240kW /5mA	20mA	5.023	5.02	-0.06%	2.0%
480kW/2.5mA	3mA	2.501	2.502	0.04%	1.2%
5MW/0.24mA	3mA	0.2401	0.241	0.38%	3.1%

Table 2: AC Current Verification

Figure 4: Current Measurement

There are other considerations when performing current measurements. When measuring and comparing low current measurements, the electrical safety tester is likely to have some leakage current shown on the display without the load resistor and multimeter connected. It is recommended that measurements be made using a load resistor that give a leakage current much greater than this residual leakage current. If measurements are made very close to this residual leakage current, then offset and vector math need to be considered. The offset function of electrical safety testers compensates measurements for this residual leakage which is attributed to capacitive leakage current.

It is important to understand how this offset is performed. Most electrical safety testers measure total current. This total current is a vector summation of leakage current due to capacitive and resistive components, as shown in Figure 5.

Figure 5: Resistive and Capacitive Current

The electrical safety tester in our above example uses the following formulas for display of AC leakage current when an offset is applied. The example assumes that the AC offset value in the tester is set to 100uA (note however that this value is programmable).

These equations show that, for offset currents < 100uA, vector math is employed. Above 100uA, straight subtraction is employed to calculate the displayed current. During verification, since the internal leakage is capacitive and the leakage due to the load resistor is considered purely resistive, vector math should always be employed if the offset function is performed.

Ground Bond Verification

The ground bond function within an electrical safety tester applies an AC current of up to 40AAC and displays the resistance of the device under test (DUT). Resistance is displayed by measuring the voltage across the DUT, current through the DUT, and using Ohm’s Law to calculate the resistance. Verification involves measurement and comparison of the output current and various calibrated resistors to verify resistance measurement. Figure 6 illustrates the setup for ground bond verification.

Figure 6: Setup for Ground Bond Verification

Figure 7: 4-Wire Connection to DUT (click image for larger view)

Four terminals are used for ground bond measurements, in the method of connection called the 4-wire Kelvin connection. Implemented in most ground bond testers, the 4-wire (Kelvin) connection eliminates errors associated with test leads and the connection to the DUT. In this connection, 4 wires are connected between the meter and the DUT, as shown in Figure 7. One set of leads drives constant current and the second set of leads senses the voltage across the DUT.

In this configuration, the test current (ITEST) is forced through the DUT (RDUT) through one set of leads called drive, while the voltage across the DUT (VDUT) is measured by a second set of leads called sense. Although some small current may flow through the voltage leads, it is usually small enough to be ignored. Since the voltage drop across the voltmeter leads is negligible, the voltage across the meter can be considered the voltage across the DUT. In essence, the resistance of the DUT (RDUT) can be measured more accurately with the 4-wire method. The voltage sensing leads should be connected as close as possible to the DUT to avoid including the effects of the voltage drop across the test leads in the final measurement.

For ground bond verification, 4-wire resistors need to be used to eliminate lead and connection errors. Figure 8 shows a typical resistor or shunt with 4 wire connection. It is important that the shunt is calibrated using an AC signal not DC. Ground bond testers use a 50Hz or 60Hz signal during testing. The AC resistance being made up of the DC resistance and reactance will always be higher than the DC. The ground bond tester will always read high if compared with DC value of the resistors.

Figure 8: 4-Wire Resistor, courtesy of Deltec Company (www.deltecco.com)

Table 3, for ground bond verification, lists the displayed resistance and displayed current from the electrical safety tester, measured voltage from the multimeter, and a calculated current column. The calculated current is calculated by dividing the Measured Voltage by the Actual Value of Load. The displayed resistance, calculated current and displayed current are all verified against the specifications listed below the table.

Load	Actual Value mW	Displayed Resistance MW	Displayed Current (A)	Measured Voltage (V)	Calculated Current (A)
1mW	0.9982	1.1	30.00	0.0299	29.95
10mW	10.77	10.7	30.00	0.323	29.99
50mW	50.53	50.5	30.00	1.516	30.00
100mW	99.55	99.6	30.00	2.987	30.01
* Verify Displayed Resistance Value is within (1% + 3 counts)
* Verify Calculated Current is within (1% + 0.3A)
* Verify Displayed Current is within (1% + 0.3A)

Table 3: Ground Bond Verification

Summary

To make sure the electrical safety tester performs as required, routine operational checks should be performed on a daily basis. Scheduling routine operational checks at regular intervals should help prevent the possibility of shipping products that have not been tested. Verification of the electrical safety tester should be performed on an annual basis or similar interval to determine if the tester meets the manufacturer’s published specifications. Remember to consider all aspects when performing verification, from the calibration of the equipment and standards used, to how the tester performs measurements. Measurement ranges, counts, resolution and accuracy all contribute to the published specifications of an electrical safety tester. g

© 2007 Conformity

Top of Page