GE

Grid Solutions

UR Family
UR Relays Percent Differential Element Testing
Application Note
GE Publication Number: GET-20105A
Copyright © 2022 GE Multilin

This document outlines testing of the UR relays Percent Differential element and results. Although content
below is shown for the B90 relay for illustration purposes, it’s fully applicable to all UR platform percent
differential relays, B30, B90, G30, G60, M60, T35 and T60 having the same algorithm.

Introduction
The B90 Low Impedance Bus Differential System is part of the GE Universal Relay (UR) product family.
The B90 provides protection and metering for busbars with up to 24 feeders. It provides multi-zone
differential protection with both restrained (percent, biased) and unrestrained (unbiased, instantaneous)
functions incorporated, together with the phase comparison operating principle. Low and high
breakpoints are connected together with a cubic spline function to provide a smooth transition between
slope 1 and slope 2. If testing of points along this region is required, use the 87T Characteristics.xlsx
spreadsheet.

87T
Characteristics.xlsx

UR Family UR Relays Percent Differential Element Testing - Application Note 1

 Figure 1: Biased operating characteristics

The differential signal is composed of the sum of individual feeder currents.

I DIF = I1 + I 2 + ... + I n

The restraint quantity is chosen to be the maximum of the individual feeder currents.

I
RES
(
= max I , I , ..., I
1 2 n
)
The biased operation has two FlexLogic operands, BUS 1 BIASED PKP and BUS 1 BIASED OP. The pickup
(PKP) operand responds to the characteristic shape. The operate (OP) operand responds to the
characteristic shape AND the directional principle or the characteristic shape and NO saturation detected.
Figure 2: Biased pickup and operate

BUS 1 BIASED PK P

DIF1

DIR
BUS 1 BIASED OP
SAT

DIF2

The differential and restraint quantities can be plotted on the biased operating characteristic curve. Since
the relay also uses the directional principle, the biased trip may not occur when the two points intercept
the curve. To test the curve, assign the FlexLogic BUS 1 BIASED PKP operand to an output that is used for
sense on the test set.
The directional principle responds to the relative direction of the fault currents; therefore, no reference
signal is required. The directional principle declares that
• If at least one fault current flows in the same direction, compared with the sum of the remaining

2 UR Family UR Relays Percent Differential Element Testing - Application Note

 currents, the fault is internal, or
• If at least one fault current flows in an opposite direction compared with the sum of the remaining
currents, the fault is external
The current is determined to be a fault current if the magnitude is greater than an adaptable threshold
based on the restraint current and the number of feeders in the bus. The current is considered if its
magnitude is greater than K * Irest or 2 times the CT rating. For bus zones with two feeders K = 0.2, for bus
zones with three to six feeders K = 0.8 / (N - 1). For bus zones with more than six feeders K = 0.16. This is
important when testing the directional principle supervision.

Setup
The example in this document tests a B90 with six zones each with a 2000:5 CT on each CT input. The
following settings are used.
Figure 3: Bus differential settings

The View button (shown in red outline) at the top of the settings opens a live view of the element.

UR Family UR Relays Percent Differential Element Testing - Application Note 3

 Figure 4: Bus differential characteristic view

Zoom into regions of the graph by clicking and holding the left mouse button on the graph and expanding
the window.

4 UR Family UR Relays Percent Differential Element Testing - Application Note

 Figure 5: Zoomed-in bus differential characteristic view

Set the trip output or spare output wired to test set sense to BUS 1 BIASED PKP to test the shape of the
dual slope characteristic.
Figure 6: Output contact setting pickup

The current inputs have a continuous rating of 4 x Inom. Do not exceed 4 x Inom
current continuously during testing, else damage to the relay occurs.
The characteristic accuracy is ±3% of the maximum circuit current or ±2% of rated, whichever is greater.

Pickup test
The pickup is set at 0.2 pu, which makes the operate current 0.2 * 5 = 1 A. The per unit value is based on
the largest CT primary rating. In this case, because all the CTs are 2000:5, the pickup is 0.2 * 2000 on the
primary or 0.2 * 5 on the secondary.

UR Family UR Relays Percent Differential Element Testing - Application Note 5

To test the pickup value, use a pulsed linear ramp test starting with two equal currents 180° apart. Then
increase or decrease one of the currents until the test set receives a sense.
1. Set the initial values to 2 A on each source, one set to 0° and the other set to 180°. This results in zero
differential current.
2. Set the pulse duration to 50 ms with a wait time of 50 ms.
3. Set the delta current to -0.01 A with a current limit of 0 A.
4. Select current source 2 as your action source.
5. Start the test.
Figure 7: Pickup test pulsed ramp setup

View the test points live on the Operating Characteristic Graph View. The restraint current is the
maximum, 1 A in this case, which is 0.4 pu. The red X starts at (0.4,0) and rises to the differential
characteristic.

6 UR Family UR Relays Percent Differential Element Testing - Application Note

 Figure 8: Pickup test view

Figure 9: Pickup test result

Accuracy is ±2% of 5 A, or 0.100 A.

Slope 1 test
Slope 1 can be checked in a similar way because the restraint quantity is less than the continuous amp
limit of the relay.
The settings have a 25% slope with a low breakpoint of 2 pu. Choosing a restraint of 1.5 pu results in a
(required differential current of 1.5 * 0.25 = 0.375 pu. The current on source 2 must then be
(1.5 - 0.375) * 5 = 5.63 A.
To test slope 1, use a pulsed linear ramp test starting with equal currents 180° apart. Increase or decrease
one of the currents until the test set receives a sense.
1. Set the initial values to 7.5 A on each source, one set to 0° and the other set to 180°.

UR Family UR Relays Percent Differential Element Testing - Application Note 7

2. Set the pulse duration to 50 ms with a wait time of 50 ms.
3. Set the delta current to -0.01 A with a current limit of 0 A.
4. Select current source 2 as your action source.
5. Start the test.
Figure 10: Slope 1 test pulsed ramp setup

Figure 11: Slope 1 test view

8 UR Family UR Relays Percent Differential Element Testing - Application Note

 Figure 12: Slope 1 test results

Accuracy is ±3% of 7.5 A, or 0.225 A.

Slope 2 test
Slope 2 begins at breakpoint 2, which is set at 8.0 pu and the slope is 50%. Since 8.0 pu is above the 4.0 pu
continuous rating of the input, a test cannot hold the restraint current at 40 A. Test set source limitations
also can come into play, requiring the user to parallel sources to get the higher magnitude currents.
For slope 2, the restraint quantity needs to be 8 pu or 40 A. The test requires using two paralleled sources
each set at half of the test value. The differential current needs to be 50%, so
Source 2 = 8 (1 - 0.5) * 5 = 20 A.
To test slope 2, use a pulsed linear ramp with three sources. Fix one source at 180° and the other two at
0°, then increase their magnitudes until the test set receives a sense.
1. Parallel sources 1 and 3 going into the relay.
2. Set the initial values to 10 A on sources 1 and 3, set to 0°, and source 2 set to 20 A and 180°.
3. Set the pulse duration to 50 ms with a wait time of 50 ms.
4. Set the delta current to 0.1 A with a current limit of 25 A.
5. Select current sources 1 and 3 as your action sources.
6. Start the test.
Figure 13: Slope 2 pulsed ramp setup

UR Family UR Relays Percent Differential Element Testing - Application Note 9

 Figure 14: Slope 2 test view

Figure 15: Slope 2 test results

Accuracy is ±3% of -0 A, or -1.20 A, but because we have current from two sources it is half, or 0.600 A.

Directional principle check

Now that the characteristic has been verified, the trip operation can be verified quickly by injecting
currents that meet the operate conditions.
First, change the output operate being used for test set sense to BUS 1 BIASED OP.

10 UR Family UR Relays Percent Differential Element Testing - Application Note

 Figure 16: Output contact setting OP

If a restraint value of 15 A or 3 pu is chosen, because there are six feeders in this relay’s bus zone, K = 0.16
so source 2 current is 15 * 0.16 = 2.4 A. Use the pulsed ramp test for this.

In firmware released after April 2021, the B90 uses an extra protection security enhancement that
requires the relay to detect an actual fault disturbance to operate. The fault detection algorithm
uses a rate of change of current algorithm to distinguish a fault from non-fault conditions.

1. Set the initial values to 15 A on source 1, set to 0°, and the other set to 180°.
2. Set the pulse duration to 50 ms with a wait time of 50 ms.
3. Set the delta current to -0.05 A with a current limit of 0 A.
4. Select current source 2 as your action source with an initial value of 5A.
5. Start the test.
Figure 17: Directional principle check settings

The relay operates when the source 2 current gets to 2.4 A.

UR Family UR Relays Percent Differential Element Testing - Application Note 11

 Figure 18: Directional principle test view

Figure 19: Directional principle test results

Accuracy is ±3% of 15 A, or 0.450 A.

12 UR Family UR Relays Percent Differential Element Testing - Application Note

For further assistance
For product support, contact the information and call center as follows:
GE Grid Solutions
650 Markland Street
Markham, Ontario
Canada L6C 0M1
Worldwide telephone: +1 905 927 7070
Europe/Middle East/Africa telephone: +34 94 485 88 54
North America toll-free: 1 800 547 8629
Fax: +1 905 927 5098
Worldwide e-mail: multilin.tech@ge.com
Europe e-mail: multilin.tech.euro@ge.com
Website: http://www.gegridsolutions.com/multilin/

UR Family UR Relays Percent Differential Element Testing - Application Note 13