Look at the first cycles of a fault or switching
What it tells you. An electromagnetic transient (EMT) simulation shows the voltage and current of each phase, sample by sample. It covers the first cycles after a fault or a switching operation. Use it to see the peak of a fault current, the voltage peaks a switching operation causes, and how an unbalanced fault looks.
Warning. EMT is not a grid code compliance tool. Every source keeps the voltage it had in the load flow, no rotor moves and no control acts. The results mean something only for the first cycles. Use the stability simulation for swings over seconds.
Before you start
Check that the network has the data below. Select an element to see its data in the inspector.
| Data | Where to enter it |
|---|---|
| Complete load flow data, because the simulation starts from the load flow | See Run a load flow |
| Machine Subtransient reactance xd″ and Stator resistance | Inspector, Short circuit |
| External grid Short-circuit power max, R/X ratio max, X0/X1 ratio and R0/X0 ratio | Inspector, Short circuit |
| Line Length, Reactance X′ and Susceptance B′, which set how a line behaves in time | Inspector, Basic data |
| For unbalanced faults: line Zero-sequence R0′, Zero-sequence X0′ and Zero-sequence B0′ | Inspector, Short circuit |
| For unbalanced faults: transformer Vector group, Zero-sequence uk0 and neutral earthing | Inspector, Basic data and Short circuit |
| The events: faults, clearings and switching, with their times | EMT events dialog, described below |
| Simulation time, Time step and Line model | Study case, under EMT simulation |
A line without zero-sequence data takes its positive-sequence values, and the report says so. Machines are unearthed. An external grid is unearthed when its X0/X1 ratio is 0. See zero-sequence data.
Set up the study
Each event has a time in milliseconds, a type and an element.
- Select the Calculate tab.
- Under Dynamics, select EMT events. The Study case dialog opens at EMT events.
- Select Add EMT event.
- In Time (ms), enter when the event happens.
- In Event, choose Fault at, Clear fault at, Trip or Close.
- In Element, choose a busbar for a fault or a clearing. Choose a line, transformer, machine, external grid, load or shunt for a trip or a close.
- For a fault, choose its type in Fault. The types are Three-phase, Line to line, Two phases to earth and Line to earth. Enter a fault resistance in R Ω, or leave it empty for a bolted fault.
- Repeat steps 3 to 7 for each event.
- Select Apply.
Events act at the nearest time step. The 4 event types behave as follows.
- Fault at connects the phases of the busbar through the resistance you gave.
- Clear fault at puts each faulted phase out at its next current zero.
- Trip opens each pole of the element's breakers at its next current zero. A branch opens at both ends.
- Close closes all 3 poles at the time you give. That time sets the point on the wave. You can close in a line, transformer, load or shunt that is out of service.
To add a fault quickly, right-click a busbar and select Add fault to EMT simulation. This adds a line-to-earth fault after 1 cycle and clears it 5 cycles later.
To change the other settings, press Ctrl+, to open Study case, then scroll to EMT simulation.
| Setting | What it does |
|---|---|
| Simulation time | Sets how long to simulate. The default is 0.2 s. |
| Time step | Sets the step in microseconds. The default is 50 µs. Shorten it for travelling waves on short lines and cables. |
| Line model | Chooses Travelling waves where they fit the step, which is the default, or π sections everywhere. |
The method is Dommel's method. See the engine description.
Run it
- On the Calculate tab, select EMT, or press Alt+W.
- Wait for the progress bar in the status bar to finish.
The EMT tab opens in the results panel. The Output tab logs the highest voltage and the peak fault current. It also warns about any event that had no effect.
To stop a run, press Ctrl+.
Read the results
The top of the EMT tab shows a pill and a summary. The pill gives the time simulated, such as 200 ms simulated. The summary gives the Step, the Highest voltage in p.u. of crest, the Peak fault current and the Time.
Use the buttons on the right of the tab to switch between 6 views.
| View | What it shows |
|---|---|
| Waveforms | The 3 phases of a voltage or current against time |
| Busbars | Busbar, Un, Peak voltage, Peak as a share of the nominal crest, At, Lowest RMS and Highest RMS |
| Branches | Branch, Model, Peak current and At. The model is Travelling wave, π section or Transformer. |
| Sources | Source, Peak current and At |
| Faults | Busbar, Fault, From, Out, Peak current, At and RMS, last cycle |
| Switching | Time, Kind and What happened, for each event and each pole that opened |
Times in the tables are in milliseconds. The Sources and Faults buttons appear only when the run has sources or faults.
Read a waveform
- In Waveforms, select a busbar or branch on the diagram. Its phases are plotted.
- To plot something else, choose a quantity in Quantity plotted. The choices are Busbar voltage, Branch current, Source current and Fault current.
- To hide a phase, select its Phase button.
- To zoom in, drag across a time range, or select Zoom in on time.
- To see the whole run again, double-click the plot, or select Show the whole run.
Read the numbers
The crest is the nominal peak of the phase-to-earth voltage. A Peak of 1.0 p.u. is the normal peak. The RMS value is taken over 1 cycle, after the first cycle.
The Busbars table highlights a Peak above 1.2 p.u. in amber and above 1.5 p.u. in red. Its filters are Peak above 1.2 p.u. and RMS below 0.9 p.u.
A fault current is the sum of a steady part and a decaying DC offset. The Peak current includes the offset. RMS, last cycle is the RMS of the largest phase over the last full cycle before the fault went out.
Check your results
- Flat before the first event. Voltages should stay at their load-flow values until the first event.
- Fault RMS. Compare RMS, last cycle with Ik″ from the short circuit study. They should be close but not equal. The short circuit study uses a voltage factor, and the EMT study starts from the load-flow voltages.
- Fault peak. The peak should lie between 1.4 and 2.8 times the RMS current. The ratio depends on the DC offset.
- Halve the time step. Run it again with half the Time step. If a result moves a lot, it is not converged.
- Line model. A line whose travel time is below the step is a π section. The Branches view shows which model each line has.
- Core saturation. The model has none, so do not trust a transformer's behaviour near its saturation.
Warning. Switching overvoltages can depend on the step. Check them with a shorter Time step before you quote a peak.
If something goes wrong
| What you see | What it means | What to do |
|---|---|---|
| Event, no effect in Switching, or a warning in Output | An event did nothing, for example a clearing with no fault | Check the element and time of the event in EMT events. |
| A warning that a pole did not reach a current zero | A DC offset kept the current from crossing zero, so the pole stayed closed | Raise Simulation time, or move the event. |
| Out shows to the end | The fault has no Clear fault at event | Add a clearing event. |
| EMT simulation cannot run, and a message about a line | A line joins busbars with different nominal voltages | Replace the line with a transformer. |
| EMT simulation failed, and a message about a zigzag winding | The EMT model does not support zigzag windings | Change the transformer's Vector group for this study. |
| An event never happens | Its time is after the Simulation time | Move the event, or raise Simulation time in Study case. |
| The pill Calculated before the last edit | You changed the network after the run | Run EMT again. |
Try it on the IEEE 14-bus sample
- Select the Help tab.
- Under Samples, select IEEE 14-bus system.
- Select EMT events. The sample has 3 events.
- Read them. A three-phase Fault at Bus 4 is at 20 ms. Clear fault at Bus 4 and Trip Line 4-5 are both at 100 ms.
- Select Cancel, then select EMT.
- Read the summary. The Highest voltage is 1.74 p.u. at Bus 9. The Peak fault current is 12.80 kA.
- In Waveforms, select Bus 4 on the diagram. The 3 voltages fall to 0 at 20 ms.
- Select Faults. The fault lasts from 20.00 ms to 108.45 ms. The Peak current is 12.802 kA at 27.55 ms. RMS, last cycle is 5.865 kA.
- Select Switching. The 12 rows show the 3 events and 9 poles opening. Each phase of the fault and of Line 4-5 opens at its own current zero, between 102.65 ms and 108.45 ms.
- Select Busbars. Bus 9 has the highest Peak at 1.742 p.u., at 105.05 ms. That is after clearing starts, not during the fault.
Check the numbers against the short circuit study. For a three-phase fault at Bus 4, Ik″ is 6.157 kA and ip is 13.650 kA. The EMT values are about 5 % and 6 % lower.
Try other fault types
Change the Fault of the first event to see the peak current at Bus 4.
| Fault | Peak current | RMS, last cycle |
|---|---|---|
| Three-phase | 12.802 kA | 5.865 kA |
| Line to line | 11.290 kA | 5.124 kA |
| Two phases to earth | 12.048 kA | 5.709 kA |
| Line to earth | 3.640 kA | 2.481 kA |
The earth fault is small because the sample's machines are unearthed.
Try it on the Riverside sample
- Under Samples, select Riverside distribution.
- Select EMT events. The sample has a Line to earth fault at Hilltop at 20 ms and its clearing at 80 ms. It trips the Capacitor bank at 120 ms and closes it at 160.1 ms.
- Select Cancel, then select EMT.
- Select Busbars, then select Peak above 1.2 p.u. The list has one row: Mill Lane at 1.236 p.u., at 160.40 ms. That is 0.3 ms after the capacitor bank closes.
- Select Faults. The Peak current is 6.210 kA at 33.75 ms. RMS, last cycle is 4.222 kA.
The RMS value lies between the minimum and maximum Ik″ of a line-to-earth fault at Hilltop: 3.823 kA and 4.450 kA.
Now check the step. Open Study case and set Time step to 25 µs, then run it again. The fault peak stays at 6.210 kA. The highest voltage rises to 1.38 p.u. at Station Road. At 10 µs it is 1.63 p.u.
The cables here are short, so a coarse step treats them as π sections. A shorter step models them as travelling waves, and the switching peak changes. This is why you should check a switching overvoltage against a shorter step.