D5 - Report on standardised methods for maintaining and periodically evaluating the synchronicity of substation / calibration equipment both at the level 1 s

Confidentiality Status: PU - Public, fully open Deliverable Cover Sheet Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or EURAMET. Neither the European Union nor the granting authority can be held responsible for them. The project has received funding from the European Partnership on Metrology, co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States. 1 of 11 21NRM02 - Digital-IT D5 - Report on standardised methods for maintaining and periodically evaluating the synchronicity of substation / calibration equipment both at the level < 1 s and => 1 s Organisation name of the lead participant for the deliverable: JV Due date of the deliverable: April 2025 Actual submission date of the deliverable: December 2025 2 of 11 Glossary TERM DEFINITION 1PPS One Pulse Per Second – A signal used for precise time synchronization, often generated by GPS receivers. ASDU Application Service Data Unit – A data structure used in communication protocols to encapsulate application-level data. BMCA Best Master Clock Algorithm CALIBRATION The process of adjusting and verifying the accuracy of a measurement instrument by comparing it to a known standard. DCA Direct Current Ammeter – An instrument used to measure direct current in an electrical circuit. DMM Digital Multimeter – An electronic measuring instrument that combines several measurement functions in one unit. DUT Device Under Test – The equipment or component being tested in a laboratory or field setup. GNSS/GPS Global Navigation Satellite Systems – a common source of precision timing. GNSS include GPS, Galileo, GLONASS and Beidou. GRANDMASTER CLOCK The primary source in a PTP network that distributes accurate time to all other devices. IED Intelligent Electronic Device – A microprocessorbased controller used in substations for protection, control, and monitoring. LPIT Low Power Instrument Transformer – A type of transformer that provides low power analog signals suitable for digital processing. PHASE DISPLACEMENT The angular difference between two waveforms, typically voltage and current, in an AC system. POWER UTILITY PROFILE A configuration of PTP designed specifically for power utility applications to ensure accurate time synchronization. PTP Precision Time Protocol – A protocol used to synchronize clocks throughout a computer network, defined in IEEE 1588. PRP Parallel Redundancy Protocol RMS Root Mean Square – A statistical measure of the magnitude of a varying quantity, commonly used in AC voltage and current measurements. SAMU Stand-Alone Merging Unit – Converts analog signals from conventional transformers into digital Sampled Values (SV) for IEC 61850 communication. SAMPLE RATE The number of samples taken per second from a continuous signal to convert it into a digital signal. SAMPLED COUNT The number of samples collected over a specific period or during a measurement process. SAMPLED VALUE A digital representation of an analog signal obtained by sampling at regular intervals. SHUNTS Lowresistance electrical components are used to measure current by producing a voltage drop proportional to the current flow. TESTBED A test setup or environment used for validating and verifying the performance of electrical devices or systems. TIC Time interval counter UNCERTAINTIES The range of possible errors in a measurement, indicating the confidence level of the result. 21NRM02 Digital -IT 3 of 11 TABLE OF CONTENTS 1Summary ................................................................................................................................. 4 2 Methods for maintaining and monitoring synchronicity < 1s ..................................................... 4 2.1 General description of digital substation timing .................................................................. 4 2.2 Example implementation of timing monitoring .................................................................... 5 2.3 Example of timing monitoring ............................................................................................. 7 3 Methods for evaluating substation synchronicity at => 1 s (leap seconds) ................................ 7 3.1 General .............................................................................................................................. 7 3.2 Example of GNSS simulator method implementation ......................................................... 9 3.3 Example of leap second performance testing ................................................................... 10 4Conclusions ........................................................................................................................... 11 21NRM02 Digital -IT 4 of 11 1 Summary Among other topics, the EURAMET Digital-IT (Metrology for digital substations) project set out to develop traceable methods for maintaining, monitoring and calibrating time keeping instrumentation in digital substations using IEEE 1588 Precision Time Protocol. No such service was to date available from any national metrology institute globally. The problem has so far been the somewhat cross-disciplinary nature of the problem, the electricity grid metrologists are not timing experts, and time and frequency researchers mostly focus on time scales and frequency transfer. The grey area between the disciplines had so far been unexplored. But timing is becoming increasingly important in operating electricity grids and the need for calibrating PTP timing in the context of digital substations has become important. Whereas latency measurements and PTP timestamp accuracy usually focus on the sub-millisecond level, timestamps can have offsets of integer numbers of seconds. Often, these offsets are related to incorrect handling of leap seconds. Therefore, methods for correctly inserting leap seconds in equipment synchronised via PTP were studied. Also, methods for ensuring synchronicity of devices that are relevant to time distribution and protection system operation in a digital substation or in calibration setups for digital instrument transformers were developed. These include methods for the insertion of leap seconds and methods for periodic and/or continuous evaluation of synchronicity of substation equipment both at the level below 1 s and above 1 s. 2 Methods for maintaining and monitoring synchronicity < 1s 2.1 General description of digital substation timing Figure 1 shows an example digital substation timing setup with two master timeservers or station clocks. The timeservers act as PTP Grandmasters (GM) on the substation LANs. Shown in this example are redundant process LANs based on the parallel redundancy protocol (PRP). IEDs designed for use on redundant process LANs are doubly connected, i.e. have two separate network ports, one for each LAN and send streams of sampled values over both LANs. In this particular example of timing provision there are individual PTP GM cards singly connected to each LAN. Other implementations of PTP over PRP may use mechanisms defined in IEC 62439-31to allow PTP packets from a single PTP GM to be copied and transmitted over both LAN A and LAN B, in effect making a single PTP GM doubly connected. Further discussion on the relative merits of singly vs doubly connected PTP GMs is beyond the scope of this document. The timeservers as shown are configured to use multiple external timing references, including GPS/GNSS, PTP timing over wide-area network, as well as a local Rubidium clock for holdover in case of failure of external sources. A built-in multi-source measurement system measures and logs external clock sources against the timeserver’s internal time scale and also provides input into the external sync source selection algorithm.2 The master station clocks are identically configured, with one important exception: The priority2 field announced by the PTP GM cards are set so that the Best Master Clock Algorithm (BMCA) of PTP slaves will preferentially select timing from PTP GM cards hosted in one particular timeserver, designated the active station clock (labelled SC_A in Figure 1). In effect, the active station clock will be the sync source for all PTP slaves’ clocks on all LANs under normal operation. Only when the active station clock becomes unavailable or if other parameters used by the BMCA (such as clock quality) change, will the backup station clock be selected by the BMCA of the PTP slaves. A benefit of this choice is that configuration changes and updates that require a reboot may first be implemented and verified on the backup station clock without imposing any timing change on PTP slave devices. Once verified on the backup station clock, similar changes can be made on the designated active station clock. 1 See e.g. https://blog.meinbergglobal.com/2021/05/27/implementing-prp-devices-and-networks/ 2 H. Hauglin, T. Dunker, A. Wallin, O. Tungland, N. Hurzuk, R. Løken, Redundant secure timing sources and timing distribution to digital power protection and control applications, Cigre Study Committee B5 Colloqium 2019, Cigre Science and Engineering, vol 17 (2020) pp 94-100. https://e-cigre.org/publication/CSE017-cse-017 21NRM02 Digital -IT 5 of 11 Figure 1. Typical setup in digital substation with two clocks. 2.2 Example implementation of timing monitoring Substation timing monitoring may serve multiple objectives: (i) Internal consistency, i.e. to verify the consistency within the substation, e.g. between IED timing on LANs and station clock timescales. (ii) External accuracy and traceability between the station clock timescales and external reference timing sources traceable to UTC. (iii) Periodic monitoring during initial commissioning/verification and/or subsequent recalibration. (iv) Continuous monitoring for running verification and fault detection as part of routine operation. Periodic/external monitoring of the accuracy/traceability of GNSS/GPS based timing sources may follow guidelines by EURAMET3. Further discussion is beyond the scope of this document. Continuous/external monitoring of multiple external sources, including outlier detection and clock combiner/source selection functions are currently being standardized by ITU WG15 in e.g. ITU Figure 2 shows an example setup for periodic/internal monitoring of substation timing using a separate time interval counter (TIC). For monitoring purposes, PTP slave timing devices with physical 1PPS outputs can be installed on the substation LANs. In a PRP context, the monitoring PTP slaves are installed as singly connected devices transmitting and receiving PTP packets only over a particular LAN. Timing inputs to the TIC is a 1PPS output of the master station clock as reference and 1PPS output from one of the PTP slave cards. The TIC also needs an accurate 10 MHz reference from the station clock in order to measure time intervals accurately (not shown). The measured time interval between master reference 1PPS and slave 1PPS documents the internal consistency between reference timing of the PTP GM and PTP slave timing on a particular LAN. Conventional stand-alone TICs have two input channels, therefore several TICs may be needed to monitor 3 EURAMET TC-TF, “Guidelines on the Use of GPS Disciplined Oscillators for Frequency or Time Traceability, Technical Guide No. 3, Version 1.0 (2016)”. Link: https://www.euramet.org/publications-media-centre/technical-guides 21NRM02 Digital -IT 6 of 11 multiple timing outputs simultaneously. Multi input timetaggers are commercially available for convenient monitoring of more timing outputs. An example setup for continuous internal monitoring is shown in Figure 3. Here the technical implementation makes use of the station clock’s multi-input measurement capabilities, in particular the use of a dedicated multi-channel pulse measurement module installed in the timeserver chassis (labelled ‘IMS PIO’ in the figure). Data from the pulse measurement card in this example was logged into daily files to be manually downloaded for off-line analysis. Data may also be automatically retrieved in real time over the station clock’s management port for use in live display or fault detection applications. Figure 2. Example implementation of an external measurement system (time interval counter) for periodically evaluating process LAN timing. Figure 3. Example implementation of continuous timing monitoring using a multi-channel measurement system integrated in the station clock (timeserver). IMS PIO is a multi-input pulse measurement card installed in the timeserver chassis. 21NRM02 Digital -IT 7 of 11 2.3 Example of timing monitoring Figure 4 shows the measurement of the time differences between LAN A and LAN B from the grand master clock PPS signal in Figure 2 setup. Here the time differences between the station clock PTP GM and the PTP slave timing is recorded during a forced switchover of GM selected by the BMCA in the PTP slave. Switchover is forced by unplugging the network interfaces that connect the ‘active’ station clock (‘SC_A’) to the substation LANs. Figure 4. Example of process bus PTP timing output during forced change of master clock. The plot shows the time scale difference between the reference master clock and the 1PPS output of PTP slave clocks on PRP process LAN A (red) and B (blue) during forced change of selected best master clock. The labels ‘SC_A’ and ‘SC_B’ indicate the selected PTP GM. 3 Methods for evaluating substation synchronicity at => 1 s (leap seconds) 3.1 General Since the start of the project, the probability of needing an added leap second has decreased, due to Earth rotation speeding up significantly4. In fact, the prospect of an unprecedented negative leap second has become a concern. The leap second mechanism is set to be changed before 2035 by increasing the maximum allowable tolerance between solar time UT1 and UTC from 0.9 s to an amount yet to be decided5. Until then the current leap second convention is in force for another decade, and it is therefore relevant to test substation timing and timing dependent systems for correct implementation leap seconds. There are in principle (at least) three methods for injecting leap second information into master station clocks, as shown in Figure 5. This is relevant for actual leap second events as well as for testing purposes. L1. Leap second information through GNSS signals. Leap second information are part of navigation data timing parameters being transmitted by GNSS satellites6. Navigation data timing parameters enable GNSS timing receivers to correct for fine and coarse offsets between GNSS system time scales and UTC. GNSS systems such as GPS and Galileo operate internally with a continuous system time scale (i.e. without leap seconds). Transmitted navigation data contain status information on current and pending future leap seconds. L2. Leap second file uploaded locally to the station clocks. L3. Leap second information from an upstream network time server over protocols such as NTP or PTP. For testing leap second updates in an integrated substation system, one challenge is to introduce leap second information for testing into the master station clocks while maintaining sync status at a level that keeps PTP clock quality parameters used by BMCA unchanged. 4 See UT1 – UT1 data in Bulletin A from the International Earth Rotation and Reference Systems Service. Link: https://datacenter.iers.org/singlePlot.php?plotname=BulletinA_All-UT1-UTC&id=6 5 Resolution 4 of the 27th meeting of the General Conference of Weights and Measures CGPM (2022). Link: https://www.bipm.org/en/-/resolution-cgpm-27-4 6 See e.g. section 5.1.7 of the Galileo Open Service Signal-in-Space Interface Control Document (OS SIS ICD), issue 2.2 (2025). Link: https://www.gsc-europa.eu/sites/default/files/sites/all/files/Galileo_OS_SIS_ICD_v2.2.pdf 21NRM02 Digital -IT 8 of 11 Another challenge is to have all sources of leap second information being consistent. Simply uploading a leap second file to a time server (as in option L2) which is using GNSS as an external timing input may impose conflicting leap second information. Disconnecting the GNSS antenna to avoid this conflict may cause the time server to report loss of sync, which in itself may affect the performance of sensors and protection/control applications downstream relying on PTP slave timing. Taking leap second information flow from an upstream time server (option L3) only shifts these issues upstream. For testing one should also keep in mind that a leap second should be the last second of a UTC month, with preference for end of June and December and second preference for end of March and September7. Timing devices in a substation system may or may not have incorporated this information as part of data integrity checks. Using option L1 for testing purposes involves feeding a time server either pre-recorded GNSS signals from a past actual leap second event or simulated GNSS signals that contain information about an (actual or fictitious) upcoming leap second. In either case, care must be taken to avoid conflicting leap second information for the test to be realistic. A time server may for instance already have stored leap second information that is dated after a replayed leap second event, and therefore discards the new but ‘outdated’ information. For replay of recorded GNSS signals from the last actual leap second on the 31st of December 2016, another concern is that newer GNSS receiver chips may not process old signals correctly, including handling of GPS week rollover. Figure 5. Potential leap second information sources for a multisource substation master clock 7 Recommendation ITU-R TF.460-6 Standard-frequency and time-signal emissions. Link: https://www.itu.int/rec/R-RECTF.460/ 21NRM02 Digital -IT 9 of 11 3.2 Example of GNSS simulator method implementation An example implementation of option L1 is shown in Figure 6 and Figure 7. Project partner JV has experience using GNSS simulators for testing the resilience of GNSS timing devices against GNSS signal interference such as jamming and spoofing. Adaptations were made to use the GNSS simulator setup for on-premises leap second testing in a digital substation in full operation. The GNSS simulator is based on commercial simulator software running on a laptop with a powerful GPU (graphics card) and streaming simulated GNSS signals to a commercial software defined radio (SDR). Additional adaptations were made to have a mobile setup for generating stable timing to the GNSS simulator. The GNSS simulator is inserted in-line with the GNSS antenna feed installed for the master station clocks. By suitable use of RF signal splitters and combiners the simulated GNSS signals may replace or be mixed with the actual live sky GNSS signals on one or both of the station clocks. The setup in Figure 7 shows how the simulated GNSS signals may replace the actual live sky GNSS signals without loss of sync. In this case the simulator timing has the accuracy/stability of a GNSS disciplined Rboscillator. The GNSS simulator may furthermore use updated satellite data relevant for the actual satellites visible. With suitable calibration and correction for reference timing and simulation/transmission delays, the simulated GNSS signals may be aligned with the actual GNSS signals to within approximately 100 ns. In this way, the simulated GNSS signals can replace the actual GNSS signals without loss of sync, but now with full simulator control over GNSS satellite data and signals, including leap second timing parameters. Beyond the scope of leap second testing in Digital-IT, GNSS signal interference in the form of jamming and spoofing has significantly gained attention since the start of the project in 2021. This is another concern relevant to substation synchronicity and the reliable operation of substation measurement, control and protection applications. The simulator software is capable of simulating ‘true’ GNSS signals (in sync with actual live sky signals) in combination with interference due to jamming and/or spoofing interference. Figure 6. Substation timing system overview with a GNSS simulator connected in-line with the substation GPS/GNSS antenna feed.