IEC 61850 Sampled Values (SV) Explained: Complete Process Bus Reference

A complete reference for substation engineers covering Sampled Values — IEC 61850-9-2 specification, frame structure, ASDU/APDU, SmpCnt and SmpSynch, ASN.1 BER encoding, multicast addressing, IEC 62351-6 security, PRP/HSR transport, and field troubleshooting

Quick Reference

Item	Value
Standard	IEC 61850-9-2:2011 (Edition 2.0) + IEC 61869-9:2016 profile
Transport	Raw Ethernet (Layer 2), no TCP/IP
EtherType	0x88BA
Multicast MAC range	01-0C-CD-04-00-00 to 01-0C-CD-04-01-FF
APPID range	0x4000–0x7FFF (default 0x4000 = “no config”)
Default VLAN priority	4 (per IEEE 802.1Q)
Encoding	ASN.1 Basic Encoding Rules (BER)
Max APDU length	1493 octets
Publisher	Merging Unit
Subscriber	Protection IED, meter, fault recorder
Service models	Multicast SV (MSVCB) + Unicast SV (USVCB)
Time sync	PTP per IEC/IEEE 61850-9-3 (preferred) or 1PPS
Security	IEC 62351-6 (optional, 28-bit reserved security field)

Introduction

Sampled Values are one of three core protocols of IEC 61850, alongside GOOSE and MMS. While MMS carries SCADA-level configuration and reporting over TCP/IP, and GOOSE carries fast binary events at Layer 2, SV carries continuous high-rate analog measurements over Layer 2 Ethernet.

If GOOSE replaces binary control wiring, Sampled Values replace CT/VT analog wiring.

This article is the complete reference, verified against the IEC 61850-9-2:2011 Edition 2.0 specification and the IEC 61869-9:2016 profile. It covers frame structure, encoding rules, addressing, synchronization, security, network engineering, and field troubleshooting.

1. What Are Sampled Values?

Sampled Values (SV) are real-time digital current and voltage measurements published as Ethernet multicast frames per IEC 61850-9-2. A Merging Unit digitizes CT and VT signals at a fixed sample rate, time-stamps each sample, and continuously streams them on the process bus where any protection IED, meter, or fault recorder can subscribe. SV replaces analog CT/VT wiring with deterministic, time-aligned digital communication.

2. Why Sampled Values Exist

Before SV, substations relied on analog CT/VT wiring — often hundreds of meters of copper from the switchyard to the control room. This created multiple problems:

• Analog signal distortion over distance
• Electromagnetic interference on long cable runs
• High copper cost
• CT burden and saturation issues
• Safety risks during maintenance (open CT secondary)
• One CT/VT could feed only one relay; sharing required wiring duplication

With Sampled Values:

1. A Merging Unit (MU) digitizes the signal at the switchyard
2. Samples travel through fiber/Ethernet, not copper
3. Multiple IEDs subscribe to the same SV stream
4. No analog distortion or noise once digitized
5. Time-aligned samples enable cross-bay protection (differential, distance)

This architecture — called the IEC 61850 Process Bus — is the backbone of digital substations.

For the device producing SV streams, see: What Is a Merging Unit?

3. SV vs GOOSE vs MMS — When to Use Each

The three protocols in IEC 61850 serve different purposes:

Aspect	MMS	GOOSE	Sampled Values
Purpose	Client/server, SCADA, config	Fast binary events	Continuous analog samples
Transport	TCP/IP (Layer 5–7)	Layer 2 Ethernet	Layer 2 Ethernet
EtherType	(uses TCP port 102)	0x88B8	0x88BA
Data type	Any (structured)	Binary status, integers	Analog samples + quality
Behavior	Request/response	Event-driven multicast	Continuous streaming
Speed	10–1000 ms	1–4 ms	µs-level timing
Bandwidth	Low	Low	High (5–18 Mbit/s per stream)
Typical source	IED	Protection relay	Merging Unit
Replaces what?	Configuration UI	Binary copper wiring	CT/VT analog wiring
Standard	IEC 61850-8-1	IEC 61850-8-1	IEC 61850-9-2

In practice, all three operate simultaneously in a digital substation:

• MMS for HMI, engineering, supervisory monitoring
• GOOSE for trips, interlocks, position changes
• SV for the continuous measurement stream that feeds protection algorithms

4. The IEC 61850 Stack — Where SV Fits

IEC 61850-9-2 defines an SV-specific communication profile using direct Layer 2 Ethernet. There is no IP, no TCP, no UDP in the SV path.

SV A-Profile (Application + Presentation Layers)

Layer	Specification
Application	IEC 61850-9-2 SV service
Presentation	ASN.1 BER (ISO/IEC 8824-1, 8825-1)
Session	(Not used for SV)

SV T-Profile (Transport to Physical Layers)

Layer	Specification
Transport, Network	(Not used — SV bypasses these)
Link Redundancy	IEC 62439-3 (PRP / HSR) — optional
Data Link	IEEE 802.1Q (Priority tagging/VLAN) + ISO/IEC 8802-3 (CSMA/CD)
Physical	100Base-FX fiber optic (recommended)

This minimal stack is why SV is fast: no protocol overhead between the application and the wire.

Client/Server Access to SV Control Blocks

While the SV stream itself bypasses IP, management of the SV Control Block (enable/disable, configuration read/write) uses standard IEC 61850-8-1 MMS over TCP/IP. This is the only place where TCP/IP is involved with SV — and even that is optional (the SVCB may be pre-configured).

5. The SV Communication Model (Publisher/Subscriber)

SV uses a publisher/subscriber model — not client/server. The publisher (typically a Merging Unit) continuously broadcasts SV frames to a multicast MAC address. Any device interested subscribes by receiving multicast traffic from that address.

Key Characteristics

1. Continuous streaming — SV never stops sending, even with no fault
2. One-to-many — a single stream feeds protection, backup, fault recorder, meter simultaneously
3. No acknowledgment — publisher doesn’t know who’s listening
4. Stateless — each frame is self-contained
5. Deterministic — frames arrive at fixed intervals

Typical Subscribers

• Primary protection relay
• Backup protection relay
• Bay controller
• Disturbance/fault recorder
• Power quality meter
• Wide-area monitoring system (PMU)
• Test equipment / simulators

All receive the same data simultaneously without any per-subscriber configuration on the publisher side.

6. Multicast (MSVCB) vs Unicast (USVCB) SV

IEC 61850-9-2 defines two SV service models:

Aspect	Multicast SV (MSVCB)	Unicast SV (USVCB)
Destination	Multicast MAC address	Specific subscriber’s MAC
Subscribers	One-to-many	One-to-one
Reservation	Open to all	Exclusive (Resv flag)
Common use	Process bus measurement distribution	Specific point-to-point flow
Field deployment	Standard	Rare

Multicast (MSVCB) is the dominant model in real-world substations. The unicast model exists for special-purpose applications where a dedicated point-to-point flow is required (typically inside an MU for diagnostic data).

This article focuses on multicast SV from this point forward.

7. The Sampled Value Control Block

The Multicast Sampled Value Control Block (MSVCB) is the IEC 61850 object that controls how a Merging Unit publishes SV streams. It plays the same role for SV that the GOOSE Control Block (GoCB) plays for GOOSE — but for continuous high-speed streams instead of event-driven messages.

Key MSVCB Attributes (IEC 61850-9-2 Table 9)

Attribute	Type	Purpose
MsvCBNam	Identifier	MMS name of this control block
SvEna	Boolean	Enable/disable transmission — TRUE = streaming, FALSE = stopped
MsvID	VisibleString	System-wide unique stream identifier
DatSet	Reference	Data set defining which samples to publish
ConfRev	Integer	Configuration revision counter — increments on any change
SmpRate	Integer	Sample rate (interpreted per SmpMod)
OptFlds	BitString	Optional fields included in each frame
SmpMod	Enum	0 = samples per nominal period, 1 = samples per second, 2 = seconds per sample
DstAddress	Structure	Destination MAC, VLAN, APPID — see Section 8
noASDU	Integer	Number of ASDUs concatenated per APDU

OptFlds Recommendations (per IEC 61869-9)

Optional Field	IEC 61869-9 Setting
refresh-time	FALSE
sample-synchronised	TRUE (ignored, kept for backward compatibility)
sample-rate	FALSE
data-set	FALSE
security	FALSE (TRUE only when IEC 62351-6 security is active)

Excluding optional fields keeps frames compact and reduces bandwidth.

8. The SV Frame Structure

Each SV message is a standard IEEE 802.3 Ethernet frame with IEEE 802.1Q VLAN tagging. The complete structure from IEC 61850-9-2 Annex A:

Maximum Frame Sizes

Configuration	Max Frame Size
No link redundancy	1521 octets
HSR redundancy	1527 octets (+6 for HSR tag)
PRP redundancy	1527 octets (+6 for PRP trailer)

9. Multicast MAC Address Allocation

IEC 61850-9-2 Annex B specifies the recommended multicast address structure:

01 - 0C - CD - 04 - xx - xx
└──┴──┴──┴── IEEE-assigned for IEC 61850
            └── 04 = Multicast Sampled Values
                  (01 = GOOSE, 02 = GSSE)
                  xx-xx = stream-specific, assigned during engineering

Service	Recommended Range
GOOSE	01-0C-CD-01-00-00 to 01-0C-CD-01-01-FF
GSSE	01-0C-CD-02-00-00 to 01-0C-CD-02-01-FF
Multicast SV	01-0C-CD-04-00-00 to 01-0C-CD-04-01-FF

This range gives 512 unique SV stream addresses per substation — more than sufficient for any practical deployment.

⚠️ Important: Each SV stream should have a unique multicast MAC address within the network. Identical destination MACs cause subscribers to misidentify streams. Most network configuration tools enforce uniqueness at engineering time.

10. APPID and VLAN Configuration

APPID (Application Identifier)

The APPID is a 16-bit identifier in the Ethertype PDU that allows fast filtering at the link layer:

• Reserved range for SV: 0x4000 to 0x7FFF
• Default value: 0x4000 (indicates “no configuration” — should be replaced during engineering)
• Recommendation: Each SV stream gets a unique APPID for source-oriented filtering

VLAN Tag

Field	Value
TPID	0x8100 (fixed per IEEE 802.1Q)
Priority	4 (default) — higher than normal traffic, lower than GOOSE if separated
CFI	0 (must be reset)
VID	0 (default) — configure unique VID per VLAN segmentation

Priority Strategy

Per IEEE 802.1Q recommendations:

• Priority 0 — Avoid (causes unpredictable delay)
• Priority 1 — Untagged frames default
• Priority 2–3 — Lower priority bus traffic
• Priority 4 — Sampled Values default
• Priority 5–7 — Time-critical (PTP, GOOSE trips)

VLAN Segmentation

Per IEC 61850-9-2 recommendations:

• SV and GOOSE should use different VIDs so they can be bandwidth-allocated independently
• Process bus VLAN should be separate from station bus VLAN
• This ensures aggregate SV traffic does not interfere with MMS or other Layer 3 traffic

11. ASDU vs APDU — The Concatenation Model

Inside the SV frame, the APDU is the “envelope” that contains one or more ASDUs.

ASDU (Application Service Data Unit)

One ASDU = one snapshot in time containing:

Field	Description
svID	Stream identifier
datset	Data set reference (optional)
smpCnt	Sample counter (16-bit)
confRev	Configuration revision
refrTm	Refresh time (optional)
smpSynch	Synchronization status
smpRate	Sample rate (optional)
sample	The actual data values per the data set
smpMod	Sample mode (optional)

APDU (Application Protocol Data Unit)

An APDU contains:

Field	Description
savPdu	Application tag
noASDU	Number of ASDUs in this APDU (1 to 65535)
security	Optional security field (IEC 62351-6)
asdu	Sequence of ASDUs (oldest first)

Why Concatenate?

Concatenating multiple ASDUs into one APDU reduces frame overhead. For example, the preferred IEC 61869-9 protection variant F4800S2I4U4 sends 2 ASDUs per frame at 4800 sps, producing 2400 frames/second instead of 4800.

The trade-off is latency: a frame with 2 ASDUs is sent only when both ASDUs are ready, adding one sample interval of delay to the first ASDU.

For protection applications, noASDU = 1 or 2 is typical. For metering, noASDU = 6 or 8.

12. SmpCnt — The Sample Counter

The Sample Counter (SmpCnt) is a 16-bit unsigned integer (INT16U) that uniquely identifies each sample within the current second.

Behavior

• Increments with every new sample
• Resets to 0 at every synchronization pulse (top of second)
• Wraps back to 0 after reaching 65535 (only for very high sample rates ≥96 kHz)

Why It Matters

Subscribers use SmpCnt to:

1. Detect missed samples (gaps in the counter)
2. Align samples from multiple MUs (sample #1234 from MU-A = sample #1234 from MU-B if both synced)
3. Time-stamp samples without needing the network refresh time

High Sample Rate Behavior (96 kHz)

At 96000 samples/second:

• Counter overflows back to 0 after sample 65535
• Continues counting up to 30463 (96000 − 65536 − 1)
• Represents the lower 16 bits of an internal 17-bit counter
• Real top of second is when SmpCnt transitions from 30463 to 0

13. SmpSynch — Synchronization Status Values

The SmpSynch attribute tells subscribers about the time synchronization source. From IEC 61850-9-2 Table 14:

Value	Meaning	When Safe for Cross-Bay Functions
0	Not synchronized (free-running or holdover expired)	❌ Local-only measurements
1	Synchronized to a local area clock (unspecified ID)	⚠️ Only if all MUs share the same local clock
2	Synchronized to a global area clock (GPS, NTP, time-traceable)	✅ Universal — all global-synced MUs aligned
3, 4	Reserved — do not use	—
5–254	Synchronized to specific local clock with this ID	✅ Same ID = same clock = aligned
255	Reserved — do not use	—

Practical Use Guidance

Application	Required SmpSynch
Local overcurrent protection (single bay)	Any (even 0 works)
Voltage protection (single bay)	Any
Differential protection (line, transformer, bus)	≥ 1 with matching clock source
Distance protection with remote endpoints	≥ 1 with matching clock source
Synchrophasor (PMU)	= 2 (global clock required)
Wide-area monitoring	= 2

The Merging Unit publishes a current SmpSynch value in every ASDU. If sync is lost, the value drops to 0 and protection logic can take blocking action.

For PTP-based synchronization, see: PTP Power Profile Explained (IEC/IEEE 61850-9-3).

14. The Simulate (S) Flag — Testing in Live Systems

The Reserved 1 field includes a single bit S (Simulate) at bit 7 of octet 0. Per IEC 61850-9-2 Section 5.3.3.4.4:

“When this flag is set, the SampledValue telegram has been issued by a publisher located in a test device and not by the publisher as specified in the configuration file of the device.”

Why This Matters

In a live substation, you may need to inject test SV streams (e.g., from a CMC Omicron, Megger relay tester, or simulator) alongside the real MU streams. The S flag tells subscribers which is real and which is test.

Subscriber Behavior

When the S flag is set:

• Test mode subscriber: Accept and process the test SV stream
• Normal mode subscriber: Ignore the test frame — continue using the real MU stream
• The decision is made per-IED based on its configured operating mode

This mechanism allows engineers to test protection schemes against a known waveform without disconnecting the actual MU. Without the S flag, test injection would conflict with production data.

15. ASN.1 BER Encoding for SV Data Types

SV uses ASN.1 Basic Encoding Rules (BER) per ISO/IEC 8825-1 with a Type-Length-Value (TLV) format. From IEC 61850-9-2 Table 15:

Basic Data Type Encoding

IEC 61850-7-2 Type	Encoding
BOOLEAN	8-bit (0 = FALSE, anything else = TRUE)
INT8	8-bit big-endian signed
INT16	16-bit big-endian signed
INT32	32-bit big-endian signed (used for SAV.instMag.i)
INT64	64-bit big-endian signed
INT8U	8-bit big-endian unsigned
INT16U	16-bit big-endian unsigned
INT24U	24-bit big-endian unsigned
INT32U	32-bit big-endian unsigned
FLOAT32	32-bit IEEE 754
FLOAT64	64-bit IEEE 754
ENUMERATED	32-bit big-endian
CODED ENUM	32-bit big-endian
OCTET STRING	20 bytes ASCII, null-terminated
VISIBLE STRING	35 bytes ASCII, null-terminated
UNICODE STRING	20 bytes ASCII, null-terminated
TimeStamp	64-bit timestamp per IEC 61850-8-1
EntryTime	48-bit timestamp per IEC 61850-8-1
BITSTRING (per 8-1)	32-bit big-endian

Key Implication for Decoders

The 32-bit signed integer representation of instMag.i gives a dynamic range of ±2 147 483 647 counts. Combined with the IEC 61869-9 scaling factors (0.001 for amps in mA, 0.01 for volts in cV), this allows:

• Current: up to ±2 147 483 A peak (2.1 MA) — handles fault currents well beyond any practical CT range
• Voltage: up to ±21 474 836 V (21.4 MV) — covers any HV/EHV system

ASN.1 TLV Example

A complete ASDU has this structure (simplified):

30 [Length] svID    [0] [Length] [Value]
            datset  [1] [Length] [Value]   (optional)
            smpCnt  [2] [Length=2] [Value]
            confRev [3] [Length=4] [Value]
            refrTm  [4] [Length] [Value]   (optional)
            smpSynch[5] [Length=1] [Value]
            smpRate [6] [Length=2] [Value] (optional)
            sample  [7] [Length=N] [Value]
            smpMod  [8] [Length=2] [Value] (optional)

Each TLV triplet allows decoders to skip unknown fields and handle optional elements gracefully.

16. Sample Rates (IEC 61869-9 Preferred Values)

The original IEC 61850-9-2 standard did not mandate specific sample rates. IEC 61869-9:2016 filled this gap with preferred rates:

Sample Rate	ASDUs/Frame	Frames/sec	Application
4000 sps	1	4000	9-2LE backward compat (50 Hz)
4800 sps	1	4800	9-2LE backward compat (60 Hz)
4800 sps	2	2400	Preferred — protection/measurement
5760 sps	1	5760	60 Hz with 96 samples/cycle
14400 sps	6	2400	Preferred — quality metering
12800 sps (deprecated)	8	1600	9-2LE legacy 50 Hz
15360 sps (deprecated)	8	1920	9-2LE legacy 60 Hz
96000 sps	1	96000	High-bandwidth DC control

The preferred protection rate (F4800S2I4U4) and preferred metering rate (F14400S6I4U4) both produce 2400 frames/second — a frequency-independent network rate that simplifies engineering.

For the variant code notation, see: What Is a Merging Unit?

17. PRP and HSR for SV Streams

Protection-grade SV networks require zero-recovery-time redundancy. IEC 61850-9-2 supports both PRP and HSR per IEC 62439-3.

PRP (Parallel Redundancy Protocol)

The MU has two Ethernet ports, one to LAN A and one to LAN B. Every SV frame is sent simultaneously on both LANs. The subscriber receives both and processes the first arrival, discarding the duplicate.

• Recovery time: zero (no detection or reconfiguration)
• Frame overhead: 6-byte PRP trailer (Sequence number + LAN ID + Suffix)
• Max frame size: 1527 octets (vs 1521 without redundancy)

HSR (High-availability Seamless Redundancy)

The MU is a node in a ring. Every SV frame is sent in both ring directions simultaneously. The receiver gets two copies; the second is discarded.

• Recovery time: zero (for any single break in the ring)
• Frame overhead: 6-byte HSR tag (Path + Size + Sequence number)
• Max frame size: 1527 octets

Bandwidth Doubling

Both PRP and HSR duplicate every SV frame on each network path. A 10-bay substation at F4800S2I4U4:

• Per LAN (PRP) or per ring direction (HSR): ~60 Mbit/s SV traffic
• Plus GOOSE, PTP, MMS
• Gigabit Ethernet strongly recommended for any non-trivial PRP/HSR deployment

For full PRP details, see: Parallel Redundancy Protocol (PRP) Complete Guide For HSR details, see: HSR (High-Availability Seamless Redundancy)

18. IEC 62351-6 Security for SV

IEC 61850-9-2 reserves a 28-bit security field (Reserved 2 + 4 bits of Reserved 1) for use by IEC/TS 62351-6, the security standard for IEC 61850.

What IEC 62351-6 Provides

Feature	Status
Frame authentication	Optional
Source verification	Optional
Anti-replay protection	Optional
Frame encryption	NOT applied to SV/GOOSE (would break real-time performance)

Why Not Encryption?

Encrypting SV would add latency that breaks protection-class real-time requirements (sub-millisecond). Instead, IEC 62351-6 uses HMAC-SHA256 for authentication only — the SV payload remains plaintext but is signed so subscribers can verify it came from a legitimate publisher.

Activating Security

When IEC 62351-6 security is enabled:

• MSVCB.OptFlds.security = TRUE
• The 28-bit security field carries authentication data
• Subscribers verify the signature; failed verification → discard frame

Adoption Status

IEC 62351-6 security is rarely deployed in production substations due to:

• Performance overhead (still measurable even without encryption)
• Key management complexity
• Limited vendor support
• Industry preference for network-level isolation (separate VLAN, separate physical network)

Most digital substations rely on physical isolation and network segmentation rather than cryptographic authentication.

19. Network Bandwidth Sizing

SV streams generate continuous high-rate traffic. Bandwidth planning is mandatory.

Per-Stream Bandwidth

Variant Code	Frames/sec	Frame Size	Bandwidth
F4000S1I4U4	4000	~150 B	~5 Mbit/s
F4800S1I4U4	4800	~150 B	~6 Mbit/s
F4800S2I4U4 (preferred protection)	2400	~300 B	~6 Mbit/s
F14400S6I4U4 (preferred metering)	2400	~900 B	~17 Mbit/s
F96000S1I4U4	96000	~150 B	~120 Mbit/s

Aggregate Bandwidth Examples

Substation Size	Streams	Total SV BW	Recommended Network
4 bays, protection only	4	~24 Mbit/s	100 Mbit/s OK
10 bays, protection only	10	~60 Mbit/s	100 Mbit/s tight, Gigabit recommended
10 bays, protection + metering	20	~230 Mbit/s	Gigabit required
20 bays, full	40	~460 Mbit/s	Gigabit required

Maximum Quantities per Stream (IEC 61869-9)

To prevent network blocking from oversized frames on 100 Mbit/s networks:

Application	Max I+U Quantities
General measuring/protection	24
Quality metering	8
DC control	24

No limits apply for Gigabit and faster networks.

20. Network Engineering Requirements

A poorly engineered process bus causes delayed samples and relay misoperation. Required elements:

Mandatory

• VLAN tagging (IEEE 802.1Q) with VID and priority configured
• Multicast filtering at switch level (IGMP snooping or static multicast tables)
• Deterministic switching (cut-through preferred, store-and-forward acceptable for SV)
• Hardware MAC filtering capable of handling SV multicast load
• Time synchronization (PTP per IEC/IEEE 61850-9-3 or 1PPS)

Strongly Recommended

• PRP or HSR redundancy for protection-grade applications
• Gigabit Ethernet for any deployment beyond a few SV streams
• Dedicated process bus VLAN separate from station bus
• IEC 61850-3 / IEEE 1613 rated switches for substation environments
• Hardware PTP timestamping in switches (transparent clock)

Avoid

• RSTP on the process bus (recovery too slow)
• Mixing corporate/IT traffic on the process bus
• Routing SV between subnets (Layer 2 only)
• Consumer-grade Ethernet switches
• Default APPID (0x4000) in production

For network switches in substations, see: IEC 61850-3 Ethernet Switch Requirements

21. SV Performance and Latency

End-to-end SV latency from primary side to subscriber:

Component	Typical Delay
Primary CT/VT response	< 1 µs
MU anti-aliasing filter	~100–200 µs
MU ADC + processing	~200–500 µs
MU total (processing delay)	≤ 2 ms (IEC 61869-9 protection limit)
Switch (cut-through, per hop)	~3–10 µs
Switch (store-and-forward, per hop)	50–150 µs
Subscriber receive + decode	~50–200 µs
Total end-to-end	typically 2–3 ms

IEC 61869-9 Maximum Processing Delay (MU side only)

Application	Limit
Quality metering	10 ms
Protection and measurement	2 ms
Time-critical low-bandwidth DC control	100 µs
High-bandwidth DC control	25 µs

Why Phase Error ≠ Delay

A common misunderstanding: SV processing delay does NOT cause phase error. The encoded SmpCnt field timestamps the sample with the actual sampling instant. Subscribers use this timestamp to reconstruct phasors — not the time of frame arrival. Thus delay affects fault detection time but not measurement accuracy.

22. Procurement Specification Template

Required Standards

• IEC 61850-9-2:2011 compliance
• IEC 61869-9:2016 profile compliance
• IEC/IEEE 61850-9-3 for time synchronization
• IEC 61850-10 conformance testing

Required Capabilities

• EtherType 0x88BA SV publishing
• Multicast SV (MSVCB) mandatory
• Unicast SV (USVCB) optional
• IEEE 802.1Q VLAN tagging with configurable VID and priority
• PRP and/or HSR support per IEC 62439-3
• Simulate (S) flag handling per IEC 61850-9-2

Required Variant Codes (IEC 61869-9)

• F4800S2I4U4 — preferred protection/measurement
• F14400S6I4U4 — preferred quality metering
• Legacy variants (F4000S1I4U4 / F4800S1I4U4) if 9-2LE compatibility needed

Subscriber Requirements

• Multicast frame reception and filtering on configured MAC + APPID + VLAN
• SmpSynch value handling (0, 1, 2, 5–254)
• SmpCnt continuity monitoring (gap detection)
• ConfRev mismatch handling
• Simulate (S) flag mode handling (test vs production)
• Maximum number of simultaneous SV subscriptions

Documentation

• PICS, PIXIT, MICS (IEC 61850-10 compliant)
• ICD file with full SCL declaration
• Performance benchmarks (max subscribed streams, max processing delay)
• Third-party conformance test report

23. Field Troubleshooting: Common Failure Patterns

When working on site, these are the patterns to check first.

🔴 Pattern 1: Subscriber receives no SV stream

Likely causes:

• SvEna = FALSE in the MSVCB
• Wrong multicast MAC address configured on subscriber
• IGMP snooping blocking the multicast on the switch
• VLAN mismatch between publisher and subscriber

Actions:

• Verify SvEna = TRUE using system configurator
• Wireshark capture on the subscriber side, filter eth.type == 0x88ba
• Check switch multicast forwarding tables
• Verify VLAN tags match end-to-end

🔴 Pattern 2: SmpSynch stuck at 0

Likely causes:

• PTP grandmaster unreachable
• PTP profile mismatch (default IEEE 1588 vs IEC/IEEE 61850-9-3)
• 1PPS signal lost or jittery beyond ±2 µs
• MU still in startup, hasn’t acquired sync yet
• Sync source lost > MU holdover time

Actions:

• Check PTP master diagnostics, clockClass = 6 (synced) or 7 (holdover)
• Verify PTP domain matches MU expectation (domain 93 if following IEC/IEEE 61850-9-3 recommendation)
• For 1PPS, measure signal quality at MU input
• Allow 30+ seconds after power-on for steady state

🔴 Pattern 3: ConfRev mismatch

Likely causes:

• MU configuration changed after subscribers were commissioned
• Data set membership/order modified
• SCD regeneration without subscriber update

Actions:

• Compare publisher’s ConfRev against each subscriber’s configured value
• Re-engineer subscribers if MU dataset changed
• Verify data set member order matches subscriber expectation

🔴 Pattern 4: Subscriber reads wrong values

Likely causes:

• Scaling factor mismatch (subscriber doesn’t apply 0.001 for mA → A)
• Wrong CT/VT ratio on subscriber
• Data set member order differs from subscriber expectation
• Wrong TCTR/TVTR instance interpretation (subscriber reading B-phase as A-phase)

Actions:

• Verify scaling: AmpSv.instMag.i is in milliamperes (multiply by 0.001 for amperes); VolSv.instMag.i is in centivolts (multiply by 0.01 for volts)
• Check substation section of SCL for phase-to-instance binding
• Confirm data set order in publisher matches subscriber’s decoding

🔴 Pattern 5: SmpCnt gaps (missing samples)

Likely causes:

• Network congestion / dropped frames
• Subscriber CPU overload (can’t keep up with frame rate)
• Switch buffer overflow
• Hardware fault on MU output port

Actions:

• Wireshark capture, check for gaps in smpCnt sequence
• Verify switch port statistics (drops, errors)
• Reduce subscriber load (subscribe to fewer streams or use lower rate)
• Inspect cable and SFP module quality

🔴 Pattern 6: Duplicate APPID conflict

Likely causes:

• Two MUs configured with the same APPID (e.g., both at default 0x4000)
• Engineering tool didn’t enforce uniqueness
• Manual configuration error

Actions:

• Each SV stream needs a unique APPID in 0x4000–0x7FFF
• Wireshark shows duplicate APPIDs immediately when capturing
• Re-run system configurator to enforce uniqueness

🔴 Pattern 7: Network bandwidth saturation

Likely causes:

• Too many SV streams on 100 Mbit/s network
• PRP/HSR doubling traffic without bandwidth headroom
• Switch CPU overload from multicast snooping
• Mixed VLAN traffic on same broadcast domain

Actions:

• Calculate total SV bandwidth (Section 18)
• Upgrade to Gigabit Ethernet if approaching 50% link utilization
• Implement IGMP snooping or static multicast filters
• Segment SV onto dedicated VLAN

🔴 Pattern 8: Test injection accidentally affects production

Likely causes:

• Test SV stream did NOT set the Simulate (S) bit
• Subscriber not configured to filter by S flag
• Same APPID as production stream

Actions:

• Verify test equipment sets S = 1
• Configure subscriber test mode to accept S = 1 streams
• Use different APPID for test streams when possible

🔴 Pattern 9: Differential protection misoperates after sync change

Likely causes:

• One end’s MU running in free-run mode (SmpSynch = 0)
• Two ends synced to different local clocks (both SmpSynch = 1, different sources)
• Time jump > sample interval during PTP re-lock causes SmpCnt discontinuity

Actions:

• Verify both ends report SmpSynch = 2 (global) or matching ID in 5–254 range
• Check PTP grandmaster coverage at both ends
• Implement blocking logic during sync state transitions

🔴 Pattern 10: Cannot decode SV in Wireshark

Likely causes:

• Wireshark version too old (< 2.0 may not have native SV decoder)
• VLAN tag stripping by capture interface
• SPAN port losing VLAN information

Actions:

• Update Wireshark to latest version
• Use proper SPAN configuration that preserves VLAN tags
• Filter eth.type == 0x88ba to isolate SV
• Apply IEC 61850 dissector preferences for custom decode

For Wireshark-based GOOSE analysis (similar technique applies to SV), see: Decoding IEC 61850 GOOSE Messages with Wireshark

Summary

Sampled Values are the foundation of the IEC 61850 process bus. They replace analog CT/VT wiring with deterministic, time-aligned digital streams that any number of IEDs can simultaneously consume.

Key Takeaways

• SV operates on Layer 2 Ethernet with EtherType 0x88BA, multicast MAC 01-0C-CD-04-xx-xx, default VLAN priority 4
• The current authoritative profile is IEC 61869-9:2016 (replaces 9-2LE)
• Publisher/subscriber model — publishers (MUs) broadcast, any subscriber can consume
• MSVCB (multicast) is the standard model; USVCB (unicast) rarely used
• Each frame is ASN.1 BER encoded with TLV structure
• APDU can contain 1 to N ASDUs — concatenation reduces overhead but adds latency
• SmpCnt identifies each sample within the second; resets to 0 at sync pulse
• SmpSynch values: 0 (not synced), 1 (local clock), 2 (global clock), 5–254 (specific local IDs)
• Simulate (S) bit distinguishes test streams from production
• AmpSv scaled in milliamperes, VolSv in centivolts (per IEC 61869-9 Tables 904/906)
• Preferred protection rate: F4800S2I4U4 = 4800 sps × 2 ASDU = 2400 frames/sec
• Preferred metering rate: F14400S6I4U4 = 14400 sps × 6 ASDU = 2400 frames/sec
• PRP/HSR doubles bandwidth on each path — plan for Gigabit Ethernet on anything non-trivial
• IEC 62351-6 security is optional; most deployments use network-level isolation instead
• Maximum SV processing delay: 2 ms for protection per IEC 61869-9