Technical Manual · Network & Protocols
Network Visualization Hub — Methodology
Every protocol parameter, frame structure, timing basis, and computed display metric behind the animated industrial protocol labs. Modbus, BACnet, OPC-UA, DNP3, PROFINET, EtherNet/IP, EtherCAT — all anchored to their governing standards. This is primarily a teaching tool; formulas documented here drive the compare-mode instrument strip and the per-protocol timing animations, not a network-sizing calculator.
▶ Open the live Network Visualization Hub01 Purpose & engineering basis
The Network Visualization Hub is an animated protocol teaching tool, not a network-sizing calculator. It renders the real on-wire frame structure, bit timing, addressing, and handshake choreography for each industrial and IT protocol, so learners can see why Modbus RTU has lower overhead than Modbus TCP, why BACnet MS/TP needs a token before transmitting, and why EtherCAT can achieve sub-millisecond cycle times where PROFINET IRT targets 1 ms.
The Lane B Industrial OT set (9 protocols LIVE) is the primary focus of this manual. The compare mode derives a normalised instrument strip — throughput, per-frame latency, header overhead — from the canonical protocol parameters tabulated in §§ 3–4. All other lanes (A: Foundations, C: DC Management, D: Security, E: APIs+Agents) follow the same animation methodology but have no computed numeric strip beyond the visualisation itself.
Alignment: CompTIA Network+ N10-009 (industrial protocols, OSI model, TCP/IP, subnetting) and CompTIA Security+ SY0-701 (TLS, OAuth, mTLS, zero-trust). Protocol specifications are the governing authority; CompTIA exam objectives provide the pedagogical framing.
02 Inputs & controls
The hub has no numeric input fields — it is parameter-driven by the protocol constants embedded in each lab. User controls select the context; the visualisation responds deterministically.
| Control | Type | Effect |
|---|---|---|
| Topic / protocol selector | Card click / lane nav | Loads the selected protocol animation lab. Each lab hard-codes the canonical frame for that protocol. |
| Compare panel (compare mode) | Multi-select (up to 4) | Renders up to 4 protocol lanes side-by-side with a normalised instrument chip strip below each. |
| Failure scenario toggle | Per-lab control (where present) | Triggers protocol-specific fault: CRC error (Modbus), token-loss (BACnet MS/TP), unsolicited retry (DNP3), ring failover (PROFINET), distributed-clock fault (EtherCAT). |
| Theme (dark / light) | Global toggle | Switches canvas background and token colours. Instrument-cyan and signal-amber are preserved in both themes. |
| Animation speed | Per-lab slider (where present) | Scales animation frame rate; does not change the protocol parameters or computed metrics. |
03 Methodology — computed display metrics
The compare mode instrument strip shows three computed metrics per protocol, derived entirely from the protocol constants in § 4. These are display metrics for comparison and education — not field measurements or engineering estimates for production traffic.
Serial-link frame transmission time
frame_bytes = PDU + framing overhead per standard. baud_rate from § 4 protocol table.IEC 61158 · STANDARD
Ethernet-link frame transmission time
link_rate_bps = 100 Mbps (100BASE-TX) or 1 Gbps (1000BASE-T) per lab configuration.IEEE 802.3 · STANDARD
One-way propagation delay
d = segment length (visualisation default 100 m). This is a pure propagation estimate.RFC 3393 · INFORMATIONAL
Link utilisation (compare strip)
Redundant-path count (topology annotation)
04 Protocol parameters & constants
Every value the hub uses is sourced from the governing standard for that protocol. Constants are hard-coded in each lab; no back-solved or estimated values are used.
Lane B — Serial industrial protocols
| Protocol | Physical | Default baud | Address space | Max nodes | CRC type | Year / std |
|---|---|---|---|---|---|---|
| Modbus RTU | RS-485 half-duplex | 9 600 bps (common); up to 115 200 | 1–247 (slave ID) | 247 | CRC-16/IBM (poly 0x8005) | 1979 · Modicon spec |
| BACnet MS/TP | RS-485 half-duplex | 38 400 bps typical; 9 600–115 200 | 0–254 (MAC address) | 128 (practical) | CRC-8 (header) + CRC-16 (data) | 1995 · ASHRAE 135 |
| DNP3 (serial) | RS-485 / RS-232 | 9 600 bps; up to 115 200 | 0–65 519 (address) | Network-wide | CRC-16/DNP (poly 0x3D65) | 1990 · IEEE 1815-2012 |
Lane B — Ethernet industrial protocols
| Protocol | Transport | Frame overhead (bytes) | Payload max (bytes) | Cycle time target | Redundancy | Std |
|---|---|---|---|---|---|---|
| Modbus TCP | TCP/IP (port 502) | MBAP 7 + FC 1 = 8 typical header | 253 (ADU data) | None mandated (poll-driven) | None built-in | Modbus.org spec |
| BACnet/IP | UDP/IP (port 47808) | BVLC 4 + NPDU 2+ + APDU variable | 1476 (1 500 − BVLC/NPDU) | None mandated | BACnet/IP failover (BBMD) | ASHRAE 135 / ISO 16484-5 |
| OPC-UA | TCP (port 4840) or HTTPS | UA Secure Channel + Message header ≥ 28 | ~65 000 (chunked) | Subscription period user-defined (≥ 0 ms) | Redundant server pairs | IEC 62541 |
| DNP3 (TCP) | TCP (port 20 000) | Data-link 10 + Transport 1 + App 2 = 13 min | 249 per segment | None mandated | Dual-path (optional) | IEEE 1815-2012 |
| PROFINET RT | Ethernet (EtherType 0x8892) | ~14 (Eth) + 4 (VLAN) + 2 (frame ID) = 20 | 1 440 | 1 ms (RT) / 250 µs (IRT) | MRP ring (IEC 62439-2) | IEC 61158-6-10 |
| EtherNet/IP | TCP (explicit) / UDP (implicit, port 2222) | ENIP 24 + CPF 12 + CIP variable | ~1 460 (explicit); 504 (I/O) | RPI user-defined (min ~1 ms) | DLR ring (IEC 62439-6) | ODVA CIP Vol. 2 |
| EtherCAT | Ethernet (EtherType 0x88A4); or UDP | 14 (Eth) + 2 (EtherCAT header) + per-datagram 10 | 44 bytes per datagram (process data) | 31.25 µs min (DC sync) | Ring topology auto-close | IEC 61158-3-12 / IEC 61784-2 |
BACnet MS/TP token-passing timing constants
| Constant | Value | Meaning |
|---|---|---|
Tslot | 10 ms | Time slot per node in token-passing cycle |
Nretry_token | 1 | Retries before skipping a silent node |
Tusage_delay | 10 ms max | Delay after receiving token before transmitting |
Treply_delay | 255 ms max | Max wait for a reply frame |
| Max nodes | 128 | Practical limit before token-rotation latency exceeds 1.28 s |
Modbus RTU frame structure
| Field | Bytes | Notes |
|---|---|---|
| Slave address | 1 | 1–247 (0 = broadcast, 248–255 reserved) |
| Function code | 1 | e.g. 0x03 = Read Holding Registers |
| Data | 0–252 | PDU payload; varies by function code |
| CRC | 2 | CRC-16/IBM, LSB first |
| Silent interval | ≥ 3.5 char times | Frame delimiter; at 9 600 bps ≈ 3.6 ms |
Modbus TCP — MBAP header
| Field | Bytes | Notes |
|---|---|---|
| Transaction identifier | 2 | Client-generated; echoed by server |
| Protocol identifier | 2 | Always 0x0000 (Modbus) |
| Length | 2 | Byte count of remaining ADU |
| Unit identifier | 1 | Slave address (1–247); 255 = no bridging |
| PDU (FC + data) | 1–253 | Same as RTU minus CRC |
EtherCAT distributed-clock (DC) sync parameters
| Parameter | Value | Meaning |
|---|---|---|
| Sync0 cycle period min | 31.25 µs | Fastest deterministic sync interval |
| DC reference clock drift | < 100 ns (typ.) | Slave clocks synchronised to reference master |
| Telegram processing delay | ~100–300 ns per slave | On-the-fly read/write per slave in ring |
| Ring latency per slave | ~ns | Telegram modified in-flight; no store-and-forward |
05 Outputs
The hub produces visual and numeric outputs. Numeric outputs appear only in compare mode.
| Output | Source | Unit | Interpretation |
|---|---|---|---|
| Animated frame chip | Canvas 2D animation | Visual | Each chip segment is colour-coded to a named protocol field (address / function / data / CRC / header). Field labels appear on hover. |
| Frame transmission time | t_frame = bytes×8 / baud | µs | Compare-strip metric. How long the frame occupies the wire. |
| Header overhead | 1 − payload/frame_bytes | % | Protocol efficiency indicator in compare strip. |
| Link utilisation | payload / frame × 100 | % | Horizontal bar in compare strip. Higher = more efficient use of bandwidth. |
| Cycle / poll period | Protocol constant (§ 4) | ms or µs | Shown as annotation on the timeline ruler. EtherCAT sub-ms; PROFINET RT 1 ms; Modbus poll-driven. |
| Redundant paths | Topology count (§ 3) | count | Topology panel annotation: 0 (Modbus bus), 1 (PROFINET MRP ring, EtherCAT ring). |
| Failure scenario state | Animation state machine | Visual | CRC fault / token-loss / retry / ring-close drawn on the canvas when failure toggle is active. |
06 Worked example — Modbus RTU frame walk-through
Master reads 10 holding registers from slave 3 (FC 03) at 9 600 bps. This is the canonical request shown in the Modbus RTU lab.
- Request frame bytes: Slave ID (1) + FC (1) + Start addr (2) + Quantity (2) + CRC (2) = 8 bytes
- Frame transmission time:
8 × 8 / 9 600 =6.67 ms - Silent interval (3.5 char times): at 9 600 bps, 1 char = 10 bits (8N1+stop) → 1/9 600 × 10 = 1.04 ms → 3.5 × 1.04 ms = ≈ 3.6 ms — animated as gap between chips
- Response frame bytes (10 registers = 20 bytes data): Slave ID (1) + FC (1) + Byte count (1) + Data (20) + CRC (2) = 25 bytes
- Response transmission time:
25 × 8 / 9 600 =20.83 ms - Total round-trip (idealized, no processing delay): 6.67 + 3.6 + 20.83 + 3.6 ms = ≈ 34.7 ms
- Link utilisation (response): payload 20 / frame 25 = 80.0%
- Header overhead (response): 1 − 0.80 = 20.0% (address + FC + byte-count + CRC)
29×8 / 100×10⁶ = 2.3 µs — three orders of magnitude faster than RS-485 serial, illustrating the throughput gain the Modbus TCP lab visualises.07 References & standards
- Modbus spec (2012) — Modbus Application Protocol Specification v1.1b3; Modbus over Serial Line Specification v1.02. Serial framing, CRC-16, function codes, address space.
- ASHRAE 135 / ISO 16484-5 — BACnet: A Data Communication Protocol for Building Automation and Control Networks. MS/TP token-passing, BACnet/IP BVLC, timing constants.
- IEC 62541 (OPC Foundation) — OPC Unified Architecture. Parts 1–14: information model, services, transport (TCP/HTTPS), security (Secure Channel), subscription model.
- IEEE 1815-2012 / DNP3 — IEEE Standard for Electric Power Systems Communications — Distributed Network Protocol (DNP3). Framing, CRC-16/DNP, unsolicited responses, addressing.
- IEC 61158 / IEC 61784-2 — Fieldbus standards family. PROFINET (IEC 61158-6-10), EtherNet/IP (IEC 61158-3-2/6-2), EtherCAT (IEC 61158-3-12).
- ODVA CIP Vol. 2 — EtherNet/IP Adaptation of CIP. ENIP encapsulation header, CPF items, CIP services, DLR ring redundancy.
- ETG.1000 series — EtherCAT Technology Group specifications: EtherCAT frame format (EtherType 0x88A4), distributed clocks, on-the-fly processing, ring topology.
- RFC 3393 (IPPM) — IP Packet Delay Variation Metric for IPPM. Framework for one-way delay and jitter. Basis for propagation delay formula
t_prop = d / v_prop. - IEC 62439-2 (MRP) — Media Redundancy Protocol for PROFINET ring failover. Recovery time ≤ 200 ms (typ. < 30 ms with fast MRP).
- IEC 62439-6 (DLR) — Device Level Ring for EtherNet/IP ring redundancy. Single-fault recovery < 3 ms (typ.).
- CompTIA Network+ N10-009 / Security+ SY0-701 — Exam objectives providing pedagogical framing for OSI model, TCP/IP, industrial protocols, TLS, OAuth coverage.
08 Assumptions & limitations
Frame sizes shown are for the most common request/response pair in each protocol (e.g. FC 03 read 10 registers for Modbus) — not worst-case maximum frames. Timing calculations assume ideal wire conditions: no collisions, no retransmissions, no queuing delay, and no switch latency for Ethernet-based protocols. Processing delay at master and slave/server is not modelled.
Baud rates shown are the most common field deployments; each protocol supports a range (Modbus RTU 1 200–115 200 bps; BACnet MS/TP up to 115 200 bps; DNP3 serial up to 115 200 bps). The compare-mode metrics change linearly with baud rate but the user cannot adjust this in the current hub — it is fixed to the default for each protocol.
Propagation velocity v_prop ≈ 2×10⁸ m/s assumes standard Cat-5e/Cat-6 twisted pair (NVP ≈ 0.67c). Fiber would use ~2×10⁸ m/s as well (multimode, NVP ≈ 0.67–0.69c). RS-485 cable velocity factor is similarly ~0.67. The 100 m segment length used in the latency display is the visualization default, not a configurable field.
This hub and its manual are an engineering education aid aligned to CompTIA certification domains. For production network design, always consult the governing protocol standard, the vendor's implementation guide, and a qualified network engineer. Frame sizes, timing, and addressing must be validated against the actual devices deployed.