Home Support Blog

6 Core Application Scenarios for PTP Switches and PoE+PTP Converged Technology

Release date:2026-11-10

A timing-capable switch earns its price in a specific set of situations: wherever two or more devices must agree on when something happened, not just on what happened. Rayin (Shenzhen Rayin Technology) is a manufacturer of GPON OLTs and industrial Ethernet switches. The IEEE 1588 precision time protocol does the distribution, but the accuracy you get depends on whether the switch timestamps in hardware — which is the single feature that separates a real timing device from an ordinary managed switch.

KEY TAKEAWAYS
  • Hardware timestamping is the dividing line. With it, PTP reaches sub-microsecond agreement across a LAN; without it, software timestamping leaves you in the tens-of-microseconds-to-milliseconds range (industry-typical).

  • The switch's role matters. A boundary clock re-times on every port and recovers accuracy hop by hop; a transparent switch only corrects its own residence time.

  • Converging PoE and PTP on one device is about cabling: power and timing on a single drop to a camera, sensor or AP, at the cost of a shared thermal and power budget.

Where does timing actually matter?

Six recurring deployments, with the accuracy class each one typically needs.

ScenarioWhat must be synchronisedTypical accuracy target
Substation automationSampled values and trip commandsAround 1 µs for sampled value streams
Multi-axis motion controlServo axes and robot cellsTens of microseconds of jitter
Machine visionCamera triggers and strobesSub-millisecond frame alignment
Broadcast IP productionAudio, video and ancillary essencesPer ST 2059 PTP profile
Distributed test rigsData acquisition channelsMicroseconds across channels
5G and TSC edgeRadio frames and coordinated schedulingMicrosecond class

In substation automation, IEC 61850 defines timing classes and sampled values are the demanding case — without tight time alignment across merging units, protection functions lose the common reference they rely on. In motion control, it is jitter rather than absolute accuracy that breaks things: axes drift relative to each other when the correction interval varies.

In broadcast, the profile is explicit: SMPTE ST 2059 defines the PTP profile used with ST 2110 essences, so audio and video stay locked. That is why a switch intended for production environments is specified against ST 2059 and ST 2022-7 rather than generic "PTP support".

What should the switch actually do with the clock?

Three roles, and the difference is not cosmetic.

RoleBehaviourWhen you want it
Ordinary clockOne port synchronised, one time sourceEnd devices
Boundary clockRe-times on every port, new timing domain per linkSwitches carrying many devices
Transparent clockCorrects for its own residence timeSimple daisy chains

A boundary clock is the workhorse: each port terminates the PTP session and starts a fresh one, so accuracy is recovered at every hop rather than degrading along the chain. Ask which role the device implements, because "supports PTP" is consistent with all three.

Profiles matter too. IEEE 802.1AS (gPTP) is the profile used with TSN; ITU-T G.8275.1 and G.8275.2 are the telecom profiles; SMPTE ST 2059 is the media one. A switch that implements the wrong profile for your application will not interoperate, however good its timestamping is.

How do you verify timing on site?

Verification is cheap and worth doing before a system depends on it. Three checks cover most cases:

  1. Confirm the grandmaster is what you think it is. A misconfigured priority lets an unintended device become the time source, and everything downstream then agrees on the wrong time very precisely.

  2. Measure offset and path delay on a live port. Most implementations report both; a growing offset under load points to a switch in the path that is not participating.

  3. Test the failure case. Remove the grandmaster and confirm holdover behaviour is acceptable, then restore it and confirm recovery without a manual restart.

Where a TSN schedule is involved, add one more: verify that the talker and listener agree on the schedule and that the switches on the path implement the same clause. Timing that is correct but unscheduled will not deliver bounded latency.

What does PoE plus PTP convergence buy you?

One cable per device instead of two. A camera, a sensor or an access point gets power and a timing reference from the same port, which removes a mains drop at every mounting point — often the most expensive part of a deployment in an existing building or a roadside cabinet.

The trade-offs are physical, not protocol-level:

  • Power budget is finite. Add up the draw of every powered device and leave headroom; a port that browns out under load behaves like a device failure.

  • Heat follows power. Delivering watts through the switch raises internal temperature, so check the derating curve at the site's real ambient.

  • Timing and power share a failure domain. Losing the switch loses both, so dual power inputs and a protected uplink matter more on a converged device.

High-voltage electrical substation equipment

FAQs

Is PTP the same as NTP? No. NTP synchronises computers to millisecond-level accuracy over a routed network. PTP with hardware timestamping targets sub-microsecond agreement on a LAN and needs switches that participate.

Why does hardware timestamping matter so much? Because the timestamp has to be taken at the physical interface. Software timestamping adds operating-system scheduling delay that varies with load, which turns into jitter.

Do I need a boundary clock? For most multi-device deployments, yes. It re-times at each hop and stops accuracy degrading along a chain of switches, which is what transparent correction alone cannot fully do.

Can one switch power devices and distribute timing? Yes, and that is the point of convergence: one drop carries power and PTP. Budget the watts and the heat, and remember both now share one failure domain.

Which PTP profile should I specify? Match the application: IEEE 802.1AS for TSN networks, ITU-T G.8275.1 or G.8275.2 for telecom, SMPTE ST 2059 for broadcast media over IP.

Related Reading

  1. PTP and IEEE 1588: Layer 2 switching for industrial timing

  2. How to choose PTP industrial switches

  3. PTP buying guide: Moxa, 3onedata and Rayin compared

  4. Latency, jitter and packet loss: the QoS targets that matter

  5. What a 90 W PoE switch can actually power

  6. LLDP: benefits and use cases in industrial networks

  7. Broadcast production networks: SMPTE 2110 on industrial switches

About the author: Sara is a Customer Manager at Rayin with over 10 years of experience in the communications field. In her free time, she enjoys badminton and swimming.

Connect with Sara on LinkedIn


About Rayin: Shenzhen Rayin Technology Co., Ltd. — company profile


Get A Quote

You have agreed to this website’s《Privacy Policy》