IEEE 1588: What’s the difference between a Boundary Clock and Transparent Clock?

The original IEEE 1588-2002 standard for a precision clock synchronization protocol describes Ordinary Clocks, which are either Grandmaster Clocks or Slave Clocks. An Ordinary Clock (OC) always has a single port.

IEEE 1588-2002 also describes Boundary Clocks. A Boundary Clock (BC) is a clock node that has two or more ports. For example, a router or Ethernet switch might be capable of operating as a BC.

A BC generally has one port in the role of a slave clock and the remaining ports in the role of master clocks. In this case, the BC recovers the time of day within the slave clock function and relays it as a reference to the master clock functions.

A practical synchronization network might consist of a Grandmaster Clock, a tree of BCs, and many Slave Clocks.

IEEE 1588-2008 introduces a new type of clock called a Transparent Clock (TC). A TC is a multi-port device that forwards precision time protocol messages, measuring the time taken for event messages to pass through the device, and accounts for this so-called residence time by modifying the message, or by sending a separate follow-up message. The other clocks in the system use the delay measured by the TC to compensate for the additional latency introduced between master and slave clocks. In this sense, the TC can be made to “disappear”, hence the name.

In a typical telecommunications network with “full on-path support” every network node or transmission system must be either a BC or a TC so that the delay through that node or transmission system does not degrade the synchronization performance.

So, which is better? The short answer is that both are capable of very high performance in a synchronization network, and one is not consistently more accurate than the other.

In some applications, a TC is significantly easier to implement because the latency measurements and residence time compensation can be done in hardware, and the node does not need to provide a full slave clock implementation. Also, in a TC there is almost nothing to configure or to tune.

There are also some advantages in using BCs:

  • The BC can provide a simple synchronization output (for example, one pulse per second) at an intermediate node in a network, allowing an operator to check and calibrate synchronization.
  • The BC can provide a selection between two or more Grandmaster Clocks in a network with redundancy.
  • BCs can be used to segment the timing network, providing distributed processing of 1588 protocols, and protecting the Grandmaster from overload.

In PTP 650 and PTP 700 the simpler implementation and ease of configuration make the TC an obvious choice. The Transparent Clock function in PTP 650 and PTP 700 provide accurate and robust correction of residence time with constant time error less than 100 ns per link regardless of traffic load, modulation mode or channel bandwidth, controlled by a simple enable/disable control.

A very attractive approach in a practical precision synchronization network is to use switches or routers functioning as BCs at each major node, interconnected by wireless links operating as TCs. This takes maximum advantages of the two types of node.

Please contact us if you have a requirement for precision timing in your network. We’d be very happy to discuss it with you.