Testing a Boundary Clock requires a Master to provide the reference time to the BC slave. The BC Slave then recovers time from PTP, optionally generates a 1pps signal, transfers time to the BC Master which then transmits this time to any connected slaves.

The performance of the BC is measured using the PTP generated by the BC Master and so verifies all of the above processing steps.

Measuring the 1pps output from the BC provides some insight into the BC Slave performance i.e. measures whether the BC is correctly recovering time from the reference Master. This is useful so as to determine whether any error is caused by clock recovery or in the transfer of time to the BC Master and the subsequent transmission of that time.

Test Configuration

Open

1pps Measurements

1pps time error is the difference between the rising edge of the reference 1pps pulse and the rising edge of the pulse to be measured.

The test equipment ensures that the reference 1pps pulse is aligned with PTP top-of-second so as to provide an accurate measurement of the DUT’s ability to recover time from PTP and correctly generate a 1pps output.

Open

The 1pps pulse from the DUT is measured at the time at which it exits the DUT. So the time error calculation is:

• 1ppsTE = T_{1ppsMeasEgress} - T_Ref

However, the test equipment can only timestamp the pulse when it arrives at the instrument. This is later than the exit from the DUT due to the delay through the measurement cable. So:

• TS_{1ppsMeasEgress} = TS_{1ppsMeasIngress} – D_cable

So we have:

• 1ppsTE = (T_{1ppsMeasIngress} – D_cable) - T_Ref

If the DUT time lags (is behind) the reference, then T_Ref will be later than T_{1ppsMeasEgress} and the time error will negative; if the DUT leads (is ahead of) the reference, then T_Ref will be earlier than T_{1ppsMeasEgress} and the time error will be positive.

PTP Measurements

PTP time error is calculated from 4 timestamps:

• T1 and T4 are from the DUT. These are the measured timestamps

• T2 and T3 are from the test equipment. These are the reference timestamps

To make an accurate time error measurement, the test equipment must know the delay to the DUT – the cable delay D_cable.

So, T1 Time Error is calculated using measured minus reference with a correction for the cable delay:

• T1TE = (T1 + D_cable) - T2

T4 Time Error is calculated using:

• T4TE = (T4 – D_cable) – T3

2-way Time Error is calculated by adding the T1 and T4 Time Error results and dividing by 2:

• 2WayTE = (T1TE + T4TE) / 2

If the DUT time lags (is behind) the reference, then both T1 and T4 will be smaller than T2 and T3 and the time error will negative; if the DUT leads (is ahead of) the reference, then both T1 and T4 will be greater than T2 and T3 and the time error will be positive.

Effect of measurement cable delay

Conceptually, the effect of the cable delay can be visualised as:

A measured value (shown as a Red number here) includes the time taken for the signal to pass through the measurement cable i.e. the cable delay (shown as a Green line).

Therefore, the actual value (shown as a Green value) is represented by the sum of the measured value and the cable delay i.e. the cable delay is added to the measured value to give correct measurement.

Related Articles