FarSync BERT | ![]() |
![]() |
---|
FarSync BERT - Test Functions
The FarSync BERT supports two test functions which are described below:
This is the standard BERT configuration and is used to test the quality of both synchronous and asynchronous lines. Industry-standard test patterns are used to identify bit error rates for the line under test. The results of the tests are collated and displayed as industry-standard metrics e.g. Errored Seconds (ES) etc.
In addition to the standard configuration, the FarSync BERT can also be configured to test the data paths of multiple nodes simultaneously on a Multi-Drop network using Industry-standard test patterns.
The Multi-Drop test must always be run in software mode so it overrides the BERT Support setting on the General tab of the Additional Configuration Dialog. The Multi-Drop test can be performed with a FarSync Flex V2 and above
The Multi-Drop test is between a Master node and one or more Slave nodes. Each slave node on the network has a unique Slave Id in the range of 1-247. The master sends the data pattern to each configured slave and waits for the slave to echo the data pattern back. The master then checks the received data pattern for errors to determine the BER for that slave. The list of slaves to be sent packets is configured in the Slaves To Poll text box.
The data pattern is sent to each slave as the payload of a packet. The length of data pattern sent in each packet is configured by setting the message Payload Length .
The packet header contains the Slave Id of the slave to send the packet to. When the packet is transmitted, it is seen by all of the slaves on the network, but only the addressed slave can respond and echo back the message.
The packets are sent to each slave, one at a time, waiting for the echoed reply before sending the next packet to the next slave.
If the master fails to receive a packet from a slave, the slave is deemed to be Unresponsive. If the Poll Unresponsive Slaves checkbox has been ticked, the slave will remain in the list of slaves to poll and will be polled on its next turn. If the checkbox hasn’t been ticked, the slave is removed from the list of slaves to poll and takes no further part in the current test. If all slaves are removed from the list of slaves to poll, the test is terminated.
Two interfaces can be used for the Multi-Drop test. RS485-2Wire and RS485-4Wire. In the 2-Wire interface the data is echoed back along the same data lines on which it was transmitted. In the 4-Wire interface the transmit and receive data lines are kept separate.
In the Multi-Drop mode, the auto-invert feature can be used to determine if the connection to a slave has been wired incorrectly.
To detect crossed wires Manchester Encoding signaling must be used.
An additional Round Trip Delay mode enables the BERT to measure the Round Trip Delay of packets to a resolution of 1ms.
In the 2-Wire configuration, if a slave’s D+ and D- wires have been incorrectly wired, the slave won’t be able to detect and echo back packets which have been addressed to it and therefore won’t be able to achieve sync.
If the slave fails to achieve sync (within 2 seconds worth of its time slice), the FarSync BERT inverts its TX and RX data. Whilst the TX and RX data are inverted, the inversion indicators TIV and RIV are displayed in the Inverted column in the Multi-Drop Results dialog.
In the 4-Wire configuration, if a slave’s RD+ and RD- wires have been incorrectly wired, the slave won’t be able to detect messages addressed to it. If a slave’s TD+ and TD- wires have been incorrectly wired, it won’t be able to echo the data back successfully.
If the slave fails to achieve sync (within 2 seconds worth of its time slice), the FarSync BERT inverts its TX and RX data. Whilst the TX and RX data are inverted, the inversion indicators TIV and RIV are displayed in the Inverted column in the Multi-Drop Results dialog.
If sync is achieved whilst one of the inversion indicators is displayed, it means:
Note that the TIV and RIV indicators in the signals pane are not used whilst the FarSync BERT is in Multi-Drop mode.
Whilst the test is running, the results are shown in the Multi-Drop Results dialog. The results for each configured slave are shown along with a merged ‘Total’.
The Round Trip Delay mode is supported by the FarSync V2, and above, Flex only.
The Round Trip Delay test measures how long it takes for a packet to be received after it has been transmitted. The timestamps are taken at the hardware level and the measurements provide a resolution of one millisecond.
Packets are transmitted one at a time with each packet being transmitted as soon as the previous packet is received or after a configured Transmit Delay.
If a packet is transmitted but not received within the configured Receive Timeout, the packet is declared as lost.
The packet comprises a header and a payload. The payload length is configured by setting the Payload Length. The payload contains a fixed data pattern. The first byte of the payload is the sequence number and subsequent bytes contain an incrementing byte pattern.
Round Trip Delay - Measurement
The accuracy of the RTD measurement is subject to limitations imposed by the hardware. Whilst every effort has been made to compensate for the delays through the hardware, there is still a degree of error in the measurement which is dependent upon the line rate.
The error is due to the varying difference in time between taking the transmission timestamp and the actual point in time at which the packet starts to be transmitted.
The delay can be up to 8/R (where 8 is the number of bits and R is the line rate in bits per second).
Some example line rates and their maximum measurement errors are shown below.
Line rate (bps) | Maximum possible 8/R error (in ms) |
300 | 26.6 |
2400 | 3.33 |
9600 | 0.83 |
64000 | 0.125 |
128000 | 0.0625 |
You should keep this in mind when selecting the line rate for any Round Trip Delay measurements.
For further details on how to configure and use the FarSync BERT click here.
Copyright © FarSite Communications Ltd. 2009-2020 |
![]() |
---|