Latency and jitter for different RS-232 converter technologies
About the publication
ISBN
9788246412184
Size
228.7 KB
Language
English
This report present the results of different tests of data transmission time when using RS-232 converters. One RS-232 to USB and two different RS-232 to Ethernet converters have been tested. The purpose of the tests has been to gain an insight into how well these technologies are suited with regards to transmission of time critical data. The tests are not meant to be a product comparison.
The RS-232 to USB converter is almost as fast as a real serial line, and does not increase the variation in transmission time significantly. It can safely be used for time critical data.
The RS-232 to Ethernet converters are of the multiport type, with respectively four and sixteen serial ports. They have been tested with high and low load on these ports, and with high and low load on the Ethernet (LAN). The results show that they are mostly unaffected by traffic on the LAN, but both are sensitive to the load on the serial ports. Both average transmission time and the variation in transmission time increases when many ports are active at the same time, and the delay may become greater than acceptable. In addition, both converters are sensitive to their configuration parameters.
The Ethernet converters’s sensitivity to serial port load and configuration leads to the following recommendation: If serial to Ethernet converters are to be used to transmit time critical data, tests should be performed on the products selected for installation using representative loads on the serial ports and the LAN, and using the same configuration as they will have in the finished installation.