[etherlab-users] Info on testing the performance of EtherCAT

[available]

I was interested in measuring this time of the order of few microseconds
using a hardware or some other mechanisim. One way that I figured out was
setting up highly synchronized clocks using PTP, between the Master and the
slave nodes and measure the difference in the transmission times for latency
and jitter values. But is there anything method more simpler?

/Able



On 10 May 2011 15:49, Armin Steinhoff <armin at steinhoff.de> wrote:

>  available wrote:
>
> Hi,
>
> I am interested in testing the latency of ethercat systems, please let me
> know how to do it with a system with one EtherCAT Master and a couple of
> Digital slave I/O nodes. I am looking at the latency of communication
> between Master and Slave node.
>
>
> The "latency is mainly defined by the size of the transmitted Ethernet
> packages. The transmission of a packet with a minimum packetsize of ~60
> bytes is in the range of 5us.
> You have still to add the cycle time of the master statemachine ... so the
> absolute smallest latency is then in the range of ~15us. With a full packet
> size you have to count with a latency of 150 -200us.
>
> Cheers
>
> --Armin
>
>
>
>
>
> /Able
>
>
> _______________________________________________
> etherlab-users mailing listetherlab-users at etherlab.orghttp://lists.etherlab.org/mailman/listinfo/etherlab-users
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20110510/318dd658/attachment-0005.htm>