[etherlab-users] EtherLab and Beckhoff Ethercat on netX500

Florian Pose fp at igh-essen.com
Thu Feb 28 11:00:35 CET 2008


Hi Lukas, hi Thomas,

On Thu, Feb 28, 2008 at 09:32:27AM +0100, Thomas Elste wrote:
> > we are trying to establish a communication between EtherLab and a
> > proprietary Bootimage from Hilscher containing an Ethercat slave.
> the EtherCAT slave firmware for the NetX from Hilscher has some
> problems in working with the IgH master. The main issues are
> the (mostly) missing slave information in the (emulated) EEPROM of
> the NetX and the fact that the NetX only supports LRD/LWR datagrams
> (instead of LWR) which the igh master does not.
> If your are interested in the details, read my posts on this list:
> http://lists.etherlab.org/pipermail/etherlab-users/2007-July/000016.html
> or in the Hilscher forum concerning some of these things:
> http://board.hilscher.com/viewtopic.php?p=886

I heard this question hundreds of times before and therefore created a
FAQ list with a single entry (for now...) :-)

http://etherlab.org/en/ethercat/faq.php#slave

> If you're interested, I can send you a patch for the master enabling
> communication with the NetX. I think I've posted it before on this
> list but that was probably for some older NetX EC slave firmware, so it
> will not work anymore.
> But this patch is just a workaround. Imho it would be a better solution
> implementing the optional LRD/LWR datagram types in the IgH master and
> to bug Hilscher as to implement a clean SII interface in their NetX
> firmware.

This is already done for the next release. The approach is, that a
domain uses the LRD datagram, if it contains only inputs, the LWR
datagram, if it contains only outputs, otherwise the LRW datagram. So
you'll have to configure separate domains for inputs and outputs (which
is generally a good idea).

Here is a link to a preliminary version of the 1.4.0 release:

http://etherlab.org/download/ethercat-1.4.0-pre-trunk-r1047.tar.bz2

-- 
Best regards,
Florian

http://etherlab.org



More information about the Etherlab-users mailing list