For all you S7 gurus

bazzab1 · Sep 6, 2007

We have a system set up as follows:

2 X S7 417H Prcessors on two racks set up for redundancy
2 X CP443-1 communications modules (1 on each rack)

The redundancy between the processors works perfectly. When we switch one processor off the other takes over, very nice.

We are using Citect to communicate with the both PLCs over ethernet. Citect is using a backend/frontend driver called PSDirect. The backend is a third party driver by ProScada. Our problem is that the backend driver does not seem to recognise that that processor has stopped and continues to get frozen data from the stopped processor. If we remove the ethernet connection from the CP443-1 module associated with the stopped processor then the driver swaps to the standby.

Anybody had any experience with this?

JRW · Sep 6, 2007

Siemens has a product called Redconnect.

I have no experience with it, but it might work for your app.

PeterW · Sep 7, 2007

A simple method maybe to create a heartbeat from both PLC's, the SCADA would then see that one of the PLC's has stopped then.

Marsupilami · Sep 7, 2007

Hello,

The proscada driver forsees a heartbeat parameter.

from the manual:

The S7 driver continues polling normally when the PLC is in Stop mode resulting in frozen
values on the screen. To force #COM’s when the PLC is in stop mode requires
configuration of the Heartbeat monitoring feature..
For each IO device (Unit) configured, a parameter of the following form is entered into the
Citect.INI file
HeartBeatAddress_<devname>=
Where <devname> is the name of the IO device in CitectSCADA and in the backend driver
And <address> is a PLC address of a variable that MUST change very 5 seconds at least if
the PLC is running. A simple way of achieving this is to increment an integer in OB1 of the
PLC scan cycle. ** the address format here is the same as the OPC server so you specify
W for Integer & D for longs etc.
Examples
HeartBeatAddress_S7PLC=MW0
HeartBeatAddress_PumpHouse=DB100,W44
HeartBeatAddress_Furnace2=DB130,D44 (long integer)

Don't forget you need to get the data from the acitve PLC if the link is broken between the PLC's.

regards,

Marsupilami

JRW · Sep 7, 2007

Its a little more complicated than that right?
His SCADA needs to switch IP addresses (from the stopped CPU to the running)
This is what Redconnect does.

Marsupilami · Sep 7, 2007

True,

it is a bit more complicated...

Thats why we used redconnect to connect to the redundant PLC pair so you dant have to worry about all the exceptions.

i am guessing that bazzab wants to save some money by using the psdirect driver.
You save:
- the special siemens CP card
- the license for redconnect
and mostly you save this twice because with redundant PLC's you also use redundant SCADA.

bazzab1 · Sep 8, 2007

Thanks for the comments guys,

The heart beat is used by the Citect frontend to indicate that the processor has stopped by putting "hash/COMS" on all animated values. There would seem to be no way to force the driver to switch from Citect.

The reason we went with the PSDirect driver was that it was sourced from one vendor, Citect. I have heard horror stories of people setting up comms between S7 and Citect using the CP card or using SimaticNet sofware. It seems that you get buried in cumbersome arrays etc. The PSDirect seemed to be a nice clean solution. I have logged a CSR with Citect support so we will see what they come up with.

Will update if I get anywhere

piwi447 · Sep 12, 2007

testing of redundancy

I have a project in next few month where i have to connect citect in the same configuration.

Citopia (the citect korean supplier and integrator) plan to use also the PS direct without redconnect nor CP1613 card.

I got information from citect confirming this choice.

I also knows from discussion with belgian siemens specialist that stopping one of the CPU in order to test the redundancy swith is not the best way(a bad case that would probably never occur). When doing this, the CP communication task are kept on.
In this case, the first CP card is still working.

I thought the IP connection redundancy was managed by the PSdirect, Is it finally working good. thank you for your feedback.

Peter

For all you S7 gurus

bazzab1

Lifetime Supporting Member

JRW

Member

PeterW

Member

Marsupilami

Member

JRW

Member

Marsupilami

Member

bazzab1

Lifetime Supporting Member

piwi447

Member

Similar Topics