RX3i Genius Communications Gateway (GCG) Fault status data issue

[yorkiiii]

Hi all, I am currently working with a system that uses genius blocks. We are connecting these genius blocks to new RX3i hardware using GCG modules. Currently we are having problems using the GCG Fault Status Data (32 bits long). The fault status data is supposed to indicate if there is a fault (fault = 1) on any of the 30 connected genius blocks (each address corresponds to one genius block fault status). The fault status data works well at first, when a block goes into a fault state the corresponding bit turns from a 0 (all good) to a 1 (fault state). The only problem is that after the fault has been cleared from the genius block AND the fault table in the CPU (CPE330 in this case), the corresponding fault status bit does not turn back to a 0 (essentially showing that block in a permanent fault state even though it is now fine). I was wondering if anyone had any knowledge to offer on this, we really need those fault status bits to work (they work using the old BEM modules, why can't GE/emerson make anything new that works properly??!). Thanks!

 

[Steve Bailey]

Can you toggle the fault bits off using Machine Edition software? If so, you might be able to set up a work-around in ladder logic by monitoring when the #IO_FLT bit transitions from high to low and writing zero to the bits when it happens.

 

[yorkiiii]

That was my first thought as well but I cannot toggle those bits. The GCG is actively writing a 1 (fault) even when there is no longer a fault on any of the genius blocks. We just found that unplugging the GCG from the profinet controller and then plugging it back in will also clear the faults. I am thinking that the GCG pulls the fault data from the CPU when it is first initializes the connection and then after that it only sends fault data. It never receives/clears fault data when the CPU clears them.

 

[Steve Bailey]

Are there any COMREQ commands available for the GCG module?

 

[yorkiiii]

A great recommendation! I should have thought about that. I just tried using the clear all bus faults comm_req and it worked beautifully. I have already contacted emerson and told them that they should possibly either change the way these fault status bits work or document in the manual that you must use a comm_req to clear these faults. Thanks so much!

 

[Steve Bailey]

Sometimes it just takes another pair of eyes.

 

[yorkiiii]

Just to close out this post with the emerson recommendations, they did get back to me and they basically said that you must use the commreqs. Even when the HHM is used to clear the fault on the actual genius block, this clear never makes it up to the GCG. The ONLY way to clear the GCG of its fault status is by using the Clear Circuit Fault COMMREQ or the Clear All Circuit Fault COMMREQ. These details are in sections 4.1.4, 5.4.2, and 6.4.6 of the GCG manual (GFK-2892G). Hopefully this helps somebody out in the future!