CLX User-Defined Major Fault

Dog7926 · Nov 11, 2010

First time poster so I apologize in advance if this topic has been discussed before. I did several searches but came up empty.

I am trying to create a user-defined major fault in RSlogix 5000 but after reading the programming manual, I am still left with questions. The manual says to:
1. Create a fault routine for the program
2. Configure the program to use the fault routine
3. Jump to the fault routine when the fault occurs.

My question is: What should the logic in the fault-routine do? Am I missing something?

Thanks in advance for any advise.

kamenges · Nov 11, 2010

Just so we're on the same page, a fault routine doesn't trigger a fault. It responds to a fault. It seems to me that fault routine mechanics are outside the scope of your original question.

Unless, that is, you already know how to trigger the fault and you need to figure out how to respond to it once it occurs. Although, that seems somewhat unnecessary when I say it like that.

Keith

Dog7926 · Nov 11, 2010

Sorry if my question was a unclear. Let me try again.

Once I detect the condition that I want to define as a major fault, how do I go about faulting the processor to stop the execution of the logic?

If I understand you correctly a fault routine would be executed after the fault has occurred, which is how I understood it. So I was thinking that once the fault condition is detected, jump to the fault routine and fault the processor. Am I on the right track? Sounds to me like the answer to that question is "no".

kamenges · Nov 11, 2010

As you said, the answer is no. The fact that a major fault is detected in the first place is enough to fault the processor. A fault routine may give you the ability to prevent the processor fault (depending on the fault detected) but it can't trigger a fault.

I've nver tried to do this before. I don't know if you can trigger a user fault or not.

Keith

kamenges · Nov 11, 2010

OK, I've got to ask. If you have already detected something worthy of a processor fault, isn't there a better way to handle it that killing the one device that may help you handle it?

Keith

Dog7926 · Nov 11, 2010

According to the Logix5000 Controllers Major and Minor Faults Programming manual (1756-PM014) page 29, "If you want to suspend (shut down) the controller based on conditions in your application, create a user-defined major fault."

It goes on to explain the steps that I listed in my original post, but I must be missing something.

Anyway, thanks for the effort. It is appreciated.

dash · Nov 11, 2010

After you create the Fault Routine, then in your normal program you will create a ladder that contains the conditions required to trigger a fault and at the end of that rung you will call a JSR to that fault routine and configure an input parameter that will be your fault code (Like 999). Then when someone goes online the major faults tab will list your fault code, 999.

Darren

The manual you reference has an example of how to do that.

Dog7926 · Nov 11, 2010

Understood. This is a client-driven requirement. The view is that because this is a safety-critical system, when this fault occurs, the system is in an unreliable state, therefore all logic execution should be stopped.

Dog7926 · Nov 11, 2010

Right, but what I don't understand is what should happen in the fault routine. The manual doesn't explain this. Is there any logic that I need to have in there? If so, what should it do?

dash · Nov 11, 2010

You do not need anything in there unless you want it to do something before it faults the controller (like reset something or maybe a counter to keep up with how many times you have triggered this or an array that contains a timestamp for the last X times this has occurred). You may need to put in a NOP (no operation) instruction to get it to compile.

Darren

mellis · Nov 11, 2010

OK, let me see if I got this right...

You trust the PLC to identify this safety critical situation when it arises.

You trust the PLC to take the appropriate action when this safety critical situation occurs.

The best you can think of as an appropriate action is for the PLC to lobotomize itself.

Seriously, if this is a safety critical situation you should be using the proper risk analysis and determining exactly what the appropriate response should be. If it's truly a safety critical system, the PLC is probably NOT where that action will take place.

Dog7926 · Nov 11, 2010

Great! I thought that might be the case but do not have a way to test right now. Thank you very much.

dash · Nov 11, 2010

I agree with mellis and Keith. This sounds like the customer may be asking for an incorrect device to be handling a safety task. Maybe you need a safety rated controller or a hardwired safety circuit. The ControlLogix can be used in a LIMITED way for a SIL-2 rated system, but you have to use certain rated I/O modules and make sure you application does not pose any more risk than than.

See this manual for a discussion of uses for ControlLogix in safety systems.

http://www.rockwellautomation.com/products/certification/safety/pdf/1756-RM001E-EN-P.pdf

Darren

mellis · Nov 11, 2010

OK, now to the fault handler...

The fault handler is a means to intercept a fault condition BEFORE it faults the processor and deal with it some way.

If you are trying to fault the processor, one easy way to do it is to move a negative number into a timer preset. That will cause a major fault.

dash · Nov 11, 2010

actually calling the fault routine and passing a user defined fault ID will fault the processor and it will show up in the major fault tab of the processor with that code. The fault routine can be used for both functions.

CLX User-Defined Major Fault

Dog7926

Member

kamenges

Member

Dog7926

Member

kamenges

Member

kamenges

Member

Dog7926

Member

dash

Lifetime Supporting Member

Dog7926

Member

Dog7926

Member

dash

Lifetime Supporting Member

mellis

Member

Dog7926

Member

dash

Lifetime Supporting Member

mellis

Member

dash

Lifetime Supporting Member

Similar Topics