SLC 5/03 Faults when Testing Edits

[scarince]

Hi all,

I have a power & free conveyor line with a series of machines / stations around it. Each station is controlled with a 5/03 connected to each other via DH+ and some e-stop hardwiring. This line has been running like this for about 15 or 20 years.

One of the 5/03's will often (but not always) fault during online editing. At the point when you "test edits" the PLC will fault. This kills the whole line....you can imagine the headache.

We don't know where to look to understand what the fault is or why it is happening. We lose communication when the fault happens and the only way we have ever been able to clear the fault is to cycle power on the plc. When it comes back up, the edit is still there and it is active.....the "test edits" command apparently is successful before the PLC faults.

We've changed the processor, and it still happens!

We can't figure out what has changed or why this happens, or what the fault is.

Does anyone have any suggestions on thing I could look at next to try to narrow this down?

I'll note that the DH+ is "active". I don't know how busy it is, but there is a lot going on with HMI's and MSG's moving status data all over the place. I can't shake the feeling that this is somehow related to communication, but I have no factual reason to think that.

I'll take any ideas.

Thank you.

Bill

 

[Aabeck]

Search the forum, I seem to remember this exact issue from before & I think it had something to do with a RSLinx setting.

What versions of RSLinx & RSLogix500 are you using.

 

[scarince]

Searchh the forum, I seem to remember this exact issue from before & I think it had something to do with a RSLinx setting.

I tried a search before posting but couldn't find anything like this related to editing.

What versions of RSLinx & RSLogix500 are you using.

I'm running RSLinx Classic Lite revistion 3.70.00 CPR 9 SR 7 and RSLogix500 version 9.00.00 CPR 9

Bill

 

[Aabeck]

Check this thread, the post about the RSLinx setting after the Ethernet posting:

http://www.plctalk.net/qanda/showthread.php?t=29662&highlight=online+edit+fails

And later in the thread there's a post about 5/04 memory/firmware revision.

 

[scarince]

Check this thread, the post about the RSLinx setting after the Ethernet posting:

http://www.plctalk.net/qanda/showthread.php?t=29662&highlight=online+edit+fails

And later in the thread there's a post about 5/04 memory/firmware revision.

Aabeck, this is great! You've given me plenty to look into. I'll work on this for a few days and report back on what I find or rule out.

Thanks!

 

[Mickey]

The SLC5/03 does not support DH+.

How are you connected to the controller?

Have you tried to program it thru its serial port?

 

[scarince]

The SLC5/03 does not support DH+.

How are you connected to the controller?

Have you tried to program it thru its serial port?

Mickey, You are 100% correct......I'm using a *5/04*, NOT a 5/03!

The comm port is plugged into the local HMI. We're connecting through DH+ through a network port on the side of the machine.

I am embarrassed by the mistake. Thanks for catching that.

I think it's too late for me to edit my first post to make this correction.

Bill

 

[Mickey]

Can you tell us more about the kind of edits you are trying to make that faults the PLC.

Have you looked at the processors status under its error tab to see what the fault (error) is? See link

http://www.plctalk.net/qanda/attachment.php?attachmentid=36745&d=1431533424

 

[Aabeck]

See, when you speed-read the original post your brain assumes it was a 5/04, had to look back & see it said 5/03.

 

[scarince]

Can you tell us more about the kind of edits you are trying to make that faults the PLC.

Have you looked at the processors status under its error tab to see what the fault (error) is? See link

Once the fault occurs we lose all comms with the PLC, so I can't see what it was. Since we cycle power to recover, we lose that info.

Unless.....is there a way to maintain that fault data through the power cycle?

 

[James Mcquade]

1.Check the entire program foe edits. if there are 2 maybe the other one is creating the fault.

2. you may have to save the program to the hard drive, and recompile it. then download back to the plc.

3. What version was the program created with and what version
are you working with now?? this can make a difference.

james

 

[Mickey]

Connect to the serial port.

 

[scarince]

Can you tell us more about the kind of edits you are trying to make that faults the PLC.

I forgot to answer this......we can change timer values but that's about it. If I try to add a new rung or just change bit from .DN to a .EN (as an example), either one can cause the fault.

3. What version was the program created with and what version
are you working with now?? this can make a difference.

This was originally written in the 1990's, so it would have been some early version of the software. I'll check the program for other edits as you mention.

Connect to the serial port.

I suppose I could do that.....I would need to unplug the HMI, but I don't think the machine will care. I'll give that a try.

Thanks!

 

[James Mcquade]

If the original program was written in the 1990's and you are running the latest version of the software, then you could get the errors you are talking about.

The newer version of software is not 100% compatible with the older version (so I was told by AB tech support) when we upgraded a plc 5/03 to a 5/05.

There may also be a firmware issue.

I would wait for an opportunity when the line is down and save the program to the pc, recompile it, and download it back to the plc.

regards,
james

 

[Geospark]

...I'll note that the DH+ is "active". I don't know how busy it is, but there is a lot going on with HMI's and MSG's moving status data all over the place. I can't shake the feeling that this is somehow related to communication...

Quite possibly communications alright. Have a look here...

14319 - SLC 5/04 -- Data Highway Plus Locks-Up and the Data Highway Plus LED Illuminates Solid Red
Access Level: TechConnect

Regards,
George