Re: [foxboro] CP-60 Fieldbus Communication Errors

We had the same problem (I work in the same plant as Ms. Heckathorn, but 
in a different unit.  Congrats, Tracy!  No more of that lurking stuff. Now 
that you're out of the closet, we expect at least a weekly posting  :). 
The problem was limited to CP60s.

The document Shelby and Tracy mentioned is probably what I've seen.  It 
mostly deals with grounding and noise issues on the coaxial fieldbus 
connections.  I thought it was all hogwash, but our field service guy came 
out and performed the retrofit, and the problem has been significantly 
ameliorated.  We were getting PIO bus failures of almost all FBMs on a CP 
at once, similar to what your system log shows, at a frequency of once a 
week or so.  The retrofit involves installing an isolated DIN rail section 
to mount the coax connector bulkheads only, and grounding this piece of 
rail with a dedicated ground wire (on one end only).  They also put chokes 
(3 of them!) around the little white coax cable segments that come out of 
the CP.  Finally, we made sure the CPs were at rev M (don't remember if we 
had to replace any).


Other things I tried: I was also told to EEPROM the FCMs to the newest rev 
(1.13 is what my SysMgmt says now).  I was also told to go to 6.5.1.  I 
did both of these before the retrofit document/kit came out, and they did 
not help.


I also was told to EEPROM the CPs as well (to 3.30 I think is what it 
ended up being).  We did it on our last turnaround, after the retrofit kit 
was installed, but haven't run long enough to see what additional effect 
it had.


Good luck,
Corey Clingo
BASF Corp.







tom.vandewater@xxxxxxxxxxxxxx
Sent by: foxboro-bounce@xxxxxxxxxxxxx
03/22/2005 08:08 AM
Please respond to foxboro

              To:  foxboro 
              cc: 
         Subject:       [foxboro] CP-60 Fieldbus Communication Errors






All knowing list,
I have a CAR in the que regarding intermittent fieldbus
communication errors associated with CP-60 fieldbus devices.  We have had
continual problems with this since we started using CP-60's at Ver 6.2.1.
We have been through a host of EEPROM updates for the FCM/FBI/DCM's, 
(1.07,
1.09, 1.11, 1.12, and 1.13).  The current recommendation to fix the 
problem
is that we upgrade our workstation software to Ver. 6.3.2 and our CP-60
software to 6.3.2 or 6.5.1 and then move our FCM/FBI/DCM, (iom82), EEPROM
level back to 1.11.  Our plant is a 24-7-365 operation and we have been
running at capacity so we have not been able to reboot the CP's since the
recommendation was made at the end of 2003.
As a result, I am asking all of you if anyone else has experienced
this problem?  If so, how has the problem been resolved?  Since I've 
already
been on many wild goose chases regarding this issue, you might imagine 
that
I am a bit skeptical when another fix is proposed that tells me I need to
install more software and go back to a previous EEPROM rev for the
FCM/FBI/DCM's.  Is there anyone out there running Ver. 6.3.2, (or later),
workstation software and 6.3.2 or 6.5.1 CP-60 image, and are still
experiencing intermittent fieldbus communication errors on CP-60's?  Your
help would be greatly appreciated.

Below is a description of our problem:

Since Foxboro has been unable to identify and fix the problem we
find ourselves deeply involved in committing OUR resources to nail down as
many of the symptoms as we can.  Not being privy to the communication
structures and error detection and reporting strategies that only Foxboro
knows, we are much like the blind man trying to describe an elephant by
touching it.  However, we have touched this elephant enough to develop 
some
theories.
Here are some of our findings:

Once all FCM's, FBI10E's, and DCM's were EEPROMed at rev 1.13, these
are the symptoms:
* Continue to have large numbers of PIO Bus Access Errors if PIO bus
switching is enabled.  (10 to 100/hour).  We have ZERO errors when running
"A or B Disable" except for the complete but temporary failure experienced
when running "A Disable" at EEPROM 1.11 on the new DCM's.
* The errors are spread out across CP's using all three types of fieldbus
communication modules, FCM10EF, FBI10E, and DCM10E's and EF's.
* The "majority" of the errors occur on CP's communicating through DCM's
* Most of the legacy FBM's are at EEPROM rev. 61.1 with a few at 40.1 (As
far as repetitive errors are concerned we see no difference between the 
two
EEPROM rev's.)
* The majority of the errors occur repetitively on the same, relatively 
few,
FBM's
* The majority of those errors are associated with legacy, discrete, 
FBM's.
(Our discrete FBM's run PLB ladders)
* Replacing the repeat offender FBM's with newer or different FBM's of the
same type does not stop that FBM from reporting errors.
* Moving the repeat offender to a different Y-Adapter does not stop that 
FBM
from reporting errors.
* Changing the CP's Primary ECB .MPOLL parameter from 7 to 0 had no impact
on the frequency of the errors, but it also looked like my change from 7 
to
0 had no impact on the Fieldbus  Scan Loading as viewed in the CP's 
Station
Block.  It should have made a big difference and I wonder if CP-60's only
use .MPOLL=7?
* The CP's involved are not experiencing any Control Loading or OM Scanner
loading overruns as displayed by the CP Station Block.
We have a few FBM's that are experiencing repetitive errors that we can
replace/reboot/move/EEPROM, with no adverse effect on the process and we
invite Foxboro to observe/troubleshoot using those.  We also have all 
System
Alarm logs over the past year that can be reviewed/analyzed.  It is our
conclusion that there are serious errors in the CP-60 fieldbus 
communication
and bus-switching strategies that will only be fixed when Foxboro decides 
to
put some concerted effort into solving them.  Our time is now, better 
spent
communicating with other users to heighten the awareness so they can 
submit
CAR's on these problems.
These issues have plagued us for over 4 years and we want to see them 
fixed.
Below is a typical System Alarm log that shows one hour before PIO bus
switching is disabled.  There are 12 PIO Bus Access Errors during that
timeframe.  The CP's then ran for over 7 hours with no PIO Bus Access
Errors.  After PIO Bus Switching is re-enabled there are 16 PIO Bus Access
Errors in a one-hour period.





 
 
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
 
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave

Other related posts: