Skip to Content

SiePortal

The integrated platform for your product selection, buying and support workflow - bringing together Industry Mall and Online Support.

SiePortalDeutsch

Search

Home
Support Support
{{(apiData.Support | filter:{'rank':1}:true)[0].title}} Support
- Menu temporary not available
- Menu temporary not available
- Menue temporary not available
- {{item.title}}

Products & Services
- Menu temporary not available
- Menue temporary not available
- {{item.title}}
Support Support
{{(apiData.Support | filter:{'rank':1}:true)[0].title}} Support
- Menu temporary not available
- Menue temporary not available
- {{item.title}}
MySiePortal MySiePortal
{{(apiData.MySiePortal | filter:{'rank':1}:true)[0].title}} MySiePortal
- Menu temporary not available
- Menue temporary not available
- {{item.title}}
Cart ({{apiData.CartCount}})

Navigation

Reset value OB 86

Created by: Andre_it at: 8/24/2010 5:09 PM (8 Replies)

Rating (1)

Thanks 0

9 Entries

8/24/2010 5:09 PM	Rate (0)
Andre_it Posts: 5 Rating: (0)	Hi, I'm new to this forum, so please bear with me. I use OB 86 for detect an error in DP slave. OB86's values are stored in a dedicate database. Those values will make a emergency shutdown. But now I have this problem: - OB86's values don't return at normal value when DP slave is Ok. This situation doesn't restore the emergency shutdown. What's the correct situation when used OB86? Can you help me? Thank you Regards
Andre_it Posts: 5 Rating: (0)
Suggestion To thank Quote Answer

8/24/2010 5:25 PM	Rate (0)
dchartier Grandmaster-Expert Joined: 9/27/2006 Last visit: 1/15/2026 Posts: 12481 Rating: (2722)	Hello André; Which OB 86 values do you store in your database records? OB86, like all diagnostic OBs, collects data for the diagnostic buffer whenever a new event talkes place that concerns it. Each new event is recorded with a timestamp and an explanation, so the events do not get reset. You could use OB86_EV_CLASS identifiers to determine if the event recorded is a new (incoming) or completed (outgoing) event. In case the OB86_EV_CLASS value is B#16#38, the remote rack problem has been cleared, so you could reset the contents of your database if you require. However, in case of multiple remote rack problems, or if the problem is intermittent (many incoming/outgoing events in succession for the same rack) then you would lose valuable information... You have to choose how you want to collect your data. Hoep this helps, Daniel Chartier
	Last edited by: dchartier at: 8/24/2010 5:26 PM
Suggestion To thank Quote Answer

8/24/2010 5:45 PM	Rate (0)
Andre_it Posts: 5 Rating: (0)	Hi daniel, thank you for answer. I think it, too. But in my simulation if I go out a module and replace it, OB86_EV_CLASS and OB86_FLT_ID don't change. They stay 16#38 and 16#C4. I don't understand why the values don't change. For this reason I suppose a my error. Can you help again? Cheers Andrea
Andre_it Posts: 5 Rating: (0)
Suggestion To thank Quote Answer

8/24/2010 7:38 PM	Rate (0)
dchartier Grandmaster-Expert Joined: 9/27/2006 Last visit: 1/15/2026 Posts: 12481 Rating: (2722)	Hello André; Could you please explain more about your project? You mention a simulation; how do you simulate removing a physical module from the rack/backplane? has this been tried on a real hardware? If so, is OB86 the only diagnostic OB called by the system on this type of event? hope this helps, Daniel Chartier

Suggestion To thank Quote Answer

8/25/2010 11:44 AM	Rate (0)
Andre_it Posts: 5 Rating: (0)	Hi Daniel, my project is a little automation (start/stop) of hydro power plant. For company policy, the CPU is 319F (failsafe mode) e HMI is a MP377. Standard automation software is in CFC, emergency shutdown software is in F-KOP and HMI software is in Wincc Flexible. For hw diagnostic I have used SFB49-50 (I saw this solution on this site). I need to detect a hw fault or an another fault situation, for to active emergency shutdown without stop CPU. For those reasons I created OB82-OB83-OB86. In my lab I assembled all hardware and I test it. When I extract a module from the rack, only OB86 is actived and when I insert it the same (I saw CPU status). The OB86 values are the same. I don't understand. If you have another solution for detect faults situation, you do not hesitate to write me. I hope you can have a answer for this problem. Cheers
Andre_it Posts: 5 Rating: (0)
Suggestion To thank Quote Answer

8/25/2010 7:56 PM	Rate (0)
dchartier Grandmaster-Expert Joined: 9/27/2006 Last visit: 1/15/2026 Posts: 12481 Rating: (2722)	Hello André; I have not worked with diagnostic blocks for CFC, so I am not aware of the information these SFB49/50 are supposed to generate when a fault is detected. Could you please describe, or provide a link to info on these blocks? Now, are the modules you are removing located in profibus slave racks? OB86 is designed to report failures of remote I/O nodes, so it should detect the disapearance of the module (in a ET 200M rack, for example) especially if it is located in the middle of the remote rack: you should lose all modules after the removed one (the bus connection is serial). If the module is on the main rack,I am not sure OB86 would be called by the operating system. Check the timestamps on the OB86 calls in the diagnostic buffer, just to see if it has been called, at a different moment. When you remove the module, does the CPU react in any way, a SF LED for example?How does the diagnostic buffer show, before and after removal? Try inserting a line of code that accesses a bit or a channel on the module you plan to remoxve, so we can see if OB122 is called in reaction. You do not have any Safety program in the CPU at this time, for your tests, do you? Hope this helps, Daniel Chartier
	Last edited by: dchartier at: 8/25/2010 7:58 PM
Suggestion To thank Quote Answer

8/26/2010 10:04 AM	Rate (0)
Andre_it Posts: 5 Rating: (0)	Hi Daniel, Below my answers: -My project has two decentral peripheries with ET 200 rack and one main rack (see file) -For a fault in AI module CPU SF led is already on. In fault simulation I extract a module from one ET 200 rack and for few second BF2 led turns on and then it turns off. -CFC software calls a channel on fault module, but OB122 isn't called. -CPU has safety program, CFC and KOP in the same time. Sorry, but some documents are in italian language. Attached zip file with: explain SFB49, project configuration, CPU status(diagnostic buffer) with list events. Thanks Andrea Attachment Project.zip (365 Downloads)
Andre_it Posts: 5 Rating: (0)
Suggestion To thank Quote Answer

8/26/2010 3:31 PM	Rate (1)
dchartier Grandmaster-Expert Joined: 9/27/2006 Last visit: 1/15/2026 Posts: 12481 Rating: (2722)	Hello André; Now I understand the "_it" in your username, I thought it meant IT, not Italy... OK, let's see what we have in the diagnostic buffer: First, OB86 was called twice, once for a Guasto Stazione 2event at timestamp 09.44.32.514, once for a Ritorno Stazione2 at 09.44.32.866; since you have only 10 evemnts recorded in the diagnostic buffer, we cannot saywhen suchearlier events have been recorded, but I have a feeling that they have, with a different timestamp. If you only record the content of the message you will not see the difference between the calls. Check that you can increase the size of the diagnostic buffer to 100 events or more in these CPUs, so try to set a higher limit on the recordings. Next, OB82 ( the diagnostic OB) has been called several times, for a "modulo/sottomodulo guasto", once for inpout address 256, once for output address 256, twice for input address 272, at different timestamps ( all around 9:42.37, so they seem related). These events arelinked to the submodules inanother module than the one you show in your screencapture, so you will have to identify them in your configuration. So, can you see a logical link between the calls of OB86 and OB82, in your manipulations? I mean, when you remove the submodule from the rack, are these events linked? And do your System Error messages show these record in your HMI display? Hope this helps, Daniel Chartier

Suggestion To thank Quote Answer

8/27/2010 11:41 AM	Rate (0)
Andre_it Posts: 5 Rating: (0)	Hi Daniel, OB82 called for analog input faults (addresses 272 and 256), because they are disconnect. This situation isn't linked with fault simulation. When I have extracted the module, the fault has called OB86. In attached file, you can found the diagnostic buffer. I extracted module at 10.00.11 (Events 1 and 2). Previous events are normal, because in my lab I can't simulate all signals. HMI showed "Slave PB2 Fault" (this event went on for few istant less than about 400ms, after it went down) and "Slave PB2, Address 13. Module extract". I see that in event 2 ID event is 16#39 and in event 1 ID event is 16#38. But CFC program doesn't see that change. CFC cicle time is 100ms. Is the small time of "Slave PB2 Fault" the problem? If it's, how do I resolve problem? Cheers Andrea Attachment Diagnose.zip (363 Downloads)
Andre_it Posts: 5 Rating: (0)
Suggestion To thank Quote Answer

|

Share this page:

Share this page on...

Follow us on