Online replacement of CPU processor card (CP334D-S3)

Your controller should be redundant, right? If it’s redundant, there’s no issue with replacing it \"online\". Manually power off the backup controller, replace it, restart it, and restore synchronization ； If redundant controllers fail simultaneously in pairs, first determine which one is the master controller at that time and immediately perform a manual switch between the master and slave controllers. If the switch is not successful, report to the workshop supervisor promptly and wait for an order to shut down the machine.

The original processor is broken; can the devices it controls still work? Is it interfering with the operation of the other systems? If the main equipment it controls can still operate in emergency mode without interfering with the operation of the other systems, then the replacement is simpler – just install a new controller with the necessary program, and it shouldn’t take long. If it does interfere with the other devices, it’s best to suspend operations temporarily to handle the issue; otherwise, things will become more complicated if the problem escalates

The 3# zhangjx_hc CPU card (CP334D-S3) is redundant and supports hot swapping; the other CPU card also works properly. At present, this fault has not affected the operation of the DCS system (it is operating under abnormal conditions; from a management perspective, the fact that critical equipment is not in standby mode represents a serious issue), and therefore it has not impacted the normal production of the plant (this facility is an important one in the plant, and any problems arising can have a significant impact). However, maintenance technicians have never carried out operations such as replacing core components of a DCS system online, and they lack the experience for this. This is exactly what we are concerned about and worried about. Stopping operations is of course the ideal solution, but is that possible for oil refining companies like ours? It’s almost impossible, and the leadership won’t agree either. Therefore, I sincerely hope that everyone can offer some suggestions to minimize risks as much as possible!

This post was last edited by jlshnlhj on 2010-3-26 09:38. The Yokogawa CENTUM CS system has been in use for about ten years now. 1. First, determine whether there is a “real” fault with the processor card: (1) Reset and restart the faulty processor card (there is a reset button on the card); see if the card can return to normal operation. (2) Cut off the power to the power supply card on the faulty side (note: it’s necessary to unplug the power plug; be careful not to plug in the wrong one). Wait for 20–30 seconds before reconnecting the power (plug the plug back in; sparks may occur, so act quickly, otherwise sparks will continue to appear), then check if the processor card can return to normal operation. 2. Replace the faulty processor card: (1) Cut off the power to the power supply card on the faulty side (unplug its power plug). (2) Turn off the battery switch of the faulty processor card, remove the card, and check the positions of all SW microswitches. Remove the battery as well. (3) Ensure that the model and specifications of the new processor card match those of the old one. Set all SW microswitches to the same settings as those on the faulty processor card. Reinstall the battery. (4) Insert the new processor card properly. (5) Restore power by plugging in the power plug; the READ and COPY lights on the new processor card should light up. After copying is complete, turn on the battery switch. These are online operations that will not affect the normal functioning of the system. The key is to be careful and not touch the side that controls the system’s operation. 3. If the processor card still malfunctions after replacement, it means that the power supply card on that side needs to be replaced. Additionally, refer to the IM hardware manual, which should contain instructions for replacing the processor card.

Generally, in systems with redundant CPUs, it is possible to replace one of them at any time: after a power outage, a new controller can be installed and then the system powered on again. Usually, the newly installed controller will automatically copy the programs from the currently operating controller. However, I haven’t used Yokogawa’s systems, so I’m not sure if that’s the case. It would be best for you to get in touch with them and have them replace the controller

5# jlshnlhj This of yours could even be used by Yokogawa as a guidebook

When the CPU freezes, if the other side is still functioning, there’s no need to rush and remove or replace the card. First, a utility software should be run; its location is at c:\cs3000\fcs\tool\.fcscdsv. Its function is to create a crashdump, that is, to save the register states of a faulty CPU in a file, after which you can send this generated file to the engineers at Yokogawa’s support department. (I believe your users should know their contact information.) If you don’t know, leave me a message. ! They can use this document to determine what exactly caused the CPU to fail and whether it can still be used. This message disappears after changing the card or restarting. This method is described on the forum

Thank you, jlshnlhj. The CPU stopped working for some reason, but it’s back to normal after restarting. I was terrified!

This post was last edited by whwglp on 2011-11-7 08:25. If it’s really broken； If it’s really impossible to shut down the machine ； If it’s really safe (although theoretically redundancy is no problem) ； If your level is good enough ； If an anomaly occurs, you should have a sufficiently good emergency response plan (of course, this refers to plans for the production process and equipment) ； If. . . . . . You can change it online; otherwise, you can just stop and deal with it

When changing the CPU, try not to download anything or switch CPUs until the COPY light goes out; otherwise, there should be no major issues.