Dear CSA,

Good wishes to you..

Pls help to diagnose the following alarm appearing on the screen of Mk6 control system while the turbine is running.

Message :- " NTP status- Insufficient system resources exists to complete the requested service". Along with this alarm message, all the screen turbine parameters goes blank. A # sign appears in all the parameter targets.A fter restarting CPU / rebooting, the entire values come back and screen gets normalised.

This message comes on its own, without anyone requesting any of the services from Mk6.

Thanking you in advance for the suggestions/solution

ESKAY

 

The FIRST question to be answered is: <b>When did this problem begin?</b> Has someone been trying to change the time on the HMI? Or has someone been building, downloading? Making changes to the configuration and/or the HMI? Is the master time signal working properly?

This being a Mark VI, the unit has probably been commissioned and in service for some time, so something has changed or been done recently.

NTP is Network Time Protocol, and this is the service that synchronizes the time of all the controllers and HMIs on the UDH.

One of the HMIs (or the HMI if there is only one) is designated to be the NTP Server and to broadcast the time on the UDH.

I have never seen this particular alarm before, and it may or may not be related to the NTP service.

But I'm pretty certain of one thing: Something has changed.

NTP is an open-source "application" and you might be able to find something on the World Wide Web (Internet) if the alarm is related to NTP.

Best of luck, and let us know how you fare.

 

I will also ask if, when the CIMPLICITY display "goes black" and shows octotrophs in the value fields, can you connect to the Mark VI using Toolbox? If so, what Diagnostic Alarms are being annunciated?

Are there any error/status LEDs lit on the Mark VI controllers when you re-boot?

Is this a SIMPLEX or a TMR installation? You say "...After restarting CPU ..." Which CPU? The HMI or one of the Mark VI processors (if it's a TMR panel)?

Is the Ethernet card for the UDH working correctly?

Are the VLAN switches working correctly?

 

it sounds like a software bug that prevented NTP to be running at that time. I would contact your vendor to get a software patch to resolve the software issue which caused the system running out of memory or resource.

 

Dear CSA,

May be the alarm related to NTP came on the screen,when an operator tried to correct the timing in HMI at the "clock" icon in status bar. There is 19 mts difference exists between actual time and the time shown in HMI.

Now the turbine is running on load and don't want to disturb any setting. We have one local and one remote HMI. CPU of remote HMI was rebooted to normalise the screen. No LED was lit in the controller. UDH and VLAN switches were working correctly. Hence, I treat this post is closed and I shall come up with new posts if the same alarm for NTP appears next time. Thanx..

 

ESKAY,

Many sites do not have a "master time" signal generator (such as a satellite receiver or an Internet-connected device/PC) which is providing the time signal to the NTP Server through a time synchronization card in the back of the HMI which is serving as the NTP Server, or over the PDH (or even over the UDH, sometimes).

You haven't told us where the NTP Server is getting it's time signal.

I've even sites with improperly configured HMIs and Mark VI control systems that have no time synch card and are not referencing the HMI's internal clock.

It's not even clear how long this problem has been persisting.

You will not trip the unit by checking or changing the time on the HMIs. GE fixed that bug a long time ago(!).

It's recommended you have someone from GE or the packager of the turbine/control system at your site to come out and review the configuration at your site and make changes or recommendations about how to correct the problem (including making any configuration changes). The whole time synchronization function can be very easily mis-configured and it will still "work" but not properly and not always.

 

If the CIMPLICITY screens are showing offline (#) for variables, then I doubt NTP issues are the root cause; most likely the NTP alarm is just a symptom as well. It sounds more like the system is out of memory or some other system resource, possibly from a memory or resource leak bug in some piece of software on the HMI. Like CSA said, I'd contact GE or the packager.

 

Dear CSA,

Thanks for ur advice and suggestion.

Let me go thro the manual to find out from where the the time signal is coming to NTP server.

I shall revert back to this problem once I get the full details on this subject with a new post.

Thanx...

 

That's why "unapproved" programs and applications are not supposed to be installed on GE HMIs. CIMPLICITY and TCI have enough memory- and resource leaks of their own.

Seriously, though, quite often when non-GE approved applications are installed on HMIs there can be problems. This is a good point.

And some versions of CIMPLICITY and TCI have been known to have memory- and resource leaks.

But mostly problems like this are caused by someone trying to change something for the wrong reason.

Since ESKAY considers this thread closed, I think we should drop it.

<b>[THUD!]</b>