Calibration screen for GCV and SRV

V

Thread Starter

voyager2_kbc

We have two MS5000 machines with Mark V HMI. When we go to the calibration screen for GCV, SRV or IGV, these screens do not work, I mean, when we try to select GCV calibration, SRV calibration or IGV calibration with a click over the target button, they do anything.. nothing happens. The HEX values that have to change when we select one the calibration screens (GCV, SRV, IGV) do not change and we can not do the valve stroke or the opening of the IGVs.

If we make the same things mentioned before in the other machine, everything goes right... we make the GCV stroke, we open and close the IGVs and we make the SRV stroke (Changing the regulator and making the download first)

Have you guys any idea of what is happening?

Regards,
Voyager2
 
Voyager,

It's not clear if you have an <I> or a GE Mark V HMI. The GE Mark V HMIs run MS-Windows and CIMPLICITY.

I'm going to presume they are <I>s, and going to suggest the displays were not configured properly >>OR<< the Password Protection is active on one <I> and not on the other.

It's also not clear if the two Mark Vs are of the same vintage. Newer Mark V panels require an additional logic signal to be a logic "1" to allow servo outputs to be stroked (L97HP0T_BYP--that sixth character in the signal name is a zero--NOT a capital 'O'). Sometimes the logic was automatically set to logic "1" in the CSP when appropriate; other times it had to be forced when the unit was at zero speed.

I also want to suggest using AutoCalibrate instead of the User Defined, or Demand, Displays. By doing so you will not have to change regulators; AutoCal does that automatically. Of course, the Password Protection must still be properly enabled.

Contrary to popular belief, AutoCal >>DOES<< work. It just has to be set up correctly (as any tool does). The topic of setting up AutoCal properly has been covered in many prior threads on control.com and can be found using the 'Search' feature of control.com.

Hope this helps! If you can provide more information, we can possibly be of more assistance.
 
V

voyager2_kbc

Both Mark V IDOS version where installed in 1994 with the installation of two new brand machines (Twins). In 2004 the Mark V IDOS version was changed with a Mark V Windows version. We do not have the AutoCal screen. Since 2004 the calibration screens, in both machines, were working properly but probably we made a change in a file or the system has a corrupted file or we changed a card and we did not configure it properly or something is missing and we do not know what is..One day we noticed that this was happening but we do not know when this started.... the fact is that we have four years without using these screens in this machine because they do not work.. we log in as administrator in both machines, we can force logics in both machines but the calibration screens do not work. They have CIMPLICITY

Voyager2
 
Voyager2_kbc,

Have a read of this thread:

http://www.control.com/thread/1403445367

You are right; something has been done to cause the problem you are experiencing. I'll keep thinking about what could be done to try to undo the thing. In the meantime, get AutoCalibrate up and running and see if it will work, and write back with the results.
 
Voyager2_kbc,

I am presuming you are talking about Demand (User-Defined) displays for stroking/calibrating SRV, GCV and IGV LVDTs. These aren't CIMPLICITY displays--they are TCI displays. That is, they are not pretty graphical displays; they are text-based displays with buttons and variable values on them. You have to go to the Start Menu, GT Control System Solutions and click on TCI or some other folder and select "Demand Displays" (or "User-Defined Displays") to open and use these displays, and there is a menu of various displays. They usually "float" in a window above the CIMPLICITY displays. Please confirm.

I'm still at a loss as to explain how these Demand (User-Defined) Displays cannot work or can become disabled. It's very strange. Very strange, indeed.

EVERY GE Mark V HMI has AutoCalibrate capability--even if it doesn't appear on some clickable screen. The file AUTOCAL.EXE should be in the G:\EXEC directory (or something similar, maybe ACALIB.EXE). And, every unit-specific directory (F:\UNIT1 for Unit 1, and F:\UNIT2 for Unit 2--if the operator interfaces are multi-unit operator interfaces) should each have a file ACALIB.DAT. AutoCalibrate uses the information in ACALIB.DAT for knowing what the ends of travel are for the SRV, GCV and IGVs when calibrating LVDT feedback. If your GE Mark V HMIs don't have ACALIB.DAT, let us know and we'll make one available for you.

If you find ACALIB.EXE or AUTOCAL.EXE in G:\EXEC, and you find ACALIB.DAT in the unit-specific directory(s), open a command prompt window on the GE Mark V HMI (click on Start | Run and type CMD in the box and press ENTER), change to the unit-specific directory, and type ACALIB or (AUTOCAL--depending on which file you found in G:\EXEC) and press ENTER, and AutoCalibrate should open. (NOTE: When you open AutoCalibrate, a few Diagnostic Alarms will be annunciated. This is because AutoCalibrate communicates directly with the RAM on the TCQA cards--and the designers of the Mark V originally wrote alarms to indicate any TCQA RAM access, and then they didn't disable those alarms when AutoCalibrate was developed. So, don't be surprised when a few "new" Diagnostic Alarms are annunciated during AutoCalibrate operation--it's "normal" (unfortunately!).)

Again, there was a wive's tale (a fairy tale; a falsehood; a myth) that AutoCalibrate didn't work properly. And that's just not true--it is an excellent tool for calibrating LVDT feedback, and for manually stroking the SRV, GCV or IGVs--without having to change regulator types and then changing them back! It saves a LOT of trouble.

Hope this helps!

I'm still at a loss as to explain why the Demand (User-Defined) Displays won't work. Is there a file in F: called DYNAMIC.BIN? If so, can you compare the sizes of the files in the two GE Mark V HMIs and let me know if they are different? What are the file date/time indications of the two files on the two HMIs? (DON'T try copying them between HMIs! Not just yet, anyway!!!)
 
C
Any changes to files in the PROM directory can cause all demand displays etc to stop functioning.

Restore the PROM directory from backup or a similar HMI.
 
curiousone makes an excellent point: The value of back-ups <b>CANNOT</b> be overestimated. For example, if one had a back-up of a known working GE Mark V HMI condition that could be used to return to a known working GE Mark V HMI condition.

I don't know as I go so far as to say that changes to <i>any</i> file in a PROM subdirectory can make Demand (User-Defined) Displays not work correctly. Certainly one should <b>NEVER</b> be making changes to any files in a PROM subdirectory <i>without assistance or direction from GE.</i> This is because the effects cannot be anticipated, and can be quite difficult to recover from--even with good back-ups. The files in a PROM subdirectory MUST match the contents of the PROMs used on the various cards in the Mark V turbine control panel. If changes are made to PROM subdirectory files and then compiled files are downloaded to the Mark V--or used to configure the operator interface RAM--the chances that something is going to be adversely affected are extremely high. And, once downloaded, it might be very difficult to do another download to undo the changes.

It might be worth a try to copy the contents of a PROM subdirectory from a working HMI to the HMI which is experiencing the problems, then perform a MK5MAKE and download USER to each of the Mark V processors and re-boot the HMI. <b>BUT, this should be done >>AFTER<< a complete back-up of the non-working HMI is performed to be used in the event something goes wrong with the attempted copy and MK5MAKE.</b>

Unless you use other Demand (User-Defined) Displays that have non-working targets, I would highly suggest switching to AutoCalibrate for calibrating LVDT feedback and for manually stroking servo-operated devices. It's much better at calibrating LVDT feedback--and it doesn't require the downloads for reconfiguring the SRV regulator that other SRV LVDT calibration methods require. So, it saves a lot of time and downloads and re-boots.
 
C
Do the HMIs have CD burners installed?

If so, you can perform some simple backups.
First shutdown the TCI with the command line: Net stop tci

Then backup your F: and G: directories to the CD drive using the xcopy command from the command line. For example:
xcopy F:*.* /s/e/v D:\back1\*.*
/s includes subdirectories
/e even empty ones
/v verify

Reboot when backup is complete.

Once you have a backup of each HMI you can compare the files to each other from each machine.

Once you have yours HMIs repaired, I will describe how to make backup hard drives for the HMIs
 
> This is because AutoCalibrate communicates directly with the RAM on the TCQA cards--and the designers
> of the Mark V originally wrote alarms to indicate any TCQA RAM access, and then they didn't
> disable those alarms when AutoCalibrate was developed.

Very useful discussion here, all true. But I have an unusual status reporting on one of our units when running the autocal on our Mark V. The Hex status value is coming back as "Status= 116" after an IGV cal, one of the TCQA status's that is not listed in the Aucalib.dat file. This is the first I have seen this error and not been able to clear it with a repeat calibration.

Is there another way of knowing the what the error is?
 
There is a potential SAFETY issue related to ACALIB vs. Demand Display

The ACALIB screens maintain the position of the device ONLY while the screen is OPEN.

This means that: if the ACALIB screen is closed (operator leaves the screen) , e.g. if somebody accidentally powers down the HMI or clicks away from the ACALIB screen: the device will go to the 0% position "very fast"!

The situation described is potentially dangerous if somebody is out there doing physical measurements (stroke, IGV position)

On the other hand, using the "demand display" puts the device in the position requested, and that position is not affected if the screen is closed, the HMI rebooted or disconnected: safer when somebody is making physical measurements.

So...
1. to perform LVDT feedback calibration without physical intervention: ACALIB is great

2. to perform physical measurements (IGV angle, actual stroke): ACALIB is not recommended, use the Demand Display instead.

Hope this helps
 
1) USE AUTOCAL.EXE on C, M drive on unit folder

2) using demand display you can do calibrate as well
 
Top