Installing remote monitoring devices to GE MkIV/V/VI turbo-generators for frequency testing

Installing remote monitoring devices to GE MkIV/V/VI turbo-generators for frequency testing

The electricity generation market in Australia is changing in 2022 and suppliers are required to prove the performance of their generators to the AEMO (Australian Electricity Market Operators).

Here at site, we have to install a device in the form of a high speed data recorder which will capture key generator parameters such as AVR, excitation system voltage and current, Governor response, active and reactive power, power factor, generator terminal voltage and current, droop response and potentially more to the electricity regulator.

  • We have selected the DigSilent Grid Code Monitoring device PFM300-32 for the remote monitoring trial.

We are required to extract the required data from many devices including the Speedtronic (MkIV/V/VI). As we have very few Speedtronic subject matter experts (SME’s) on hand, we are getting remote support from an overseas SME.

We have been asked by the SME to supply the following:

The SME requires the UNITn (where n is the unit number) directory complete with all the subdirectories (especially the PROM) without the UNITn and the PROM directories we cannot prepare the software modifications remotely.

The SME requires the m6b file.

We have provided the SME the Elementary Drawings but he requires the MLI A010 (Control System Specification) drawings also needed for each case.

Can my fellow engineers at assist in helping me how to extract the information the SME requires and the format we can provide him as we have limited experience on-site? This is appears difficult in equipment way before HDD & USB!!

Note: I am looking at basic MkIV/V/VI training for technicians and engineers at site but organising this and thereafter gaining experience in these legacy controllers for younger engineers shall take time. General assistance in these old Speedtronics is a challenge.

Many thanks for the continued support. KWS (Perth, Western Australia)
Ken Symmons,

Er, ..., Um, ..., Why can't the (over-paid) SME tell you how to get the information they need.?.?.? It would seem if they have the information they gave you, for the tidy sum you are paying them they should be able to tell you how to obtain it. But, I digress. Again.

We would have to know what type of operator interface(s) the Mark V('s) have, and whether or not the floppy disk drive(s) or the USB ports or the CD burners even work. If they are GE Mark V HMIs, with Ethernet NICs (Network Interface Cards) installed for connection to an Ethernet network, are they connected to an Ethernet network and can you shar files over the network to another device which can accept the files/directories? But, there will be a drive letter, F:, on the HDD of whatever operator interface the Mark V('s) have and you will find the UNITn (where 'n' is the unit number, usually beginning with 1, sometimes only 1) and its PROM subdirectory. Then you will have to find a way to copy the contents of those directories onto some kind of removable media to send to the (over-paid) SME.

The Mark VI should be the easiest of the bunch, because it more than likely has an Ethernet NIC installed in it even if the USB port(s) don't work, or the CD burners don't work. You would need to use MS-Windows Explorer to search each of the hard drives to find a folder most likely called GTn (where 'n' is the unit number), and then search in the folder (and any subfolders) for a file with a .m6b filename extension. If there is a drive called 'MASTER' you particularly want to look in the folders on THAT drive for the .m6b file. (The .m6b file is the file that contains the majority of the configuration information and programming (application code) that is downloaded to the Mark VI and used to control and protect the turbine and auxiliaries.) Sometimes that file can be several MB in size; sometimes it's only about 1 or 2 MB in size.

The Mark IV is going to be the most difficult. The ML A010, Control Specification drawing may, depending on the age of the unit, actually be three documents (paper documents). I'm sure the one the (over-paid) SME wants is the Control Specification-System Settings document. In the last versions of the Mark IV, there was just one Control Specification document which contained the information in the System Settings document AND the System Adjustments document. You will most likely find the Control Specification document(s) in one of the Operations and Maintenance Manuals. There are usually four (4) volumes; the third being the maintenance section of the Manuals, and the fourth volume being the Parts Lists. You should concentrate your efforts on Vol's. I and -II, and if you can't find the document(s) in one of those Volumes, then look in Vol. III. (GE is NOTHING if not consistently inconsistent. There was usually a tab in Vol. I or Vol. II (sometimes Vol. III) called 'Control Specification.')

If you cannot find the Control Specification document(s), your only option at that point is to hope the line printer in the Mark IV turbine control panel works, has ribbons, and paper. (You can use fan-fold paper as long as it has the perforated tractor feed edges as most all of the line printers GE provided also had tractor feed mechanisms. Whether or not you use rolled paper OR fan-fold tractor feed paper, you need to make sure that it is PROPERLY lined up going into the printer AND coming out of the printer, or it will snag, and go askew, and you will have a mess on your hands (if it's even still working).)

Then unplug the printer from the mains power and let it sit for about 5 (five) minutes with no mains power. This will clear the tiny bit of volatile memory the printers had, so that if it's memory is full and it has been trying to print but couldn't the memory will be clear.

With a cleared memory and properly aligned printer paper, and mains power then you can got to the Main Menu (I think it was called) on the <OPM> (Operator Interface Module) on the left front door of the Mark IV turbine control panel and go to the 'Control Constants' display, and beginning on the first page, press the PRINT PAGE softkey . Once the printing has started (or finished), scroll to the next page and press PRINT PAGE and continue until all of the Control Constants have been printed.

At this point, the last and the "easiest" thing to do is to make a photocopy or a scanned .pdf of the Mark IV Speedtronic Elementary to send to the SME. The Mark IV Speedtronic elementary is the document which depicts hardware and software used to control and protect the turbine and auxiliaries, can have a couple hundred pages, and is usually 11"x17" in size (sometimes people with really good eyes would make 8-1/2"x11" (A4) size copies of the drawing--but they are pretty useless because one can't write notes on them very well--and that's the best thing about printed copies of the sequencing (being able to make notes!)). There are ovals with Control Constant names on the software sheets of the elementary that can be used to look up the value of the Control Constant being supplied/passed to the software.

And that should give the (over-paid) SME what they want. You could scan the Control Constant printout and send that electronically if you scan the Mark IV Speedtronic Elementary.

Myself, I'm not curious enough to look up the equipment you mentioned to see how it gets data--but everything you asked for is NOT always available in an analog, discrete or even proprietary communications protocol format on every Mark* turbine control system. You would be better off getting copies of the Generator Control Panel elementary drawing, or the Generator Protection Panel elementary, and finding where to get the information from the PT and CT circuits needed to derive most of the signals you listed.

Droop response, Governor response (which, in my mind are basically the same) and "potentially more" are going to be the most difficult "signals" to find. Droop response can be most easily measured by monitoring two signals in the Mark* turbine control: TNR (turbine speed reference, in percent of rated speed) and DW (it might be DWATT) (the typical signal name(s) for generator load, which come from a transducer monitoring PT and CT signals in either the Generator Control Panel (for Mark IV systems), or the Generator Protection Panel (for Mark V and Mark VI systems)). You want to change TNR (by using the RAISE- or LOWER SPEED/LOAD switch on the Generator Control Panel (for Mark IV systems) or clicking on the RAISE- or LOWER SPEED/LOAD targets (on the HMIs of the Mark V and Mark VI systems) and observing how much the load changes. For example, if you increase TNR by 0.5% (say from 101.5% to 102.0%, or from 102.1% to 102.6%, as examples) the load for a machine with 4% droop should change by 12.5% of rated (on an isobaric day, and on a turbine in a new and clean condition with clean turbine inlet air filters and a negligible exhaust duct back pressure). Or said a different way, every 1.0% change in TNR should result in a 25% change in load (of rated load on an iso day, for a new and clean ....). This is also true when decreasing TNR by some amount.

You will ALMOST NEVER find the droop percentage (regulation) number specifically defined in GE documents. People like to "tweak" the Droop setting for reasons they THINK will change the performance of the machine, and in the process they cause a slew of knock-on effects--which are ALWAYS the turbine control system's fault (NOT!!!). So, GE doesn't make it obvious and a lot of the problems just never happen. There are ways to "back-calculate" the Droop setting from various Control Constant values, but what you are looking for, anyway, is ACTUAL Droop percentage (regulation), and you have to prove what it is. And if it isn't what it should be, then you will likely have to change it. And the (over-paid) SME should be able to help with that. The one good thing is, though: All the Mark* turbine control systems you listed are programmable digital control systems--and parameters (Control Constants) DO NOT drift over time. Either someone mucked with it, or was changed during commissioning. But, the only way to really know what the actual Droop percentage/regulation is is to test the system. And, if it ain't close to what it should be--then start worrying about what to do.

Fuhgettabout training. One of the beauties of the Mark* systems is that it rarely needs service, and when a company spends a LOT of money to get employees trained the people who attend the training rarely use the information the learned--and when they need it, it's a couple of years after the training, or more, and they're lost and confused. Or they got promoted, or quit; but mostly, they just won't remember what they learned because they don't use it very often. It's a waste of money, and if you get the training from GE or BH--it's a colossal waste of money. If the people you want to spend money to train up on Mark* turbine control systems rarely have to work on the systems and they have other equipment and control systems they are responsible for, especially don't do it. Don't spend the money for Mark training. Find a local resource on Oz or in Asia (I REALLY intensely dislike recommending that, because there are so few who really are competent on Mark* systems in Asia, even working for the OEM) and use them. Sure, it will cost you money--but when you call them they have knowledge and experience they regularly use and you won't be calling them often (because the Mark* is genuinely a reliable and robust control system (yes, even the Mark V is reliable and robust)).

This is the (un-paid) SME signing off! That's all I got. Best of luck!

Don't let the dingoes bite you!
Ken Symmons,

I didn't answer the questions about format.... For Mark V, when you copy the directories and their contents to removable media they will include a LOT of ASCII text files--which anyone can open and view with any basic text editor (think MS-Notepad, or even MS-Word (just don't make any changes with MS-Word or save the changes that might unintentionally be made). That covers the Mark V.

The .m6b file is in a proprietary format, which begs the question: How is the SME going to be able to view the file, because they will need the GE program Toolbox or (Legacy Toolbox) or ToolboxST (depending on what version of software is in use on the GE Mark VI HMIs at your site(s)). Either they have to have some version of Toolbox (or ToolboxST) or they have some magic I'm not aware of. So, this is interesting that they asked you to send ONLY the .m6b file (.m6b stands for Mark VI binary file--because of its format). I also think you are going to need to tell the (over-paid) SME what version of Toolbox (or TooboxST) is in use for each Mark VI being covered in this study. (Cause they are most likely going to need to be able to know which version(s) of Toolbox (or ToolboxST) to use to open and view the information in the .m6b file(s).

As for the Mark IV, the ONLY way you are going to find the ML A010, Control Specification, is in printed form (UNLESS someone at your site scanned the printed version into electronic format, or paid someone/a company to scan the printed version into electronic format). It DOES NOT exist electronically in the Mark IV on any chip anywhere. AND, the Control Specification is a printed document which likely was never issued as an as-built after the unit(s) were installed and commissioned--so they only list the FACTORY values which were in the Mark IV when it was shipped to site. Many times, the actual values of Control Constants get changed during commissioning, or after. And those changes never get made in the printed version of the Control Specification in the Operations & Maintenance Manuals, or sent to GE to use to issue an as-built printed document.

SOOO, this means that if you ONLY give them the Control Specification document it probably won't accurately reflect the ACTUAL Control Constant Values in the Mark IV.... So, I would definitely recommend you provide BOTH the Control Specification document (if you can find it), AND the printed list of Control Constants from the Mark IV. That way the (over-paid) SME has BOTH the actual Control Constant values AND the factory, as-shipped Control Constant values to use for their analysis.

Another thing is: EVERY Mark* can have different operating parameters (Control Constants). So, you are going to need to provide this information (for every Mark*) for every turbine being covered by this study. While the Control Constants may have the same names in different Mark* turbine control panels, they don't have to have the same value.... So, you should be providing the same information for each and every Mark*/turbine-generator which will be covered in this study. (Hopefully, the (over-paid) SME told you this, and you just didn't tell us....)

Hope this helps! This sounds worse than it will actually be; you, and your personnel, just need to be patient and methodical and you will be able to gather all of the required information (at least the information you listed above which the (over-paid) SME asked for). It's not all in one place; it will need to be unit-specific (meaning you will need to provide the information for each Mark*/turbine-generator being covered in this study); and it will found in different formats, as described above. But, at least you will be able to gather and send the information the (over-paid) SME asked for (per your original post).

Patience. Perseverance. You will get through it. Finding out how to get the information from older GE Mark V HMIs (via removable media) is also going to be difficult. A lot of floppy disk drives and CD burners have just quit, or because of lack of use and lack of maintenance on the operator interface CPU (keeping it cleaned--regularly) they just won't work. And, the Mark IV printer--which SO many people just abhor and (incorrectly) call "old technology" and have ignored for decades (letting the ribbons dry out and not buying replacement ribbons; not keeping the paper in the machine; just generally ignoring the printer (because it continually prints large volumes of alarms, and not knowing how to stop that (at least temporarily) or resolving dithering alarm conditions)) is going to present a problem. You could take digital pictures of each Control Constant Display screen--but you are going to have to make sure the pictures are CLEAR and the information is readable. And, that, of course, presumes the Mark IV CRT works and is readable.... There are ways to digitally "download" Mark IV information (NOT the Control Specification, though), but that would just take too much time and back-and-forth in this forum to be able to get your personnel to a point to be able to do that.

Best of luck! There are dingoes in this task; just be patient and persevere and you can keep them at bay.

Please write back to let us know how you fare!