Min Value Gate Operation

Hi experts,

I would like to know the explanation for below inquiries with regards to Min Value Gate (MVG) for GE.

1. What is it, a software based or hardware based comparator?

2. How does it recognize the low value from all the inputs?

3. When the input supposedly to be high, can it somehow misinterpreted bh MVG as low and hence giving the wrong FSR?

4. At certain sequence such as Start Up, does other inputs are forced to be HIGH?

>1. What is it, a software based or hardware based comparator?


>2. How does it recognize the low value from all the inputs?

It compares all of the inputs and selects the lowest of them all to be its output.

>3. When the input supposedly to be high, can it somehow
>misinterpreted bh MVG as low and hence giving the wrong

No. (I presume 'bh' was supposed to be 'by.'

>4. At certain sequence such as Start Up, does other inputs
>are forced to be HIGH?


It is very kindly suggested you somehow find a way to get access to Toolbox and the .m6b file for at least one of the Mark VI turbine control systems/units at your site. Things will become very clear after a steep learning curve, though if you're at all familiar with relay ladder logic it shouldn't be too difficult (since GE uses some relay-ladder graphics to illustrate how some Function Blocks work). You are sorely hampered without access to the application code, AND the Item- and Block Help files, in Toolbox.

You might post your email address here and ask if someone could send you an electronic print-out of their .m6b file (it will be several MB). And, since you won't have the benefit of using Toolbox to search and have Help available, it might be VERY difficult to learn too much more. BUT, at least you will have some idea of what things look like, if not the total picture. GE generally shows inputs to a function block or macro on the left, and outputs from the function block or macro on the right (flow is from left to right through the block). But, really, without Toolbox to switch to Item- or Block Help and to be able to see internals of blocks and macros in a graphical representation, you're not going to learn very much. You really need access to Toolbox and a .m6b file for a unit.

We can't post pictures are images to control.com to help you visually. And, really, the detail you are asking for is pretty excessive for a free World Wide Web forum. You need to get your employer to send you to some training and get some hands-on experience with the "tools" of Mark VI (namely, Toolbox). The kinds of questions you are asking require lots more detail than we can provide (easily), and some questions require lots of background and further detail for explanation--and that's really difficult without being able to point to specific blocks and functions in Toolbox and have you do some study and investigation yourself.

I know this sounds harsh--but you are to be commended for trying to learn without the materials and experience and training you need. You may have been successful at other sites and on other equipment, but most people will tell you: Teaching yourself Mark* turbine controls is difficult to impossible without a LOT of late-night work, and access to the tools provided for configuration and troubleshooting. In fact, I'd say it's pretty damn near impossible without access to Toolbox. (Normally, I tell people to study and learn and commit the system P&IDs to memory at the beginning of their journey to understand the Mark*--and you will also find it necessary at some point to do that, if you want to be a good technician (or even a good operator). People don't like P&IDs, but if you want to be a good technician (or operator) you are going to have to learn them, or be content with not being a good technician. The P&IDs show most of the inputs and outputs and field devices connected to the Mark*--and it's important to know what they do, and, at least schematically, where they are (though it's best to know physically exactly where they are!). Many things in Toolbox become clearer when you understand the P&IDs.)

Best of luck in your endeavour. You need to get access to a .m6b file and Toolbox. You can get some basic concepts from reading the blurbs and system descriptions in the Manuals (which is a lot more than many people do!), but you're NEVER going to get to the level you seem to want to be at without the "tools" necessary.


>Why do we select the lowest?

We don't select the lowest (of multiple inputs to be the output). The MIN VALUE, or MINIMUM SELECT, block selects the lowest of multiple inputs to be the output.

And that's a philosophical choice--a control philosophy. For example, the thing that decreases the hot gas path parts life the fastest in a heavy duty gas turbine is excessive hot gas temperature entering the first stage turbine nozzles. (It affects all downstream hot gas path parts, as well.) The thermal stresses of starting and shutting down are also very hard on the hot gas path parts, but all other things being equal when the turbine is running at rated speed and producing torque for the device it is driving (a generator or a pump or a compressor) it's the temperature of the hot combustion gases entering the first stage turbine nozzle that needs to be controlled "limited."

I believe a reason it was desired to use a "minimum" selection function was that in the event that some control reference incorrectly called for excessive or maximum fuel flow that it would NOT be selected as the output to the fuel flow control system. (This may have been the most important design criteria--again, based on the fact that it's best to keep the hot gas temperatures entering the turbine below a certain value--I'm not old enough to know that for sure (if it was the most important design consideration, but I'm sure it was very important.)

Now, there are several control "loops" that can control the fuel flow-rate into the combustors--in other words, multiple fuel flow-rate references depending on the operation of the machine at any given time. There is Start-up control, Acceleration Control, Speed Control, Exhaust Temperature control, to name a few. And, if logic switches were used to change between them when the fuel control needed to be changed it's very likely that there might be "bumps" or "dips" in the fuel flow-rate being switched to. The fuel being switched to might be higher (either slightly higher or much higher), or it might be lower (slightly lower or much lower). And, of course it is desirable for fuel flow-rate to be smooth and not have the "bumps" or "dips" at all.

So, it was decided to feed all of the various potential fuel flow-rate references into a MIN SEL block and let the block choose the lowest value at any given point in time to be the output--and the "final" reference to the fuel control valve. It was believed, and has proven to be true, that the "bumps" and "dips" experienced during the switching from one reference to another would be minimized or eliminated. And, the fuel flow-rate references were also designed so as not to be the lowest value when they weren't needed for control. In the case of Shutdown fuel control, when the unit is NOT shutting down, there is logic that is holding it at Maximum so that it will not be the lowest value, and in the case of Start-up control once the unit reaches approximately 100% speed there is logic that also ramps the Start-up fuel flow reference up and "out of the way" so that it will not be the lowest input to the block.

But, now let's say that the unit is on Droop Speed Control producing power and the generator breaker suddenly opens. The very first thing that's going to happen is the turbine-generator shaft speed is going to start increasing, and very quickly--towards overspeed. Well, Acceleration control, which is looking at the rate of acceleration, senses the sudden acceleration and moves to reduce the fuel flow-rate to slow down the rate of acceleration.(In later Mark* turbine control systems for GE-design heavy duty gasturbines when the Mark* senses the generator breaker opening the Speed Control fuel flow-rate control automatically moves to 100.3% which should also help to limit the acceleration--but just in case it's not doing a good enough job, Acceleration control is there to be an "upper limit" the fuel flow-rate even more until the acceleration rate is "under control" and Speed Control can then move back into the "lowest" input "position.")

But, for the most part, under normal rated speed operating conditions, Speed Control, Acceleration Control and Exhaust Temperature Control are always active inputs to the MIN SEL block.

Many of these kinds of "decisions" are based on operating and/or control philosophies that were established early in the design of GE heavy duty gas turbines and have just carried over to modern controls. I'll bet (because I haven't any experience with them) that early electronic GE Mark* turbine control systems used diode selection of multiple analog signals for choosing the lowest input to be the output of the "block" ("block of diodes"), and when it came time to design a similar software function, well that was used as the basis for the design--minimum selection).

Minimum Selection, or Median Selection, or High (Maximum) Selection are all control philosophies--and they all have their uses.

Usually, these kinds of decisions are based on years of operating experience and, usually, operational failures. And, again, back in the early days of electronics diode selection was about all they had for something like this--which lent itself very well to choosing the lowest value (back in those days one also had to be an electronic expert to design and even to work on . Over the decades GE has "tended" to make very few changes to their overall heavy duty gas turbine controls philosophies (GE Belfort is a MAJOR recent exception to this long-standing condition.) Actually, when the Mark II Speedtronic turbine control system was introduced GE went from relay-ladder logic diagrams to and-gate and or-gate type diagrams--and many experienced people (GE Field Engineers and operators/technicians) really ... disliked (to use a "nice" term) the new elementary drawing scheme. So, when the Mark IV came out (which was the next version of Mark* Speedtronic turbine controls) it used RLD format, which made many people who were familiar with older GE elementary drawings and people who know RLD very happy. So, GE tried changing something--and it wasn't well received. (And, in the beginning, the Mark IV wasn't well received either, but mostly because a lot of signal names (important ones) changed. But GE didn't relent on that.

And, the new control systems for the newest GE-design heavy duty gas turbines are using new and different signal-naming conventions, as well. And, they are being well-received. (KKS-based, I'm told--and you don't want to know what KKS stands for in the controls community!)

Again, it's mostly a philosophy (and maybe even an electronics-induced method back in the days of diodes and analog signals) that drives some of the choices and methods like this.

Hope this helps!