from the Control and Instrument department...
Interpreting FSR's BBL
 Posted by Pranatia Wisnu Heryuda on 16 February, 2010 - 9:23 pm
Dear CSA,

I'm heryuda, a newbie in speedtronic mark V frame 9 control system. i have a lot of difficulties in interpreting the BBL of FSR. i need to know the interpretation of FSRV3, the fuel stroke reference, L60FSRV2, the FSR rate of Change, FSRMINV2, the minimum fuel stroke reference, FSRACCV1, the acceleration control of FSR, FSRMANV2, the manual fuel stroke reference, FSRSUV1, the startup of FSR, FSRSDV1, the shutdown of FSR, and the last is FQROUT V2, the liquid flow command.
i hope you can help me.

my best regards.

 Posted by CSA on 17 February, 2010 - 3:04 pm
Learning a new language is always difficult, and this is nothing more than a different way of expressing paths, and comparators, and mathematical operations (addition, subtraction, multiplication, division, square root, etc.), and some logical operators which prevent or permit signals to pass through the blocks.

It's a very graphical language, and in general the inputs to the blocks are shown on the left side of the block and the outputs from the block are shown on the right side of the block (so signals "flow" from right to left, generally).

It just takes some time and patience and study, and you will find that it's not that difficult. At first it seems daunting, but then everything new seems (somewhat) difficult at first, doesn't it?

In the Appendix in the back of the Mark V Application Manual, GEH-6195, there is a section on Mark V fixed-point mathematics (which can really be the most confusing aspect of the Mark V, all those nasty "shifts" and such!) and blocks (primitives (ADD, SUB, MULT, etc.) and "Big Blocks" (algorithms)). While not every block is explained, and the ones that are are often very cryptically explained, you should be able to study similar blocks and develop an understanding that will serve you will in working through those that are not described there.

Every time I have tried to document what happens in one of these blocks for someone, it always degrades into some kind or argument. Mostly because once people have a "glimmer" of what happens in the block and then have a more studied look at it, they develop their own "understanding" and interpretation and insist theirs is the proper explanation. (In fairness, I'm not always right.) So, I'm not going to even attempt to provide any detailed explanations of these blocks.

However, if you have specific questions after you review the blocks, I and others here on control.com will be happy to confirm or clarify your understanding of the blocks.

One of the things that people who will take the time to study the CSP and blocks always remark on later is that there is good detail in the blocks. Many, many control system vendors represent these functions with empty rectangles, and use written descriptions which can be very difficult to understand. So, the graphical representations used in Speedtronic panels actually provide some very useful information once people become familiar with them, and those who have experience with other similar control systems usually comment that the graphical representations used in Speedtronic control systems are very, very informative.

But, you have listed a very large group of blocks. I suggest you pick one and study it. Tell us how you think it works, and we will help you confirm or clarify your thinking. And, then you will see how that understanding can be used to work through other blocks.

As I said earlier, one of the very confusing things about the Mark V is that it uses fixed-point mathematics, which makes the use of "shifts" necessary to scale values so they can be used in subsequent blocks. In many blocks you will see notation such as "-2" or "1", and this is referring to a shift of the binary "decimal" place (how's that for a term!) of the value. Again, this is all explained in one of the Appendix chapters in the back of GEH-6195. Don't get hung up on the shifts and floating point math; it will just confuse you. Just know that if your turbine is running properly (and it is or has been running properly at some time!) that all the necessary shifts and manipulations were done properly and that it's virtually impossible for them to "drift" or "change" and cause subsequent problems.

Also, the Mark V was provided with a control Specification document, and that document has some very limited, but sometimes helpful, descriptions of some of the blocks you listed. Find that document and get familiar with it.

Yes, there's no question about it: The learning curve is very steep to begin with. But I assure you it really only takes a few hours (concentrated, focused hours) to get the basics down and then you will be surprised at how quickly other blocks can be understood.

Make photostatic copies of the blocks. I even recommend that you hand-draw the blocks one "signal" at a time. Make notes on your copy of the block about how you think it works. Obtain a printed copy of the CSP and make it yours by putting your notes in the document. Write all over it! It's yours, and those notes will help you many times over as you learn and progress in your job.

And, if you find something useful or helpful, or even if you find it to be completely wrong, tell us. "Feedback is our most important contribution" (c) here at control.com. It's what makes the posts here useful to others in the future if they'll use the 'Search' feature hidden at the top of the webpage. It also helps those responding to know if they've been helpful, or missed completely, and refine their answers now and in the future.

Welcome to the community!

 Posted by CSA on 17 February, 2010 - 7:11 pm
> in general the inputs to the blocks are shown on the left side of the block and the outputs from the block are shown on the right side of the block (so signals "flow" from right to left, generally). <

er, ... uh, ... If the inputs are on the left and the outputs are on the right, then the signals would "flow" from left to right (not right to left). Apologies for any confusion that might have created.

When the turbine is running, FSRSU and FSRSD are ramped out of the way (to 100%, usually) so that they don't interfere with speed- and temperature control). They are two of several inputs to the FSR "minimum select" function. See the FSR display on the Mark V operator interface for some help. Also, usually in the front of the Control Specification there is a drawing of how most of the blocks you listed all work together (depending on the vintage of the Mark V). So, it's really worth a look at that document.

The fuel splitter is the block that ramps one fuel up and the other is ramped down when you are trying to transfer fuels (such as from gas to liquid, or liquid to gas).

It also helps if you are having a particular problem if you tell us what the problem is.

 Posted by Heryuda on 18 February, 2010 - 3:08 am

1. in the BBL of FSRV3 (Fuel Stroke Reference)there is a minimum select and State logic block. Why a minimum selection is used for the control of a gas turbine?

2. In the same BBL (FSRV3), value of FSRSD is 100.00, and the FSRMAX,FSRNIP,FSRCPD,FSRCTD, and load_lim is 32767. What is theese values mean?

3.What is the relationship of minimum select and logic block at the FSRV3 with the FSRMINV2? FSRMIN V2 is the Minimum fuel stroke reference.

Thank you very much for your help.
My best regards.

 Posted by CSA on 18 February, 2010 - 1:20 pm
1. The Speedtronic continually calculates many values of FSR simultaneously and then the Min Select function chooses the lowest of these values as the Fuel Stroke Reference which will control the fuel flow-rate.

There is an ASCII text file on the operator interface in F:\UNIT1 called LONGNAME.DAT. This file contains the descriptions (limited to 80 characters, I believe) for most (not all) of the CDB (Control Signal Database) signals. It may be useful for you to refer to that for descriptions. The CSP Documentor uses the information in this file to populate the descriptions of the signal names in the CSP.

`FSRSU   FSR Starup (used to control the fuel during start-up and acceleration after firing and warm-up)FSRN    FSR Droop Speed ControlFSRNI   FSR Isochronous Speed ControlFSRACC  FSR Acceleration Control (used to "catch" the turbine shaft should it start to accelerate excessively)FSRT    FSR Temperature ControlFSRSD   FSR Shutdown (used to control the fuel during a normal, fired shutdown)FSRMAN  FSR Manual Control (used for testing purposes)FSRMIN  FSR Minimum (used to try to prevent flame-out when fuel is being reduced because of overspeed or load rejection, etc.); this signal's usage has changed over the years`

If you look at the FSR Display on the operator interface when the unit is running, you should see that FSRSU and FSRSD and FSRMAN are all at 100%, which means they are very high. If the unit is operating at Part Load (not on Base Load, exhaust temperature control) then FSRN should be the lowest of all the FSRs. FSRACC is usually just a couple of tenths of a percent greater than FSRN in this case, again, just to kind of "catch" the machine should it accelerate too quickly (not likely). FSRT should be greater than FSRN at Part Load.

When the unit is operating at Base Load on exhaust temperature control, then FSRN will be greater than FSRT, and FSRT will be the lowest of all the FSRS and will be controlling the fuel.

The concept of using the lowest of the FSRs as the controlling reference for fuel control is to protect the turbine. If any FSR is calling for "too much" fuel, it could damage the turbine. So, they designers decided to use the Min Select function as a kind of safety function, choosing the least of the various FSRs as the one that will control the unit.

If you look at the FSRSU and FSRSD blocks, you will see that they are set to "max" during normal operation of the turbine, effectively disabling them during normal operation (because start-up is complete and the unit is not in shutdown).By doing this, these two signals are "removed" from the Min Select function (because they are much higher than any of the others), thereby hopefully preventing any unintended interference with normal operation.

FSRMIN's primary function is to set a lower limit on FSR so that if any of the FSR calculations tries to reduce the fuel below the limit that it will not let the fuel go below FSRMIN. The purpose of this is try to prevent flame-out by reducing the fuel too quickly. For example, if the unit were at Base Load and suddenly the generator breaker were opened (but the event opening the breaker was not a turbine trip), then the fuel has to be cut back very quickly because the turbine will start to accelerate very quickly due to the high fuel flow. If the fuel were to be cut back to a very low value, the flame in the combustors could be put out to low or no fuel flow or even that the fuel flow was so low that the air flowing through the unit would blow the flame out. So, FSRMIN is calculated to try to set a lower limit to try to prevent flame out. (In this example, the turbine speed will still increase, slightly, because of inertia and the time required to reduce the fuel flow and "catch" the acceleration and reduce speed back to 100%, but it shouldn't reach the overspeed setpoint.)

Recent Speedtronic panels (starting with later versions of the Mark V) actually calculate a value of FSRMIN while the unit is running and feed that value to the FSRSU block and the FSRSD block. This is to try to help reduce thermal stresses on the turbine during start-up and shutdown that might result from excessive fuel flows. It has actually served to greatly increase the complexity of the control schemes, but it does do it's desired function. Way too much detail for this forum to accurately try to describe exactly how it works for every usage, and all the versions of it's increases usage.

2. I think the above should help with the answers to these questions. 32767 is one of the "maximum" values. One thing everyone should know about Speedtronic panels is that they are built to be used for lots of different turbines and applications. As such, a lot of signals are put into the "generic" database, but not all of the signals are actually used on every job. So, there will be some signals which will appear in the CSP and on displays which aren't actually used on a particular application. It makes understanding the CSP difficult at first, but, it's one of those things that you should be making notes on your copy of the CSP about and then it will become a more useful and meaningful document.

3. I think some of the answer to 1. should help with this, too.

 Posted by heryuda on 21 February, 2010 - 4:45 am
Dear CSA,

I hope this will be my last post in this topic, my last questions is:
1. what is the formula to calculate the L60FSRH, on the BBL of L60FSRV2, the FSR rate of change.

2. what is the formula to calculate the serv_ref, on the BBL of FQROUTV2, the liquid fuel flow command.

on next 40 hours will be my skill test, i really need your help.
Thank You Very much..

my best regards.

 Posted by CSA on 22 February, 2010 - 1:06 pm
heryuda,

In all the years I've worked on GE Speedtronic turbine control systems, I have never encountered a problem with the actuation of L60FSRH or L60FSRL. So, I've never really looked at the BBL. I don't have access to all my Mark V files at this time, but if either of these signals are being actuated (going to a logic "1") I would suspect that some Control Constants have been changed that shouldn't have been.

So, unfortunately, I can't provide much information on this at this time. Except to say, I would find it very unusual for either of these signals to be actuated under any circumstances without some ill-advised changes being made to some Control Constant values.

As for the liq fuel servo-valve output, I believe that the formula is:

(FSR1 * (TNH/100)) + FQKOUTNO + 0 = FQROUT

The above presumes the unit is running (L4 = "1"), liquid fuel is enabled (L20FLX = "1") and flowing above the minimum limit (FQKOUTMN).

It would be very helpful to know what problem or problems you are trying to solve.

 Posted by heryuda on 18 February, 2010 - 7:06 am
Dear CSA,

after i read again the manual,now i have more specific questions:

1. In the Big Block Language of FSRV3 (Fuel Stroke Reference), there is a Minimum Select and state logic. Why did the selection of fuel stroke reference is the most minimum value of the inputs?

2.when i click the "show value" button, there are values for:
FSRSD: 100.00

FSRACL, FSRNIP, FSRCPD, FSRCTD, LOAD_LIM :32767.

what is the meaning of these values? i saw this value during startup, but why did the value of FSRSD is the lowest?

best regards.

 Posted by Heryuda on 17 February, 2010 - 7:52 pm
Dear CSA,

I really thank you very much for your tips and help. i've read the gas turbine operator's manual and tried to interpret the BBL. as you said earlier that i've to pick one or two to be focused,but some specific block did not included on the appendix then i chose the startup FSR (FSRSUV1) and the shutdown FSR (FSRSDV1). In the case of interpreting the math operation, now i've no problem because of your tips, but i still need to know how this BBLs work to control the startup and shutdown of a gas turbine.

Thank you very much for your time.

My Best Regards.

Your use of this site is subject to the terms and conditions set forth under Legal Notices and the Privacy Policy. Please read those terms and conditions carefully. Subject to the rights expressly reserved to others under Legal Notices, the content of this site and the compilation thereof is © 1999-2013 Nerds in Control, LLC. All rights reserved.

Users of this site are benefiting from open source technologies, including PHP, MySQL and Apache. Be happy.

Fortune
"Amnesia used to be my favorite word, but then I forgot it."
Help keep our servers running...