In order to understand the control system of GE-design heavy duty gas turbine, I'm starting to learn how to read Control Specification. Can someone please explain how RPM is converted to counts in the following control specification sheet? And what values Mark VI will use for calculation in various algorithms?
Moderator's note: scroll down to see table.
02.01.03 SPEED SIGNAL: HP TURBINE (PERCENT)
ALGORITHM: PR_MEAS TNH ELEM SH: 42M
THE TNH_PR SIGNAL IS RESCALED TO PERCENT OF RUNNING SPEED AT LINE FREQUENCY,
THE GOVERNOR SPEED, TNH.5728, B15
(ADJUST TNKH2 FOR 26214, B15 AT LINE FREQUENCY)
SPEED %SPEED COUNTS CONSTANT NAME
5106* 100 26214, B15 TNKH2
6381 125 32767, B15 TNKH3
*NOTE: TNKH2= ---------------------
RATED TURBINE SPEED (WHERE: SPEED = RPM)
CONSTANT K VALUE UNITS DESCRIPTION
TNKH2 K 11551 CNT 15
TNKH3 K -2 CNT 15
The section you provided seems to be from a very early Mark IV Control Specification. The Mark IV used fixed point math, and the documentation was very cryptic and could be confusing. Fixed point math requires the use of shifts to keep the binary maths and the decimal maths "together."
I found it best to ignore the "extra" binary scaling information and shifts. Since we can't "see" what's going on in the binary maths (which is what the Mark IV actually uses) unless you are trying to write new code/rungs which use maths it's best to just presume the Mark IV microprocessors are doing the binary maths correctly (which it always does).
Use your preferred Internet search engine to look for information on fixed point binary maths if you want to pursue the matter further. It would consume a small text book to try to explain it here, and I'm sure there are better (and worse) explanations on the World Wide Web.
The Mark IV uses fixed-point mathematics. Not floating-point point maths we are most accustomed.
Because the Mark IV was an early digital control system thr designers decided to describe the binary information for many control signal values, because troubleshooting without that information is very difficult.
The really great news is that the Mark IV and Mark V (which also used fixed-point maths) we're both very well designed and didn't (don't) need the level of troubleshooting they designers anticipated.
You can find all sorts of information on the World Wide Web about fixed-point maths. There is also a very good description of fixed-point maths in the Mark V Application Manual, GEH-6195, which can be downloaded from many sites on the World Wide Web.
Hope this helps.