Hello Again,
Our GT-2 is SpeedTronic MK II Controlled... In Each shutdown we calibrate the cards of the TCP.. (Turbine Control Panel)... We use a Calibration Manual that enlists step-by-step calibration procedure... But there are certain procedures whose physical significance is not properly understood... I seek some help in understanding one such section... Following is the exact writeup as given in GE's Spec Manual...

---------------------------------------------
The Manual Quotes,
"
Section 13.7 : Excessive Start up Fuel Trip

During this calibration, Gas Fuel will be permitted to flow thru the turbine. To prevent Ignition the spark plugs must be disabled.

13.7.1 : Connect a Digital Voltmeter to the SRV's Servo Drive Voltage Pin.

13.7.2 : Temporarily disable the VMPA (Trip Card) Function.

13.7.3 : Place the Operation Selector Switch in "FIRE" and give the unit a normal start signal. When the SRV and GCV open and stabilize at their Firing Fuel Positions, Record the Voltage at "SRV's Servo Drive Voltage Pin".

13.7.4 : While Still in the Firing Sequence, Establish VCE = 10V. (In MK II this can be done by shorting two pins of a Card). This will cause GCV to go full open and SRV to open further (Pl. Note that in MK II SRV Opens fully at 9.25V). When the SRV position stabilizes, record the Voltage at "SRV's Servo Drive Voltage Pin". As soon as the reading is taken shut of the Gas Fuel flow by turning the Operation Selector Switch back to "CRANK" position.

13.7.5 : Calculate the "Excess Fuel Trip Voltage" as follows :

Excessive Fuel Trip Setpoint = [ Fire Stroke Volts, Step 13.7.3 ] + 0.5*[ (Full Stroke Volts, Step 13.7.4) - (Fire Stroke Volts, Step 13.7.3) ]

13.7.6 : Now Set VCE to Zero using Manual VCE POT. Allow this VCE to control the position of SRV by jumpering J1 and J2 on SRV Drive Card. (In MK II, by jumpering these points VCE is directly fed to SRV).. Remove GCV Drive Card to ensure that GCV remains shut.

13.7.7 : Slowly increase VCE untill the voltage at "SRV's Servo Drive Voltage Pin" equals the Excess Fuel Trip Voltage calculated in Step 13.7.5

13.7.8 : Observe the "Fault" light on the SRV Drive Card and adjust The Trip Reference Pot so that the "Fault" light just comes on.

13.7.9 : Give the Unit a stop signal.

13.7.10 : With Spark plugs removed, all cards in place and all jumpers removed, give the turbine a normal start signal with master operation switch in the "Fire" position. Confirm that the unit does not trip with the GCV in its normal firing position.

13.7.11 : While still within the Firing Time, turn the Manual VCE POT untill VCE is reduced slightly from it's firing value. Install a jumper to make VCE Maximum. (In MK II this can be done by shorting two pins of a Card).

13.7.12 : Increase VCE with the Manual VCE POT and confirm that the unit trips when the voltage at "SRV's Servo Drive Voltage Pin" reaches the value calculated in step 13.7.5.

13.7.13 : Remove all jumpers and test equipment. Reset Max VCE as per control specs.

13.7.14 : Purge the turbine by cranking the unit for 2 mins.

13.7.15 : Reconnect spark plugs.

---------------------------------------------

First of all, what's the exact significance of this calibration procedure... Is such a thing taken care of in MK V system too??

au revoir
Rahul

 

Funny you should ask... This alarm caused this author some grief several times this past Monday on a Mk IV-equipped unit.

The primary purpose of this calibration/alarm is to prevent a condition where the GCV was "wide open" during firing and the SRV would open excessively to maintain the P2 pressure. This would result in "excessive fuel flow" into the turbine and exhaust duct, and many bad things could happen, as you could imagine, if the fuel ignited....

Many such alarms, trips, and configurations resulted from an unanticipated catastrophic failure somewhere in the world--this one most likely on an unmanned site (many units were/are remotely started while someone is sent to site to observe/operate the unit while it is running). No doubt, this can be a very valuable safeguard if the on-site operator is not paying proper attention, but it can also be costly to crank the unit (electricity costs money--even in a power plant), and flow gas through the unit into the open air.

In the Mk II days, after disabling the spark plugs one actually opened the GCV and admitted fuel (LOTS of it) into the unit through the SRV taking measurements and making calculations to determine what position the SRV might go to under such conditions and adjusting the alarm setpoint accordingly. LOTS of work; LOTS of natural gas going unburnt up the stack; lots of energy spent CRANKing the unit.

However, because the Mk II is basically an analog control system--with potentiometers and resistors and LOTS of termi-point and wire-wrap connections--it's important to simulate as realistically as possible many of the conditions and calculate and adjust the pots as necessary, and even to periodically check the accuracy of the calculations and pot settings.

In the Mk IV, Mk V and Mk VI world, there is a similar alarm condition (which results in a turbine trip--emergency shutdown) if the SRV is open "too far." The "too far" amount is a calculated value, calcluated at "the factory", and is sometimes set too low, depending on site conditions and must be adjusted if/as necessary. The alarm text message reads something like "Start-up Fuel Flow Excessive-Trip", but when one looks at the elementary/CSP one can see that fuel flow-rate feedback is NOT monitored, simply SRV position.... But, it's not so critical to perform a similar check/calculation--and, without lots of jumpers and batteries and I/O Configuraton changes, it's virtually impossible on a Mk V-equipped unit!

The present-day alarm text is simply a carry-over from the early SpeedTronic days....

With digital control systems, this isn't quite so critical any longer. You've already noticed that there aren't nearly as many calibration procedures in the Mk V as there were for the Mk II.... That's the result of many things; some good things, and some bad things....

The Control Specification documents began to really go downhill during the Mark IV production days, until they've reached a critically low bottom point now with the Mk VI. What was once an excellent source of information from three drawings (Control System Settings, Control System Adjustments, and Operating Instructions and Sequences), is down to a single document which is full of inaccuracies and missing critical bits and pieces of information.

You could probably evaluate how you operate your units and decide, based on how much adjustment you make each time you perform this procedure, whether or not its worthwhile to perform on a regular basis. Certainly, if you replace any of the cards in that circuit you should perform that procedure (as you should have any time you replaced any of the cards in that circuit since component tolerances vary).

markvguy

 

Thanks for a - as always - nice elaborate reply...

Do you think, if this section (Excessive startup fuel flow) is imporperly calibrated, it can lead to a following condition?

Condition:
Recently GT-2, was to be taken up for annual preventive maintenance... When the STOP command was initiated to the Turbine, it stalled at 4000 RPM... The turbine stopped decelerating and maintained 4000RPM... 14HS didnt drop out... Variety of options from depressing 5E (Emergency) Push button to manually closing the inlet SDV (Shutdown Valve) were discussed... But finally one attempt at a second stop was taken and it coasted down to zero speed.... We later stroked the GCV/SRV but it was found to be ok... The turbine, at the time of stop command, was running on "Lean gas fuel" as it always does... We checked with the process people to confirm if there was an increase in the clarofic value of fuel... Even that wasnt the case...

Any clues from this info??

au revoir
Rahul

 

No clue whatsoever--except that 14HS NORMALLY drops out at about 94% of rated turbine speed, and 4000 RPM would be MUCH LESS than 94% of rated speed.

And, we all know that rated speed at your site is a matter of long unresolved dispute!

markvguy

 

Yes, there is an ambiguity with the MK V machine's FSNL speed value... But MK II machines were always running at 5100 RPM at FSNL... However, in coming months we'll get a chance to deal with the MK V machine also during it's scheduled first HGPI (Hot Gas Path Inspection) since commissioning... The Mech. Maint. team has sought help from the OES... We shall, too, have a chance of checking some of the parameters as discussed in the earlier posting related to my query on Freq and speed mismatches, on this website...

Usually when the machine is running, the process personnel dont even permit to take the machine in FSNL... They either want to export or consume... So any elaborate testing is always fraught with risk for an individual who tries to "expand his own learing" at the cost of generation...!!!! But that's how most maintenance jobs go in our country where an individual gets a chance to learn only when the machine "Trips"... And then it's expected of him to be "all knowing" so that the machine is started in as small time as possible...!!

But this forum has indeed helped cos it encourages one to ask even the most obvious and fundamental questions... Thanks to the postings of mkvguy and others... Hope this association blossoms...

au revoir
Rahul

 

Rahul, it's virtually the same everywhere in the world. Fuel costs money, and the purpose of running the unit is to generate beans (er, uh, electricity) to make the bean-counters (er, uh, accountants and owners and investors) happy and pay for that fuel. It's always difficult to convince Operations/Management that just a couple of minutes to gather data is beneficial in the long run--and gathering the information asked for previously would only take about two minutes if you had your meters in place before the unit was started.

In fact, in order to synchronize the unit, the speed/frequency will HAVE TO BE matched to the grid frequency regardless of the configuration settings so you could just monitor the whole process and record the speed pick-up frequency and the generator frequency during synchronization using two digital frequency meters (Fluke 79s or 8060s or equivalent).

It is understood that it is difficult to attach meter leads to the Phoenix high-density terminal boards used in the Mark V turbine control panels, but many turbine control panels (possibly yours??) supplied by GE licensees with GE-design heavy-duty gas turbines used "marshalling" panels as intermediate terminations to allow better troubleshooting access. The larger, more traditional, terminal boards used in the Marshalling Panels make it easier to connect meter leads to monitor values.

With regards to the rated speed of Frame 5s and Frame 6s, as has been explained before, the axial compressors of these machines were designed to be operated at _APPROXIMATELY_ 5129 RPM. Load (Reduction) gear manufacturers could examine the metal "slug" being used to create the bull- and pinion gears and decrease or increase the number of teeth (increase the tooth size or decrease the tooth size, respectively) in order to maximize the strength of the gears. Some Load Gears operated at 4994 RPM, some at 5100 RPM, some at 5105 RPM, some at 5129 RPM, some at 5130 RPM, some at 5134 RPM, some at 5135 RPM, some at 5143 RPM to produce rated generator rotor speed/frequency (50 Hz or 60 HZ). (Those are only the speeds which this author has encountered; they are likely others...)

As with start-up sequencing, it's not a good idea to generalize too much.

markvguy