Electro Hydraulic Controller (Functional description) –
Introduction :Electro – hydraulic controller is an integral part of EAST (Electronic Automatic for steam turbine) supplied with KWU designed turbi nes. As the name suggest the controller combines the advantages available in electrical as well as in hydraulic system and provides following additional additional advantages over conventional governor. Increases the life of turbo set by conservative operation with the aid of TSE. Reliable operation of isolated power grid by automatic switch over of the load controller to frequency control. Precise maintenance of the rated frequency of the power grid by means of an exact frequency load curve. Low speed deviation under all operational conditions. Support of the pressure control system. Function wise electro – hydraulic controller (EHC) can be divided in to following sub section. 1) Speed measurement 2) Speed controller 3) Load controller 4) Pressure controller 5) Control Transfer 6) Position controller 1) Speed measurement :Speed measurement system should be commissioned before barrin g gear as it involves work front bearing pedestal. For speed measurement four numbers of hall probes are mounted around a disc containing 120 magnets (60 N – Pole and 60 S – Pole placed alternatively). Advantage of the same is that one rotation of disc will generate 60 puses. So by counting the pulses for one second we can directly detect the turbo set speed in rpm. All the four Hall probes are will upto pedestal junction Box. Out of these four Hall probes, three probes are used for measurement purpose and one is kept as spare. The gap between hall probe and disc should be maintained around 0.8mm. The three Hall probes which are used, are wired upto three pulse converter Junction Boxer. (In the previous set all the three pulse converter were mounted in one Junction Box but as these are mounted in an our tight enclosure in the field due to excessive heat, cards used to fail Due to this reason in new sets three separate Junction Boxes are pr ovided).
The Hall probes contains 4 wires. A constant current (30 – 80 mA) is fed to the two wires and across output wires a voltage is generated. The output voltage polarity and magnitude depends upon the magnetic field. The internal resistance of the Hall probe is around 33 r. When the magnetic disc rotates around the Hall probes, Hall probe output is a sine wave with a peak voltage of ± 700 mv to ± 900 mv. In the pulse converter card these – sine wave are converted into a square wave with a pulse voltage of ± 10.0 v. Each pulse converter card generates three isolated output out of which two outputs are
used and one is kept as spare. In EHC cabinet the speed measurement is done in two independent but identical channels. One channel is used for controller and another for indication and limit value monitors. Both measurement channels receive signal form all the three pulse converters. The three signal processed and one, out of the three signals, is selected for further processing with the left of a specially designed card ADA 11. Along with the selecting the healthy signal for further processing this card also g enerates selective fault alarm if any of the input signal is faulty without affecting the system. This fault can be acknowledged by switch 511 if signal becomes healthy and turbine speed is more than 13 rpm. If speed fall below 13 rpm the fault is memorised and it can not be acknowledged. Advantage of the above system being that at stand still when all the channel are giving no pulses the card will indicate the defective hall probe and the same can be attended. After processing the frequency signal is converted into voltage signal for internal use and then into a current signal for external use Speed Controller :Speed controller is used to increase the turbine speed form barring gear speed to rated speed in a controlled manner and to assist Auto – synchroniser in synchronising the M/c. After synchronisation speed controller can be used for taking up the load upto 100% but as a normal practice approx. 10% of load should be taken with speed controller and then load controller should be taken into services. It is do ne because during speed controller mode a small frequency change will cause corresponding change in valve opening (hence in load) as the valve opening is directly proportional to the difference between grid frequency and speed set point. Speed controller can be divided into following sub-section. a) Speed setpoint b) Speed setpoint control e) Dn/Dr monitoring d) Speed controller loop e) No load correction
Following is the brief description of various sub sections. a) Speed setpoint :Speed setpoint can be changed from cabinet, desk, Auto synchroniser and SGC turbine. The setpoint is called nr. The setpoint changes at a gradient of 2160 rpm/min upto a setting of 2820 rpm. After 2820 rpm speed setpoint changes at a gradient of 360 rpm/min. This enables a accurate setting of setpoint in the working range and enable smooth loading with speed controller after synchronising speed setpoint indicator is available at desk in to ranges (i) 0 – 3300 rpm (ii) 2700 – 3300 rpm speed setpoint remains in follow made during following conditions. i) After turbine trips speed controller follows the actual speed with the difference of 120 rpm. The trip command is initiated from pressure switch MAX51CP011 or MAX51CP012 (in old sets from P.S. MAX51CP013). This follows mode ensures that setpoint is below the actual speed when the trip circuit is normalises and rolling does not takes place till raise command is given to speed setpoint. After synchronisation when load controller takes over speed controller follows the actual speed between 49Hz & 51Hz. This ensures that speed controller does not interfere in the operation as the output of speed controller will remain at 0.0v. As per original scheme in this mode speed setpoint used to remain at 3015 rpm but as our grid frequency is not constant speed controller used to interfere in the operation. Speed setpoint control :The setpoint fixed in the earlier section cannot be passed to speed control loop directly as it is only the desired value which can be changed at a very fast rates and it does not takes care of turbine margins. To incorporate above facilities an integrator (Closed loop) is used with the help of the integration a maximum desired rate of acceleration can be set during commissioning (normally it is set at 600 rpm/min). The integrator has also got a provision to receive TSE signal and controls the rate of change
of output depending upon the TSE signal. If TSE margin is more than, 30°K then the output changes at a rate of 600 rpm/min of margin falls below 30°K the rate of change of speed is as per fig. 1.
The output of the integrator is known as NRTD or NRLIM. The NRTD indicator is available at desk. The range of the indicator is 0 – 3300 rpm. The integrator has got following extra provisions. i) Fast calibration is provided during both the follow mode to enable fast matching of input (NR) and output (NRTD) ii) Stop mode to block the integrator during disturbance in TSE or steam parameters. iii) Simulation of 0°K TSE margin (hence blocking the integrator output where it was) of output of integrator (NRTD) increases more than the actual speed by 45 rpm (originally 17 rpm) before generate breaker is closed. This ensures that before synchronisation speed controller will not give output more than 30%. (iv) Generates signal NR> NRTD and nR
10.00 ------- = 22.22 0.450 This gain should be adjusted by AE131 R26. The system response. Can be adjusted by Potentiometer R12 & R13 of AE 131 by using a high speed recorder (Oscillomink). The output of controller is known as hrnc. e) No load correction :As the speed controller loop is PD (P) so an error in the input is a must to keep the valves open for running the m/c at the rated speed. To avoid this error between setpoint and actual value a pressure depended correction signal is giver. As the valve opening required for running the m/c at rated speed is directly depended on steam pressure. Hence for correction purpose practical reading should be taken when the m/c is running at rated speed different pressure. For example :Turbine running at 3000 rpm Output of speed controller at (50kg/cu) = 1 .5 v Output of speed controller at (150kg/cu) = 0 .8v The pressure transmitter range is 0 – 250 = 0 – 10v The correction input is given with a gain of 0.05 and the overall gain of speed controller loop is 22.22 so net gain for correction signal = 0.05 x 22.22 = 1.11 1.5 hence correction input required at 50 kg/cu = ---- = 1.35 V 1.11 0.8 and correction input required at 150 kg/cm2 = ---- = 0.72 V 1.11 Now two equation can be made with Y = m x +c y correction signal X pressure signal 1.35 = m x 2 +c (1) 0.72 = m x 6 +c (2) By solving above two equation m= - 0.16 c = 1.67 Gain (m) should be adjusted by Potentiometer ABO99/RSS – ive polarity of (m) is taken case by feeding positive signal to negative point of comparator. Biasing (c) can be adjusted by potentiometer AB115/R13 (3) Load Controllers :Load controller is used to maintain the load set by the operator or Automatic load dispatch center (optional). Load controller is also responsible for taking care of frequency changes and correcting the load. The controller comes into operation only after synchronise before synchronisation if remains in the follow mode. The total loop can be divided in following sub section. a) Load setpoint (PR) b) Load gradient and TSE margin influence c) Load setpoint control (PRTD) d) Frequency influence e) Load Limiter f) Actual load Acquisition g) Final control loop a) Load setpoint :Load setpoint can be changed from cabinet, desk, or for Auto dispatch center (optional). The setpoint changes at a gradient of 100 MW/min. The gradient remains
same through out the range. The load setpoint indicator is available at desk. The range of the indicator is 0 – 250 MW. Load gradient and TSE influence :Load gradient is an additional feature available in load controller. The load gradient acts in parallel with TSE margin with a minimum selector. Load gradient range is 0 – 25 MW/min. While for TSE margin 0 – 30°K corresponds to 0 – 25 MW/min Both load gradient and TSE margin have got an On/Off Switch so if any one is desired to be off taken out of service can be switched off. The one which remains on, controls the output. If both are off the PRTD changes at fixed ratio of 25 MW/min. Load gradient setting can be changed from desk or cabinet. Load gradient setting is common for increasing as well as decreasing. In case of TSE upper margin controller the increasing rate and lower margin controller the decreasing rate. The limiters for the margins are provided in such a way that – ive upper margin is allowed to control PRTD (hence it can unload the m/c) while – ive lower margin is not allowed to control the m/c (hence it can not load the m/c). If TSE influence is on and TSE develops any fault immediately a block command is initiated so that wrong TSE margins are not allowed to control PRTD. To restore it the TSE influence should be switched off. Before again switching on the influence the TSE fault should be rectified and acknowledge. c) Load setpoint control (PRTD) :The function of this sub section is similar to the one already described in speed controller. The maximum rate by change here is 25 mw/m in both directions. The follow mode and fast calibrate signal are used for different purpose as explain below. i) If load gradient is off and any change in Pr is made, PRTD changes by 5% (of full range) immediately and remaining change takes place at the rate given by TSB. ii) The setpoint controller remains in follow mode when generator breaker is off or load controller off or load is less than station load. During this mode PRTD follows actual load. This follow mode is a must so that the PRTD is brought back to zero after breaker opens. After synchronisation PRTD tracks the actual load (which is taken with speed controller) so that more time is not wasted in bringing the PRTD from zero value to working value. iii) During above follow mode fast calibration signal is given so that PRTD follow Pact without any delay. iv) STOP command is generated if TSE fault appears or any disturbance takes place in steam parameter. v) Generator blocking command for TSE test when PRTD & PR are not matching vi) Indicator for PRTD is available in desk with a range of 0 – 250 MW. d) Frequnecy Influence :Frequency influence is used to correct the load setpoint. Depending upon the frequency deviation a load correction signal is generated. The frequency influence has got a droop of 5% (It can be changed from 2.5% to 8.0% in steps of 0.5% on line) hence a deviation of ± 0.5 hc will cause load change of ± 40 MW. The correction signal is limited at ±40MW. The circuit has got a filter so that normal hunting in frequency will not cause hunting in load. But if frequency drifts by even a small amount will cause load change. The frequency influence can be switched ON/OFF depending upon the requirement. Frequency influence remains ineffective below a frequency o f 45 Hz. Indicator for the correction signal is available at desk. The range of the same is ± 100 MW. The correction signal is known as (pro f). e) Load Limiter :Final setpoint (addition of PRTD of Prof) is fed to a minimum to selection with the load limiter. As the name suggests if the final setpoint becomes more the load limiter setpoint the limiter will control the load. The limiter also can be used for fast reduction of load in emergency cases. While doing to PR (interior PRTD) also should be reduced, so that during loading load incr as per PRTD and not as per load limiter. f) Actual Load Acquisition :The actual load is mea in cabinet CJJ08. Three load transducers are prov for thi s purpose. All the three load transducers get signal from PT and CT and give the output in mA. The output range for these transducer is 4 – 20 m1 which corresponds to 0 – 250
mw. These three signal are passed on to EHC cabinet for selection and fault indication. The middle value of the three is selected for further processing. If any value drifts from the other value by more then a certain fixed amount (5%) then a selective alarm is generated and indicated. The fault does not disturbs the working of the controller as the selection circuit discards the defective channel. The fault generated can be acknowledged by switch 511 available outside an selection card once. The fault is rectified. g) Final controller loop :The final loop is a PI controller. The loop can be divided into two major parts consisting of two cards ARL12 and ARL13 specially designed for this purpose. Responsibility of ARL12 card is to receive the final setpoint and the actual load. From these two signals an error signal is generated. The gain of the loop is set in this card. This card has got an inbuilt frequency influence. Whose responsibility is to prevent the over speeding of the turbine. Hence this frequency influence acts only in one direction i.e. if frequency is going up it gives a correction signal to close the valve. If frequency is going down it will not take any action. Unlike the frequency influence described earlier this influence can not be switched off. ARL12 card also detects the large error between setpoint and actual value (if it exceeds 10% and frequency is increased beyond 51 Hz) it changes the controller to isolated grid mode. In this mode the gain of the loop is increased to take fast correction action. The output of ARL12 card is fed to ARL13 cards. ARL13 card provides the integrator part of the loop. Apart from the integrator, ARL13 has also got a filter so that it does not takes any action if output of ARL12 is hunting at higher rate than the one set for filter. For this purpose the hunting time prior should be recorded and the same should be set by potentiometer R11/R12 of ARL13. The setting of R11/R12 should be same to avoid beat formation hence if the setting is to be changed it one line than if should by done in very small steps alternatively. ARL13 card also has provision for follow mode. Two type of follow modes are available. Following above (hvo) :This mode is in operation during following conditions. (i) Load controller off or (ii) Generator breaker off or (iii) Pact < STL and DP>5% During the above condition the load controller output tracks the speed controller output tracks the speed controller output all the time with a constant error of – 150mv. This follow mode ensures that during above conditions the load controller should not come into action. Following Low (hvu) :This follow mode is present once the breaker is closed (synchronised) and speed controller is still in action. During this mode also the load controller output tracks the speed controller output with error of – 150.0 mv. The only difference in this mode is that the load controller output difference in this mode is that the load controller output can go more than speed controller output (hence load controller can take over) but it can not go 150 mv below the speed controller output. The advantage being that due to this follow mode the load controller can be taken into service any time without any delay after synchronisation if desired. The output of load controller is limited to + 10V thru first minimum selection circuit (explained in control transfer). The load controller output is known as hRDC. 4) Pressure Controller :Responsibility of pressure control loop is to maintain the pressure within the working range at all loads pressure controller has got two makes of operation. a) Initial Pressure mode :A pressure deviation signal (difference of requi red pressure as per load and actual pressure) is fed to the controller. As long as the deviation is more than + 150mv (Actual > required) the pressure controller output remains at saturation. If the deviation fall between 0 - + 150mv the output of the pressure controller jumps from saturation to the output of final minimum selection output (hence it is made reading to take over). As
soon as the deviation becomes – ive the output of pressure controller starts decreasing and it takes over the control and it will reducer the said as. This mode should per sets be used during start up and initial loading of load the turbo set. b) Limit Pressure mode :The function of this mode is similar to the initial pressure mode. The only difference being that a biasing of 10 kg/ur is given in this loop. Hence all the action start after the pressure drops by more than 10 kg/cm2. This mode should be used once the load on turbo set is stablised. Output of pressure controller is known as hrPRC 5) Control Transfer :This loop receives the signal from speed controller (hrnc), load controller (hrPC) and pressure controller (hrPRC). These signal passes.
Thru a set of MAX/MIN selection and then the final value selected is passed on to position controller. The functioning of the system is as follows hrnc and hrpc pass thru a MAX value selector and the value selected is passed on to first MIN value selection. 10.5V is added in hrnc and this value is also fed to first M1 value selector (This is d one to prevent the overspes of the turboset. As soon as the turbine speed increase above the speed setpoint then speed controller wi ll give output. This output can not pass thru MAX value selector, hence it limits the output thru first MIN value selector. The value selected in first MIN value selected is passed on to the second MIN value selector along with the output of pressure controller output (hrPRC). The value selected here is passed on to positive controller. All the three selector (one MAX and two MIN have the provision to indicate that which value is selected from this information respective controller (which is in service) is indicated. The final output of control transfer loop is known as hr. 6) Position controller :Position controller is the final control element in EHC. It receives the signal from control transfer. It receives the feed back signal from Collins transmitters. For this purpose two Collins transmitters are provided in the plungers coil. The output of the two Collins passes thru a MIN value selector (logic MAX selector) as the range is –ve and the value selected is passed on to the controller. The controller compares the setpoint and the feed back and depending upon the error signal it bring the piston in the desired position. Some important points in this loop are as follows. (i) Plunger coil supplied in EHC is an integrator type. The balance point of the coil is – 1.0 V (should be set any where between – 0.8V to – 1.2V). Similar to an integrator if this voltage (-1.0V) is applied across the coil, the pi ston remains where it was. For moving the piston in open direction the voltage has to be increased so if the voltage is made – 0.9V (-0.9V is more than – 1.0V) piston will keep on moving towards opening direction till the voltage is again made – 1.0V. As soon as the voltage is made – 1.0V piston will remain where it was. Similarly if voltage is made – 1.1V ( - 1.1V is less than – 1.0V) the piston will keep on moving towards closing direction till the voltage is again brought back to – 1.0V. (ii) The Collins transmitters used for a very accurate feed back of the piston position. The output of the Collins varies from – 8.0v to + 8.0v which is converted into 0 – 10.0v – 0.0v corresponds to positive at which control valve just starts opening and – 10.0v C to the position at which last control valve is just open. A constant b iasing of + 200.0mv
is given for Collins transmitters so as to a void frequent change from one Collins to another. In case of wire to in any Collins transmitters, the respective card generates a fault alarm. If at any time the between the two Collins transmitters exceeds If a fault alarm is generated. iii) There is a provision to switch ON/OFF the voltage to plunger coil But switching ON or OFF can be only if the output of position controller is more + 5.0v. This ensures that at the time of OFF the piston is full open and there will not jump after switching OFF. Similarly it ensures that switching ON, the command for full open is the hence piston will not go down (hence valves close) once the coil is switched ON. (to facility working on the controller after of the coil supply). Problems Faced and modifications done :1) The hall probes fails frequently. This is due to mainly due to following reasons a) Fragile design of Hall probes b) Failure of MOP bearing (due to this magnetic disk touches the hall probes and damages it) 2) Speed load setpoint does not changes its position after giving the command. This happen mainly due to dust. This can be rectified by increasing the setting Potentiometer R22 and R23. 3) Dn/Dt monitoring operates when speed is increased for from 600 rpm to 3000 rpm. For this the limit value monitor ABO59/x25/R14 should be increased to 700 rpm (2.10v) as the original value is two close to normal soaking speed. 4) Speed controller interferes in the operation if frequency changes even by a very small amount. As explained in speed controller, to over come this problem a follow mode has been included so that speed setpoint follows actual speed between 49Hz to 51Hz. 5) Speed controller gets a trip command in the initial rolling even though the turbine has not tripped. This happens due to momentary different in the trip oil pressure. To avoid this a timer of 2 sec. Is introduced in the follows command. 6) The pressure controller output is taken via a MAX value selector with a fixed value of 0.8 (10% load). This is done to ensure that pressure controller does not reduces the load below 10%. 7) Blocking command for load controller output during change over from load controller to speed controller is removed. This is done so that load controller is allowed to take action of any disturbance takes place duri ng change over time. 8) Collins transmitter card (AKK-11) generates spurious fault alarm. For this card AKC-11 has to be modified 9) In some projects wiring mistake in the feedback of Collins transmitters was noticed. For rectifying it, measure voltage at X14 with respect to X24 in the respective AKC-11 Card. Voltage should be +4.0v to + when values are fully close. In full open condition it should be between – 4.0v to – 8.0v. If it is reverse wiring should be changed in Junction box. 10) In some cases mistakes in plunger coil wiring was also found. To detect this feed + 1.5V at pin CB53L : Z2 with respect to CB53L : Z6, the piston should open fully. If not change the wiring in junction box. 11) Setpoint block command persist and load/speed setpoint cannot be changed. It is not due to any problem/mistakes. This command can be generated due to di sturbance in steam parameters or TSE, If it is from steam parameter it can be acknowledged by master setpoint release. If it is from TSE, influence should be switched OFF and master set points release should be acknowledged. Before taking TSE in services again, TSE fault should be rectified. The line for 0.0 or 4.0 mA selection was not in correct position. The CT ratio selection was also found to be an 10 Amp while it should have been at 5.0 Amp. After above two rectifications load controller started giving correct output.
