Automotice Changeover to Startup Fuel Initiated

S

Thread Starter

Shahab Qader

Our gas turbine 9E starts up on distillate fuel and "change over to low calorific gas" is available at 37.5 MW GT load. We are initiating changeover on 50MW. Now the problem is that Gas turbine successfully changeover to gas but immediately gas pressure P2 low alarm appears and Automatic Changeover to startup fuel is initiated. We have noted that it is happening due to Cavity backflow protection which says that if P2 pressure is not 5psig more than CPD within 20 seconds of L3GRV becomes 1, auxgas is locked out. We have noted that P2 has reached this threshold in 273 seconds instead 20s stipulated time. During last successful changeover it took 19.52s. I have noted that frcrout is increasing very slowly from initial value of -40%(SD_ref) (when L3GRV became 1) to -8%(when pressure reached its threshold).Valve opening was from -0.07% to 0.2%. My vendor says it should reach to 0% within 1 sec or less. He is analysing this anomaly. Anybody has some idea. We have done the stroking of SRV and it is closely following the position reference.

Shahab
shahab.qader [at] gmail.com
 
When did this problem start?

What is the gain of the servo regulator for the SRV?

What is the gas fuel supply pressure doing during the transition?

How stable is the gas fuel supply presssure during load changes and during the transfer back to liquid fuel?

How long is the fuel transfer timer set for?

During a fuel transfer it's customary for there to be a slight time delay on opening the fuel valves of the fuel being transferred to, or to open them slowly at first, in order to allow some time for the fuel lines to fill and pressurize. If the valves open too quickly initially in relation to the other fuel valves there can be too much fuel initially, and if they open too slowly in relation to the other fuel valves then the load can drop because there is insufficient fuel flow as the lines pressurize and fill. It's not clear if the delay you are speaking of is related to fill times or not, but I would say the SRV shouldn't go from -40% to being open extremely quickly. This could cause the gas fuel supply pressure to drop and/or could cause excess fuel to flow into the turbine and cause a load increase and possibly even an exhaust over temperature trip.

What is the load doing during the transfer from liquid fuel to gas fuel, and then from gas fuel back to liquid fuel?

On the face of it, it would seem that the gas fuel supply pressure is not sufficient to support the flow required at 50 MW. If the SRV is full open trying to make P2 equal to P2 pressure reference, then the gas fuel supply pressure is most likely insufficient to support the flow required at 50 MW.

If you enable the transfer at the minimum load setpoint, how does the unit behave?

The SRV should be pretty fast-acting to respond to changes in supply pressure in order to maintain the P2 pressure reference. This presumes there is sufficient gas fuel flow available, meaning that as the flow increases the supply pressure (upstream of the SRV) should remain fairly stable or at least 0.5-1.0 barg above the P2 pressure reference. The SRV needs to be able to increase P2 pressure quickly if the pressure drops for some reason (usually because the GCV opens) and it also needs to be able to decrease P2 pressure quickly if P2 pressure increases for some reason (like the GCV closes quickly or the supply pressure drops suddenly). The SRV action should be fairly quick.

Gas fuel supply pressure also needs to be pretty stable as flow changes and the SRV is opening during the initial transfer. If the gas fuel supply pressure regulator isn't capable of maintaining pressure as flow increases someone may have tried adjusting the SRV regulator gain to improve the stability of the gas fuel supply pressure.

Sometimes, people <b>improperly</b> adjust the SRV regulator gain in order to try to help stabilize supply pressure when the gas fuel supply pressure/flow is unstable or can't be increased/decreased as required for normal operation. Instead of fixing the root cause of the problem (insufficient flow and/or supply pressure), they try to slow down the SRV which ends up causing other, unintended problems.

The SRV regulator is a pressure-control loop with position feedback. The regulator will put the valve at whatever position is required in order to try to make actual P2 pressure equal to the P2 pressure reference, provided the gas fuel supply pressure and flow is sufficient and the regulator gain is set properly. The accuracy of LVDT calibration is unimportant with the SRV because the regulator is trying to control pressure and not position, and will put the valve at whatever position is required to make the actual P2 pressure equal to the P2 pressure reference, provided the gas fuel supply pressure and flow is sufficient and the regulator gain is adjusted properly. (This also presumes the servo is the correct servo for the application (GPM and slew rate) and hydraulic pressure is sufficient.

One other thing which might be "choking" the SRV is that hydraulic pressure has to flow through 20FG-1 to get to the SRV servo and actuator. If the ports of the solenoid are plugged or restricted, this might cause some response problems, also.

Hope this helps!
 
Top