Perhaps no one understands your job-related challenges better than colleagues with the same engine. That’s what makes the Frame 6 Users Group’s compressor, turbine, combustion, and I&C roundtables particularly valuable. These sessions allow you to describe issues of concern to fellow users and let them suggest possible solutions based on their experiences. Think of it as free consulting provided by the industry’s top O&M experts.

While no safety roundtable is scheduled for the 2021 conference, be aware that there are several safety threads posted to the organization’s online forum, hosted on the Power Users website—including experience with optical flame detectors, how to deal with ill-fitting compartment doors and hardware replacements to correct, functional tests to confirm proper operation of water-mist fire-suppression systems during unit commissioning, opening of compartment doors with the CO2 system activated, etc.

Another way to come up to speed on the safety aspects of Frame 6 O&M is to become familiar with the OEM’s safety-related Technical Information Letters (TILs) and Product Service Safety Bulletins (PSSBs). These are identified in the sidebar. If you don’t have copies of the pertinent documents, request them from your plant’s GE representative. And since you can’t remember everything, consider having one or more safety professionals assigned to your plant during outages.

Safety TILs and Product Service Safety Bulletins affecting 6B gas turbines

TIL 2101, Modification of manual lever hoist for safe rotor removal.
2044, Dry flame sensor false flame indication while turbine is offline.
2028, Control settings for GE Reuter Stokes flame sensors.
2025, GE Reuter Stokes FTD325 dry flame sensors, false flame indication.
1986, Braid-lined flexible metal-hose failures.
1918, 6B Riverhawk load-coupling hardware and tooling safety concern.
1838, Environmentally induced catalytic-bead gas-leak sensor degradation.
1793, Arsenic and heavy-metal material handling guidelines.
1713, 6B, 6FA, 6FA+E, and 9E false-start drain system recommendations.
1709, 6B load-coupling recommendations.
1707, Outer-crossfire-tube packing-ring upgrade.
1700, Potential gas-leak hazard during offline water washes.
1633, Load-coupling pressure during disassembly.
1628, E- and B-class gas-turbine shell inspection.
1612, Temperature degradation of turbine-compartment light fixtures.
1585-R1, Proper use and care of flexible metal hoses.
1577, Precautions for air-inlet filter-house ladder hatches.
1576-R1, Gas-turbine rotor inspections.
1574, 6B standard combustion fuel-nozzle body cracking.
1573, Fire-protection-system wiring verification.
1566-R2, Hazardous-gas detection system recommendations.
1565, Safety precautions to follow while working on VGVs.
1557, Temperature-regulation valves containing methylene chloride.
1556, Security measures against logic forcing.
1554, Signage requirements for enclosures protected by CO2 fire protection.
1537-1, High gas flow at startup—Lratiohy logic sequence.
1522-R1, Fire-protection-system upgrades for select gas turbines.
1520-1, High hydrogen purge recommendations.
1429-R1, Accessory and fuel-gas-module compression-fitting oil leaks.
1368-2, Recommended fire-prevention measures for air-inlet filter houses.
1275-1R2, Excessive fuel flow at startup.
1159-2, Precautions for working in or near the turbine compartment or fuel
handling system of an operating gas turbine.

PSSBs

2018-1003, Online collector-maintenance awareness.
2018-0709-R2, Observation of hexavalent chromium on parts during outage.
2016-1220, GT upgrade—Impact on HRSG.
2016-1209, Gas-turbine water-cooled flame sensor false flame indication.
2016-1117, Lifting and rigging devices.
2016-1104, Gas-turbine operational safety GEK update.

Some of the GE material pertinent to Frame 6 owner/operators goes beyond the basic engine. Example: PSSB20161220, “GT Upgrade Impact on HRSG,” presents the experience of an owner that learned an engine upgrade had been implemented without sufficient evaluation of the safety impacts on the boiler. Specifically, the new steaming capacity was greater than the nameplate rating and the relieving capacity of the existing safety valve.

This is a serious concern. But don’t expect to get a meaningful HRSG discussion going at a meeting focused on gas turbines. For that, you should participate in the HRSG Forum with Bob Anderson. Join the discussion at the first HRSG Forum in 2021, on May 3, by registering for the two-hour virtual event at no cost.

Finally, remember that there’s a fast amount of safety-related information readily available to owner/operators on the CCJ website, where you can find best practices submitted by colleagues over the years.

Below are a few of the discussion topics pursued, and thoughts shared, at recent meetings of the Frame 6 Users Group. Some you may have missed and are of current value, others might trigger some ideas to discuss at the upcoming roundtable sessions May 4 and May 18 from 10 a.m. Eastern to about noon. Access the complete agenda and registration form (no cost to users) on the Power Users website.

Proper electrical and I&C wiring inside the compartment important to unit reliability. When troubleshooting failing or failed sensors, technicians sometimes find that the temperature limit of their wiring is less than the compartment temperature. Poor-quality conduit should be avoided, too. One contributor to this discussion said that at his plant sensors are wired to relays to identify failed sensors.

Avoid water washing your compressor before an outage to minimize the possibility of corrosion. However, do water wash after an outage.

Clear the bellmouth drain after a compressor wash. You don’t want a couple of feet of water accumulating at the compressor inlet where it can be sucked into the unit on restart.

Relocate compressor bleed valves from inside the package to the outside for better reliability.

Check exhaust thermocouples during startup for possible problems ahead. If you a T/C lagging the others by about 100 deg F, and eventually catching up, consider replacement at your next opportunity.

Failure to restart after a unit trip. Check for sulfur buildup in stop/speed ratio valves.

Trip on low lube-oil pressure. A root-cause analysis revealed that regulator valves had not been serviced in more than three decades. Diaphragms became brittle and failed. Recommendation: Add diaphragms to your PM checklist if not already there.

Fire protection is a perennial topic. A user noted that the CO2 system at his plant discharged before the alarm activated. Having reliable alarms and external lighting to warn of a release is critical to personnel safety. One got the impression from the discussion that controls for fire-suppression systems—water mist and CO2—may not be as reliable as they should be. It can be difficult to find qualified vendors to maintain these safety systems, according to a few participants. One said he double-checks third-party certifications and any work done on the system.

Attendees were urged to check package integrity for leaks because if leakage persists—at louvers, for example—you can’t maintain the inert atmosphere while the unit cools. Louver mechanisms on legacy units were identified as a problem area and characterized by one user as being “rinky-dink.”

Difficulty in synchronizing a black-start unit revealed the following to investigators: The Mark VI auto-synch feature was not turned off and the breaker closed with electricians in the generator auxiliary cabinet—a safety no-no. The outcome from this incident was a modified startup procedure that requires operators to confirm excitation at 50% speed on black-start units. Electricians also must check the GAC to confirm there are no faults prior to startup. Finally, a warning sign was hung on the cabinet door and operators are required to issue stop-work notifications to electricians during engine starts.

Unit trip on high oil temperature without alarm notification. Recorded data did not indicate any change in oil temperature. The alarm for high oil pressure was found faulty. The gremlin was a loose wire. Termination strip was repaired and the unit returned to service the same day. User sharing the experience said termination strips can take just so much abuse and suggested that the person you assign to work on them should be someone you trust with a screwdriver.

Locate safety boxes at strategic locations around the plant to retain PPE-use requirements for specific tasks and equipment. Also, consider locating specific tooling at use locations. One example given was the placement of toolboxes on top of the HRSGs to reduce the need for technicians to travel back and forth to a central location, saving time and reducing the risk of injury.

BASF-Geismar’s operations staff was challenged to develop a tool that provides a simple and intuitive display of the performance of the Utilities Dept’s systems and equipment. Critical to the development of a user-friendly performance dashboard are the following:

• • Identify the proper key performance indicators (KPI) to monitor.
  • Model equipment/system performance accurately to provide appropriate target values under varying operating conditions.

The KPIs selected for monitoring included gas-turbine output and heat rate, boiler efficiency, steam venting, steam letdown through PRVs, specific power consumption for compressed air, and purchased steam.

Some KPI target values were a constant value—such as zero for steam venting and 70,000 lb/hr as the target for purchased steam. However, many KPI targets vary with operating conditions.  For instance, the expected efficiency of a boiler is not a constant value but varies based on boiler load, changes in fuel composition, etc. Similarly, gas-turbine output varies considerably with ambient-air temperature.

Equipment and system performance models were developed for a wide range of operating conditions. At BASF-Geismar Utilities, most of the modeling was based on real-life operating data collected when the equipment was known to be operating well. Thus, the operational targets are proven performance metrics and not necessarily based on new equipment design data, which in some cases may not be appropriate.

Because of the varying targets for different operating conditions, performance indices were developed for many of the KPIs. A performance index is a calculation to gauge how well a piece of equipment, or process, is meeting its defined expectation—or more simply, its target performance. A performance index of 1.0 indicates the equipment/process is exactly meeting its goal; a higher score, exceeding expectations; a lower score, not meeting expectations.

For processes measured by a value that increases with improved performance, the performance index is the actual performance divided by the target performance. To illustrate, if at a given condition boiler efficiency is expected to be 82.0% but the actual performance is 82.4%, that performance index would be 82.4÷82.0 = 1.005.  The result is greater than 1.0, indicating satisfactory performance.

For processes measured by a value that decreases with improved performance, the performance index is the target value divided by the actual performance. An example is gas-turbine heat rate, a measurement of fuel consumption divided by the unit output. If the expected heat rate of a gas turbine is 12.0 million Btu/MW and the measured (actual) performance is 12.25 million Btu/MW, the performance index would be 12.0÷12.25 = 0.9796. The result, being less than 1.0, indicates poor performance.

The performance index is not useful for comparing the performance of two unlike pieces of equipment. For instance, if equipment A, which normally produces 80 units per day instead produces 85 units is compared to equipment B which normally produces 100 units per day but instead produces 95 units, the performance index score for equipment A would be higher yet equipment B still produced more units. But if one does not look at the performance index value, one might think equipment B is doing well because it is out-producing equipment A while in fact it is underperforming its expectations.

The dashboard created (figure) shows KPI data together with a green, yellow, or red indicator light—to provide an instant indication of performance. An Excel spreadsheet was used to download the necessary process data from AspenTech Explorer and perform the necessary calculations.

The performance data displayed shows average values for periods of one, four, 12, and 24 hours along with the current performance. Providing data in this format allows performance trending.

Note that the small button with the “T” is a quick link that opens a trend graph for that particular parameter. Another quick link at the bottom of the dashboard opens a troubleshooting file which can be used as a guide to correct poor performance.

To create some competitive spirit among operators and shifts, there’s a “score” in the top right-hand corner showing the number of green lights compared to the maximum possible number of green lights. Current performance data are not included in this score as it changes on a minute-to-minute basis.

For the dashboard shown, the first three rows of performance data are KPIs monitored using performance indices with target values that vary with operating conditions; the bottom row of data are KPIs that have fixed target values.

From this display you can see boiler No. 4’s efficiency performance was unsatisfactory but improved. Similarly, steam was vented hours ago but the vent is now closed. Condensate return rates dropped and are still marginally low and should be investigated/addressed.

Data displayed in this manner does not tell, for instance, which boiler is operating most efficiently, but rather indicates how the actual boiler efficiency compares to the expected performance.

Boiler combustion controls are designed to optimize the air flow to the fuel flow rate such that sufficient air (oxygen) is available for complete combustion and the amount of performance-robbing excess air is minimized. Recall that insufficient air results in the formation of excess carbon monoxide, a regulated parameter, as well as a loss in efficiency because of incomplete combustion.

Part of the air-flow controls programming at BASF-Geismar allowed for operator adjustment of the stack O2 set-point value—an O2 bias value. An operator could enter a negative bias in an attempt to lower stack O2, thereby increasing boiler efficiency. But if the O2 value were reduced too much, insufficient air would be available for combustion, thereby producing an excessive amount of CO. By contrast, if the O2 bias value was set too high, an excessive amount of air would be used, leading to inefficient combustion.

Powerplant board operators have many duties and do not have the time to babysit the O2/air controllers to fine-tune the O2 bias value and to make adjustments each time boiler load or fuel composition changes. To avoid nuisance CO alarms, operators typically would set the O2 bias to a high value, contributing to inefficient operation.

Initial solutions revolved around trying to give operators target efficiencies to hit, thus letting them know at about what point the CO would “break-through.” This was only moderately effective because the target efficiency varied with boiler load and fuel composition, and occasionally CO break-through would occur before the target efficiency could be reached—for one or more of several minor reasons.

Plant personnel ended up modeling boiler efficiency over a wide range of operating conditions and fuel compositions and created characterization tables that could calculate an accurate efficiency target for the given operating conditions. This model was used to generate a set point for an efficiency controller, which when in automatic, would compare the actual boiler efficiency to the efficiency target (set point) and adjust the O2 bias automatically.

The efficiency controller automatically lowers the O2 bias value until (1) the target efficiency is reached, (2) the CO level starts rising (at which time the efficiency set point is adjusted lower, thereby increasing the O2 bias value), or (3) the minimum O2 bias value limit is reached. In essence, the new efficiency controller automatically adjusts the O2 bias value to achieve the target boiler efficiency, and the CO controller is configured to adjust the efficiency set point on the efficiency controller should CO emissions rise too high.

Control schemes, before and after the staff effort, are illustrated in the diagram.

Results: The efficiency controller and CO override controller have worked very well. Operators no longer have to adjust the O2 bias value as it now is generated automatically. The boilers operate at the targeted values and any CO excursions are handled automatically without operator intervention.

BASF Geismar – BASF Chemical Co
160-MW, gas-fired, 2 x 1 combined cycle cogeneration facility location in Geismar, La.
Plant Manager: Jerry Lebold

The sharing of best practices among owner/operators contributes to safer working conditions and to increases in unit availability and reliability fleet-wide. The Frame 6 users have been proactive in this regard, contributing their experiences during the annual meetings as well as in the group’s online forum, now hosted on the Power Users website.

J C Rawls, a technology engineer in BASF-Geismar’s utilities department and a member of the Frame 6 User Group’s steering committee, has been particularly helpful in explaining the details of his work on boiler and powerplant performance improvement to colleagues at the annual meetings and with the industry at large through CCJ’s Best Practices Program.

Three entries submitted by Rawls recently have been recognized with Best Practices Awards and may be of value to you. One discusses a home-grown boiler efficiency controller that improves performance through process automation, another describes a performance dashboard that tells at a glance if a particular system or piece of equipment is meeting operational expectations.

Finally, in “How to configure controls for economic steam dispatch,” Rawls discusses a proven controls scheme for the efficient production of steam from six boilers (unfired HRSG, duct-fired HRSG, and four conventional boilers) for two-dozen chemical manufacturing units typically requiring from 630,000 to 860,000 lb/hr of superheated steam at 615 psig.

Turbine Tip No. 9 by Dave Lucier, owner/GM, PAL Turbine Services, applies to General Electric package power plants (PPP), including the following: Model Series 5001P, 6001B, and 7001 B-EA.

Suppose your company has purchased a pre-owned PPP and must move it to a new location. Such resales have become more popular recently as the original purpose of this equipment (peaking and emergency power) has ended for many electric utilities. These plants can be relocated successfully, but it is important to engage a knowledgeable crew supervised by an experienced senior field engineer as technical director. This is no job for amateurs if you want to retain your asset’s value.

Preparation work should reverse the OEM’s original “as-installed” procedure from perhaps 40 or 50 years ago. Nobody working at your plant likely remembers that installation process. Plus, there may not be any GE installation records, field-engineer reports, or information available on how to move this unit to a new foundation.

Bear in mind that to safely uproot and move a GE frame gas turbine, specific procedures must be followed. Among them are those described below:

• • Remove the side panels adjacent to the turbine shell (Fig 1), taking note of the two shell supports, bolting, and dowels. Each support foot has four vertical bolts and one dowel (Fig 2).

• • Loosen the bolts for the two support legs shown in Figs 3 and 4 and install mechanical jacks temporarily at each of the two vertical-joint locations. Next, jack up the shell about 15 mils so a shipping pin can be installed in the lower centerline gib key. Shell lifted, tap loose the shims under the support feet, which may be rusted in place. Note that the turbine must “ride” on a shipping pin in case it is “humped” during transport. This can happen when the machine is lifted by crane or transported by truck or train. Damage could occur to the compressor blades and bearings if the unit is not prepared properly.
  • Bolts for the exhaust seal should be loosened, otherwise they may “snap” when the shell is jacked up. The bolts may be rusted, so be prepared to replace them after tapping the holes (Fig 5).

• • Notice the loosened fasteners and elevated dowel pin between the bolts at the left in Fig 6. The dowel remains in place, shims tapped loose, and support legs are not supporting any turbine weight. Fig 7 shows the pin installed to support the turbine shell during shipping to prevent “humping” of the shell. Red tag is to remind personnel that the dowel should be removed after jacking at the new location. Then the legs can be bolted down to support the shell.
  • The horizontal gib key bolts in Fig 8 can remain in place on each side as long as a 25-mil feeler gage can be slipped in to assure the shell (not shown) is not “pinched.” This will maintain the horizontal centerline and shell position. Once they are installed, the mechanical jacks supporting the casing can be removed.
  • Front flex-plate bolting need not be disturbed.
  • The rotor must be pushed axially against the active thrust bearing (located internally on the opposite end at the No. 1 turbine bearing) to keep the rotor secure during shipment. The axial internal clearance between thrust bearings is about 16 to 19 mils. Use a dial indicator (not shown) to “thrust” the rotor backwards against the active thrust bearing (Fig 9).
  • There’s no need to disturb the front compressor flex support when preparing for the move (Fig 10).

Turbine Tip No. 10 by Dave Lucier, owner/GM, PAL Turbine Services, applies to General Electric package power plants (PPP), including the following: Model Series 5001, 6001, and 7001.

GE PPPs originally had Cardox CO2 fire protection (Fig 1) with individual bottles of suppressant connected in two systems—initial burst and sustained delivery—to three compartments: accessory base, combustion, and load gear and exhaust area. Frame 5 and 6 engines with reduction gears had two overhead “trap doors” that closed when the discharge occurred, to mitigate CO2 flow into the air-cooled generator from the load-gear compartment, thereby protecting the generator from contamination.

Each of the three compartments was equipped with color-coded temperature sensors indicating the ambient temperature allowed (Fig 2). If the temperature exceeded the setting in the tip sensor, the bottled suppressant would be discharged to extinguish the fire.

Location, location. Most GE gas turbines had fire protection systems installed inside their control cabs, taking valuable space away from plant operators. Owners typically didn’t like having the bottles so close to personnel, often moving them to an adjacent building (new or existing), thereby opening up space for a desk and chair. Fig 3 illustrates a system and batteries that were moved outside the control cab.

Enclosures for PPPs are supposed to be sealed to contain any fire that might occur. The goal is to entrap the fire and smother it with suppressant, extinguishing it as soon as possible. Thus, door and panel seals must be maintained in good condition (Fig 4). Insurance companies may require plant operators to “prove” the integrity of the sealing system.

Many insurance companies now are requesting that owners consider replacing antiquated fire protection systems. But before doing this, be aware of the following onsite considerations with the existing enclosures and fire sensing systems:

• • Test the existing system for leaks—that is, deliberately discharge suppressant into the compartments to locate leaks, holes, rusted panels, etc. It is very likely that the door and panel seals have rotted, and the panels no longer fit properly to envelop and contain the compartment.
  • Refer to the schematic piping diagrams for the tip ratings of compartment temperature sensors. Typical settings are as follows:
    • • Accessory compartment, 45FA-1 and 45FA-2 (200F nominal).
      • Turbine combustion compartment, 45FT-1 and 45FT-2 (600F nominal).
      • Load gear (exhaust plenum) compartment, 45FT-3 and 45FT-4 (500F nominal).

Caution: The CO2 fire protection system is passive. It remains in standby mode until a fire occurs in one of the compartments.

Personnel should “pin” the system whenever personnel are working in the area to prevent accidental discharge. If you think this can’t happen consider the following experience: During cranking checks on MS5001L fuel-regulator controls, a flexible coolant line feeding the diesel cranking engine burst, igniting the ethylene glycol. A flame ball roared through the accessory compartment and the CO2 system discharged to extinguish the fire. The technician was not injured seriously—thankfully—and no lost-time accident was recorded.

Turbine Tip No. 12 by Dave Lucier, owner/GM, PAL Turbine Services, applies to General Electric package power plant (PPP) models MS5001D, L, and LA.   

GE installed compressor bleed valves (a/k/a recirculation valves) on these legacy gas turbines. Axial-flow compressors for the earliest units—those with 15 or 16 stages—had valves to recirculate air from the 10^th stage to the fourth, for “unloading” the machine during startup and shutdown to mitigate vibrations caused by the surge phenomenon. Two valves were installed for this purpose in the turbine compartment adjacent to the compressor casings (Fig 1).

Later turbine models, those with 17- and 18-stage compressors, were equipped with valves to bleed air from the 11^th stage to the turbine exhaust during startup and shutdown.

The recirculation valve shown in Fig 1 is open during startup and shutdown and later closed by compressor discharge pressure tapped from the 16^th stage via the small line on top of the valve (arrow). Depending on spring strength, the valve is fully closed at a PCD of about 30 to 40 psig. Note that the acronym for compressor discharge pressure, PCD, as found in early instruction books, was changed to CDP in the mid-1980s.

These recirculation valves do not have position “indicators.” If either or both valves should remain hung-up in the open position during online operation, performance (turbine power output) would suffer, because air flow to the combustors would be lower. However, no damage to the compressor should be experienced even if not fully closed when the unit is online.

My late former partner, Charlie Pond devised a simple solution to assist operators in knowing whether the valve is open or closed (Fig 2):

• • Remove the valve cover and drill a 9/16-in.-diam hole in the top beside the PCD tubing line.
  • Relieve the hole opening to accommodate an O-ring.
  • Determine the length of rod to be used as the indicator and mark it for the fully open and closed positions.
  • Select an appropriate O-ring to help seal the rod from excessive air leakage.
  • Grease the rod and O-ring, making sure the rod moves smoothly as the valve strokes from open to closed.
  • Install a pressure gage in the PCD supply line.

Note that it’s best to test the valve travel while the gas turbine is shutdown.