By Victor de Biasi

Humid and dry air injection represents a relatively low-costtechnology for increasing gas turbine power output and re-ducing heat rate over a wide range of ambient temperatureand part-load operating conditions.

S everal companies are evaluating

the technology and cost effective-

ness of injecting externally com-pressed, heated and humidified air (canalso be dry air) to increase gas turbineoutput during hot day operation.

Air augmentation is a relativelylow.:cost technology applicable tosimple cycle and combined cycleplants on a new installation or retrofitbasis.

Enhanced performance is most dra-matic at higher temperatures and alti-tudes but is also attractive at lowertemperatures:

D Increased power. Humidifiedair injection can increase gas turbineoutput by up to 20% at 95°P ambienttemperature.

turbine hot day loss is that inlet air los-es density, thereby reduceing mass flowand hence power output.. Compressorsection power consumption also in-creases because it must work harder asair density falls off, further reducing netplant output.

In effect, with air augmentation, gasturbines operate as though at muchlower ambient temperatures. Dr.Michael Nakhamkin, president of En-ergy Storage and Power Consultants,developed and owns the patents onthe use of air injection technology fqrgas turbine power augmentation.

Last year, Calpine obtained a non-exclusive license from ESPCto applyhumid air technology to its servicefleet of simple cycle gas turbine andcombined cycle power plants.

The company has since validatedperformance improvement guaranteesand is getting ready to run more ex-tensive tests at higher power ratings.

Calpine conducted its validation

0 Lower heat rates. Can reducegas turbine heat rate by 8% to 15%over entire operating range from fullto part-load output.

power augmentation are similar tothose for steam injection.

The same restrictions apply to makesure that the incremental increase inpower does not exceed operationallimitations.

Incremental increases in power mustbe kept within the bounds of maximumallowable torque, compressor surge,combustion parameters, flame stability,and electric generator capacity.

Air augmentation works by introduc-ing an external supply of pressurizedair into a gas turbine upstream of thecombustors to increase mass flow. Theincremental power added is propor-tional to the mass of the injected flow.

The air can be either humidified ordry. Humidification has the addedbenefit that it reduces the amount ofsupplementary air needed to increasethe gas turbine total flow and saves oncompressor work otherwise needed topressurize a higher flow.

The simplified explanation for gas

D Lower emissions. Can reduceNOx emissions by 15% (lbs per kWh)compared with simple cycle gas tur-bine output.

0 Installed costs. Turnkey cost ofhumidified air augmentation systemestimated at roughly $180 to $220 perkW.

Engineering and mechanical aspectsof injecting air into a gas turbine for

12 GAS TURBINE WORLD: March-April 2002

Humid and Dry Air Injection Technology License

C alpine has a non-exclusive license to apply humidified and dry air

augmentation technology to its fleet of simple cycle and combined cy-

cle plants.As part of the license agreement, Calpine ran a series of validation tests

on a PG7241 FA peaking installation, Unit #3, at its Broad River Energy

Center in Cherokee County, South Carolina.Test measurements were made while operating with humid air injection,

dry air injection, steam injection, and without any injection at all to measureactual power output and heat rate at different ambient temperatures undereach of those operating conditions.

Results said to have confirmed and exceeded air augmentation perfor-mance guarantees at 95°F and 60% relative humidity site conditions with3.5% humidified air injection - which is the maximum moisture level that

GE specifies for steam injection operation.Output confirmed the projected power augmentation of 18.3 MW when

operating on "dry" control which is consistent with a projected output of

22.3 MW while operating on "wet" control. No known operational limits

were exceeded.Calpine is expected to run follow-up tests this summer to evaluate in-

creased levels of humidification and to optimize operating conditions. An

extrapolation of test results indicates that increasing the humidity level toonly 5% will increase output to over 28 MW on "dry" and to over 32 MW on

"wet" control.

Humid air injectionSimplified schematic of humidified air injection plant configuration. Major com-ponents include a compressor train (commercial off-the-shelf compressors ormodules), a saturation column, heat exchange unit, and balance of plant equip-ment such as interconnecting piping, valves and controls.

Gas Turbine

Filter t

AmbientAir

Fuel4-"

t~~:-7:-Compressor Train t

HeatRecoveryUnit

~' J.

~Moist Air

~AirSatura

Ambient Air ~

tests on a PG7241FA gas turbinepeaking installation, unit #3, at theBroad River energy center in SouthCarolina.

This particular gas turbine isequipped with a once-through steamgenerator and ports for injecting high-pressure steam that is generated intocombustors for power augmentation.

Those same steam ports are easilyadapted for air injection so that pro-ject test engineers are able to evaluategas turbine performance with andwithout humid air injection, dry airinjection, steam injection and, for areference base, no injection at all.

Special measures were taken tomake sure air injection would notdamage the gas turbine during the testruns, including instrumentation tomonitor and record compressor flowdisturbances, mechanical and thermalstresses, vibration levels, and combus-tion dynamics.

As a precaution, for this initial se-ries of tests, the pressurized and hu-midified air injection for power aug-mentation was limited to the same3.5% moisture limit that GeneralElectric specifies for Fr7FA steam in-

jection power augmentation.Calpine's external air pressurization

and humidification system consists oftwo motor-driven air compressors withintercoolers, an air pre-heater, and amixer supplied with steam to humidifythe pressurized air for injection (doesthe same job as the air saturator).

Test results confirmed an 18.3-MWincrease in power output at 95°F inletair temperature with the 3.5% humidi-ty limitation.

Upcoming tests are expected to runat humidity levels of 5% and then goprogressively higher to establish com-fortable operational limits for opti-mum performance.

It is reasonable to believe that someof the tests will also be directed to-wards determining whether turbinerotor inlet temperature control withair augmentation should be referencedto "wet" or "dry" curve operation.

Normal control routine when usingsteam injection for power augmenta-tion is to operate on the "dry" curvewhich reduces turbine rotor inlet tem-

t~ Stack

InterCoolers f

-Make-up Water

133AS TURBINE WORLD: March-April 2002

~lor r

perature by 30oP to 50oP to guardagainst over-temperaturing the turbineblades.

Steam has a much higher heat trans-fer factor than air. As a result, saysN akhamkin, any steam coming intocontact with surrounding surfaces ofcombustors, transition pieces and tur-bine blades can create localized ther-mal spikes that overheat and, in time,cause damage.

Pockets of steam can also cause sig-nificant pressure oscillations in thecombustors. Humidified air does notpose any of those same problems. Itsheat transfer characteristics are lowerand therefore more benign than withsteam. There is not the same risk of lo-calized overheating as there is withsteam.

Technically, at least, a good argu-ment can be made that humidified airshould be able to have higher moisturelevels than deemed acceptable for

steam injection.It also suggests that humid air inject-

ed gas turbines should be able to oper-ate on the nonnal "wet" control curvewithout having to reduce turbine rotorinlet temperature.

Calpine's Fr7FA test results, for ex-ample, measured an 18.3 MW increasein power while operating on the "dry"curve (reduced firing temperature).

On "wet" control, the increase inpower output at 95°F would have beencloser to 22.3 MW - even with only3.5% humidification.

Dr. Michael Nakhamkin, president ofESPC, says that increasing the humidi-ty level to only 5.0% would have in-creased net power output by more than28 MW even even with "dry" curvecontrol.

Operating on the "wet" curve,Calpine should expect to see a morethan 32 MW increase.

Dry air is not as effective as humid

air injection. But dry air also can in-crease power output at any ambienttemperature - again subject to torque,

compressor surge, combustion, torqueand electric generator limitations.

At 95°F ambient, for example, thePG724lFA operates at 150.4 MW out-put with a heat rate of 9760 Btu/kWhwithout an evaporative cooler, inletchilling, or inlet fogging.

Dry air injection, studies indicate,can increase output to an estimated166.4 MW with a slight improvementin heat rate to 9720 Btu/kWh.

Dr. Nakhamkin sees steam injectionas a major but less effective competingtechnology for gas turbine power aug-mentation.

Humid air provides at least 35%more power augmentation than steaminjection, he claims, and slightly betterheat rates.

Steam injection also has some majordrawbacks when it comes to servicelife, operating costs, and operationalflexibility, he says.

Reportedly steam augmentation cancut the life of "hot gas path parts" inhalf, even for operation at lower tur-bine inlet temperatures.

There should be no impact all on ser-vice life with dry air injection, hemaintains, and significantly less im-pact with humid air injection. .

Further, humid air injection does notrequire demineralized water so thatthere is a significant savings in terms oftreatment, personnel, and operating andmaintenance costs otherwise associatedwith the use of demineralized water.

Lastly, air injection can introduce in-cremental power instantly as oposed tothe 30 to 40-minute delay typical forsteam injection, says Nakhamkin.

Instant start-up may not be an advan-tage for base load operation but is im-portant for short-duration daily peak-ing cycle installations.

Humid air injection is also said to besuperior to inlet fogging and wet com-pression for improving hot day outputand heat rate.

Water is introduced after the air hasbeen compressed, he explains, whereasfogging and wet compression adds wa-ter to the compression cycle.With fogging or wet compression, the

14 GAS TURBINE WORLD: March-April2002

Humid air injection NOx emissionsLogarithm plot of test combustor results shows dramatic decrease from closeto 100 ppm NOx without any humid air injection to less than 10 ppm at 10%moisture level and less than 5 ppm at 15% moisture level. Combustion be-came unstable at around 30% humidity. Tests sponsored by EPRI were con-ducted by Textron and Aero Industrial Technology.

100 ppm'- ~ ~ ~ ~ ,In

C0'inIn

'Ew><

0z

~ ""-10 ppm

... .ppm ..

Moisture Levels (Water-to-Air)

10% 20%1-0.1 ppm 0% 30%-1--

iiii'YYJJJJJ,JJIilJJiJJEstimated Fr7FA Humid Air Plant Costs """"" "

Turnkey cost for supply and installation of Fr7FA-sized humid air plant based onnon-competitive quotes for engineering, construction and equipment. Total costis estimated for budgeting purposes at around $4.75 million.

.$ 2,940,000

..1,000,000. . . . 18,000

. . . 700,000

. . . .600,000

. . . . 50,000

. .. 45,000

. . . . 80,000

. .. 29,000

. . . . 98,000

. . . 320,000

Mechanical equipment. . . . .2.35 MW compressor train.

auxiliary circ water pumps. . .

heat recovery unit. . . . . . . . .

air saturation column. . . . . .

wet cooling tower. . . . . . . . .

circulating water pump. . . . .

make-up water tank (erected)

make-up water pump. . . . . .

water pre-treatment system.

piping, valves, insulation. . . .

. .$310

...253

. . . .34

. . . .23

Electrical equipment. . . . . . .components and sub-systems

instrumentation. . . . . . . . . . .

control system (PLC)

$1,496,000. . . . . . .203,000

. . . . . . .485,000

. . . . . . . .85,000

125,000. . . . . . . .58,000

. . . . . . . . 60,000

. . . . . . . .80,000

. . . . . . . 400,000

vaporized water must be pressurizedalong with the air, significantly in-creasing compressor power require-ments. In addition, only demineralizedwater must be used to avoid blade ero-sion attacks and compressor fouling..

Both dry air and humid air augmen-tation technology can be retrofitted toexisting installations. Humid air injec-tion is particularly effective for gas tur-bines not equipped with dry low NO;:combustors.

Non-DLN combustors allow opera-tion at higher moisture levels leadingto higher relative power augmentationand lower heat rates.

In general, older gas turbines havemore forgiving surge margins that cantolerate higher volumes of air injectionfor greater output without interferingwith compressor operation.

TV A engineering studies indicatethat gas turbines like Fr7Es that havedecent heat rates, stand to benefit themost because they can be readily up-graded to be more competitive withadvanced FA, G and H technologysimple cycle peakers and combined

cycle plants.Air augmentation may also make

sense for gas turbine powered pipe-line compressor stations where unitsizes are relatively small and water isnot available. It offers a relatively in-expensive way of expanding stationcapacity without have to permit andinstall additional gas turbine sets.

For stations with multiple gas turbineinstallations, dry air injection systemcomponents could be optimized toserve overall power augmentationneeds - with significant cost savingsand increased system reliability, avail-ability and flexibility.

An initial budget for a Fr7FA-sizedpackage, based on bids submitted bycompetitive suppliers for major itemssuch as the compressor train and satu-rator, came in at $4.75 million total.

That is still probably on the highside, says Nakhamkin, as evidenced bythe fact that Calpine was able to shave$400,000 off the estimated cost of thecompressors they installed. On that ba-sis, he considers $5 million as a goodconservative number for budgetingpurposes. .

Other project costs. . . . . . .

foundations, civil works. . .

erection, contingency, profit

project management. . . . .

engineering. . . . . . . . . . . .

logistics and shipping. . . .

training overhead and warranty. . .

contingency and profit. . .

.$ 4,746.000Total cost

GAS TURBINE WORLD: March-April 2002 15

,000

,000

,000

,000