20060228Dempsey direct.pdf

: '
: r"
:, " " ,:;:..; ;" :-;: ~
DAVID J. MEYER
VICE PRESIDENT, CHIEF COUNSELFORREGULATQ~yji!LS C U;\I,SiCiiAND GOVERNMENTAL AFFAIRS
AVISTA CORPORATION
O. BOX 3727
1411 EAST MISSION AVENUE
SPOKANE, WASHINGTON 99220-3727
TELEPHONE: [phone]
FACSIMILE: [phone]
BEFORE THE IDAHO PUBLIC UTILITIES COMMISSION
THOMPSON RIVER CO-GEN, LLC
a Colorado Company,CASE NO. AVU-05-
COMPLAINANT
vs.
A VISTA CORPORATION dba A VISTA DIRECT TESTIMONY
UTILITIES , a Washington Corporation
THOMAS C. DEMPSEY
RESPONDENT.
FOR A VISTA CORPORATION
I. INTRODUCTION
Please state your name, employer and business address.
My name is Thomas C. Dempsey. I am employed as Manager, Thermal
Engineering, by A vista Corporation and my business address is 1411 East Mission Avenue
Spokane, Washington.
Please state your educational background and professional experience.
I graduated from the University of Texas at Austin in August 1993 , with a
Bachelor of Science in Mechanical Engineering, with a major in thermal/fluid systems, of
which power plant design and analysis is a component. I have taken numerous training
courses during the course of my career, including ThermotlexCID software training.
ThermotlexCID is an industry standard power plant design and performance modeling software
package sold by Thermotlow, Inc.
In December of 1993 , I accepted a position in Houston Lighting & Power
Performance Analysis group.My responsibilities in that position included performance
testing of various types of generating units, including natural gas-fired Rankine cycle units
natural gas-fired combustion turbines, and coal-fired facilities. Testing of those facility types
encompassed performance efficiency and emissions testing and tuning activities.
I joined Washington Water Power (now A vista) in December of 1996 as a Production
Engineer in the Generation Engineering group. My job responsibilities included management
of engineering projects for the Company s hydro and thermal facilities as well
performance testing for Avista s wood-fired facility in Kettle Falls, Washington. In that role
Dempsey, Di
Avista Corporation
I also had responsibility for development of the preliminary conceptual design for Avista
Kettle Falls Generating Station combined-cycle project.
I received my Professional Engineering license in January, 1998. In 2001 , I was
promoted to the position of Combustion Turbine Specialist, which included responsibilities
for supervision of operation and maintenance for Avista s Northeast and Rathdrum natural
gas- fired combustion turbine facilities and the Boulder Park Generating Station, a natural
gas-fired reciprocating engine plant. In 2004, I became Manager of Thermal Engineering
which, in addition to the operations and maintenance responsibilities described above, added
responsibilities for engineering supervision for all of Avista s thermal generating facilities
(Kettle Falls Generating Station, Rathdrum Combustion Turbine, Northeast Combustion
Turbine, Coyote Springs 2, Boulder Park Generating Station), with the exception of Colstrip.
What is the scope of your testimony in this proceeding?
I will further explain the basis for the Company s determination that the net
output capacity of the TRC project is greater than 10 aMW even with the new environmental
permitting requirements. My testimony will explain the relationship between boiler heat
input and steam output levels and the amount of generation net output. I will provide an
overview of TRC's emissions test results for NOx and SO2. I will also review the Montana
Department of Environmental Quality s (MDEQ) recent air quality permit Preliminary
Determination on the TRC project's request for increased levels of NO x and SO2 emissions.
I will review the most recent TRC emissions test data and discuss why it is reasonable to
expect that the emission requirements outlined in the MDEQ Preliminary Determination will
not reduce the project's ability to produce a net output in excess of 10 aMW.
Dempsey, Di
Avista Corporation
I am sponsoring the following exhibits listed in the table below, which were prepared
under my direction:
lExhibit No.Description
211 TRC Hourly Net Output - September 2005
212 TRC - Hourly Net Output - 66- Hour Test Period (9/1 0-9/12)
213 Summary Comparison - TRC Air Permit Requirements
214 TRC - NOx(36% Reduction) - 66-Hour Test Period (9/10-9/12)
What review of the TRC project did you perform?
On page 1 of a letter to the MDEQ, dated January 4 2006, TRC explains that
the "emission rate proposed as NOx BACT (best available control technology) is based on
the September 2005 boiler testing performed by Luke Conner, owner of CPL Systems." The
Company has, therefore, based its review of the project's emissions and net generation over
this same September testing period.
I performed a review of hourly generation net output, NOx emissions, heat input, and
steam output for the September 2005 testing period. In particular, I reviewed NOx emissions
heat input, and steam output data from the MDEQ records for the 66-hour period from
September 10, Hour 6 through September 12, Hour 23. I also reviewed the hourly net output
data from the same 66-hour period and found that the project net output averaged 11.4 aMW
Why did you pick this 66-hour period during September?
The project operated a total of eight days in September. The 66-hour period
spans three days and represents the last multi-day continuous test period for the project and
Dempsey, Di
Avista Corporation
includes those hours after start-up with the project fully stabilized and generating at nearly
full output levels.
How was the hourly net output data for the TRC project during the
month of September obtained?
Exhibit No. 211 shows hourly net output graphs for every day of testing
during September 2005. These numbers were obtained from TRC's Response to Production
Requests Nos. 29 and 36. Each hourly data point includes the metered Northwestern kWh as
well as the hourly average Thompson River Lumber energy consumption. The Thompson
River Lumber daily consumption varied, but averaged 0.399 MWhIhr each day during
September.
As the project was in a "testing and tuning" mode, it was often ramping up or down or
otherwise simply operating at other than full net output levels. Those hours of reduced net
output effectively lowered the daily net output average amounts below the level of what
would otherwise be the case for a continuously operating project. Consequently, a review
the hourly net output data shows an even higher sustained capability than the daily average
amounts addressed in Mr. Perk's testimony.
What was the hourly net output for the TRC project during the 66-hour
continuous operation period?
Exhibit No. 212 shows the hourly net output for the 66-hour period in September. The
project averaged 11.4 aMW
Was the Thompson River Lumber steam load being served by the TRC
project during this September 10 through September 12, 2005 66-hour period?
Dempsey, Di
A vista Corporation
No. Based on TRC's response to a data request for steam sales information, it
was indicated that the Thompson River Lumber steam supply system was not on line during
any of the September 2005 testing period.
Will you please review the net generation effects expected when steam is
supplied to Thomson River Lumber.
Net generation is reduced when the TRC project supplies steam to the
Thompson River Lumber Company; this represents an energy loss from the perspective of the
steam turbine generator cycle. In order to quantify this effect, the Company developed a
computer model of the plant using ThermotlexCID power plant design and modeling software.
Components for the model were chosen to match the components shown in the TRC project
heat balance that we had observed during a site visit. Using the ThermotlexOY model, the
Company chose various lumber mill steam demands to calculate the reduction in net output.
As stated previously, during the September testing, it was noted that the TRC project was not
supplying steam to the lumber mill. Using ThermotlexOY to simulate a 9000 lb/hr steam
supply to Thompson River Lumber reduced the predicted net generation by approximately
35 aMW, compared to conditions where no steam is sent to the lumber mill.
Please discuss the boiler output levels observed during the 66-hour testing
period.
Based on boiler load data obtained from the MDEQ, it was also observed that
the boiler was operating below full output, by an average amount of approximately 6%,
during the 66-hour testing period discussed above. Again using the ThermotlexOY model, the
Company determined that the average net generation during that 66-hour period would have
Dempsey, Di
A vista Corporation
been approximately 0.8 aMW higher if the boiler had been operated at its Maximum
Continuous Rating (without exceeding its heat input limit) for steam supply throughout the
period.
After simulating steam supply to Thompson River Lumber and assuming
operation of the boiler at its Maximum Continuous Rating for steam without exceeding
the 192.8 million Btulhr limit on fuel, what is your calculated average net generation for
the period?
Using the ThermotlexOY model to simulate both the anticipated reduction in
net generation due to a 9000 lb/hr steam supply to the lumber mill and the anticipated
increase in net generation that could be realized if operating the boiler at its maximum
continuous rating, yielded a predicted average net generation during the period of 11.8 aMW
The hours immediately preceding and following this time period were start-up and shut-down
periods where boiler load and output were both substantially below design and were not
considered. The net output level of 11.8 aMW is even higher than the 11.4 aMW discussed
earlier in my testimony, and reflects the net effect of further adjusting for both steam service
to Thompson River Lumber and operating at the maximum continuous rating.
Will you please review the TRC project test results for NOx?
The MDEQ rejected TRC's proposed NOx limit increases in itsYes.
February 10, 2005 Preliminary Determination regarding TRC's application for a modified air
quality permit.The MDEQ instead required a NOx BACT (best available control
technology) limit of 0.178 Ibs/MMBtu as measured on a 3-hour rolling basis and installation
of a Selective Non-Catalytic Reduction (SNCR) system. In calculating the NOx limit, the
Dempsey, Di
Avista Corporation
MDEQ assumed a SNCR NOx reduction efficiency of 36%. Exhibit No. 213 shows a side-
by-side comparison of NO x and SO2 emissions limits under TRC's current air quality permit
TRC's November 16, 2005 proposed air quality permit modifications, and the MDEQ'
Preliminary Determination requirements.
The Company has reviewed TRC NOx emissions levels from MDEQ data records for
the month of September 2005. The Company has, in particular, reviewed test data from the
66-hour period, September 10 2005 through September 12, 2005. As mentioned previously,
that period represents the last multi-day continuous test period for the project and includes
those hours after start-up with the project fully stabilized and generating near full output
levels.
NOx emissions averaged 0.137 Ibs/MMBtu for this 66-hour period when data is
adjusted to reflect a 36% reduction in NOx using technology referenced as BACT by the
MDEQ Preliminary Determination. Exhibit No. 214 shows the NOx levels for each hour of
operation. After the project is up to steady-state, full load operation and when NOx emissions
are adjusted downward by 36%, the NOx levels all fall below the 0.178 Ib/MMBtu MDEQ
Preliminary Determination limit for NOx. This indicates that the project can still operate at
levels in excess of 10 aMW. even with the environmental permit limits.
Will any SO2 limits required by the MDEQ result in reduced boiler
capability?
The MDEQ states on page 3 of the Preliminary Determination that
, "
SO2
emissions from the boiler shall be controlled by a flue gas desulfurization (FGD) unit when
combusting coal. ..." The MDEQ specifies on page 2 of the Preliminary Determination a
Dempsey, Di
Avista Corporation
206 Ib/MMBtu limit on SO2 emissions. The MDEQ references a number of technologies
that are capable of meeting the SO2 emission limits in their Permit Analysis section. On
page 26 of the Permit Analysis section, the MDEQ states
, "
For LSD FGD guaranteed 85%
SO2 removal can be obtained from control equipment manufacturers.... (emphasis added)
Therefore, technologies are available to TRC to limit SO2 emissions, without otherwise
impacting the project's ability to generate in excess of 10 aMW. (TRC will have to employ
these technologies, in any event, to operate the plant.)
By way of summary, can you comment on the expected impact of the new
NOx and SO2 technologies on project net output?
On page 7 of its Application for Air Quality and Operating Permit
Modifications dated November 11 , 2005 TRC notes that a SO2 limit expressed in
Ibs/MMBtu does not limit fuel input into the boiler. Correspondingly, the steam generated is
not limited. The MDEQ Preliminary Determination maintains this type of specification with
respect to both the SO2 and NOx limits. The MDEQ did, however, maintain the boiler heat
input limit of 192.8 MMBtu/hr. The Company s simulations already reflect this limitation as
well as the maximum continuous steam rating of the boiler. Accordingly, even with the
environmental permitting constraints of the MDEQ, the project's net output will exceed 10
aMW
Does that conclude your pre-nIed direct testimony?
Yes it does.
Dempsey, Di
A vista Corporation
SEE CASE FILE FOR
DEMPSEY EXHIBITS
211 - 212 213 - 214
WHICH COULD NOT BE
SCANNED SUCCESSFULLY