Big Sandy Plant

1
Big Sandy Plant

• Update
• Loesche Dynamic Classifier
• 2012

2
Executive Summary

• 20 increase in fuel thru put was noticed on this
  pulverizer
• Total overall project cost 1 million (labor
  and material)
• Project time line - 8 weeks (one shift) 6000 man
  hours
• Initial testing shows installation on all 6
  pulverizers would produce a spare pulverizer
  (based on this one)
• System wide revenue potential 9.5M based on
  lost MWHs from GADS data
• Other potential benefits include NOx reduction
  with improved fineness, increased mill
  performance and increased boiler efficiency thru
  improved combustion
• Project under review by AEP Program Management
  Office as an Engage 2 Gain idea
• All craft hours removal and install

3
Background

• Big Sandy Plants Unit 2 is an 800 MW Foster
  Wheeler Boiler with six MB23 hybrid MPS 89
  Pulverizers.
• Full load requires all six Pulverizers.
• One Pulverizer O/S is approximately 100 MW
  curtailment.

4
History

• MPS mills in 2010 accounted for 286,080 MWHs of
  curtailments on the 800 and 1300 series units.
• The 800mw series without spare mills lost 263,850
  MWHs while the 1300mw series with spare mills
  only lost 22,230 MWHs
• Inspections and rebuilds accounted for 150,000
  MWHs of curtailments in 2010 on the 800mw
  series.
• Pulverizer Inspections (only) were the 4th
  Largest driver of Plant EFOR in 800mw units in
  the last 7 years.
• Given these statistics AEP is looking for ways to
  reduce these curtailments.

5
Project

• Loesche approached AEP and it was agreed to
  install one of their Dynamic Classifiers at Big
  Sandy Plant on the 21 Pulverizer.
• Loesche claimed 15 - 20 increase in thru put at
  the current fineness or improved fineness at same
  thru put.
• If successful and installed on all mills, this
  would allow minimal unit load curtailment when
  one mill is out of service.

6
Removal of existing classifier

• During the fall outage of 2012 the installation /
  removal began on Pulverizer 21. All the coal
  piping, shutoff gates and walkways on top of the
  mill had to be removed to clear the top area and
  the classifier reject chute was cut loose from
  the inside. The existing seal air piping for the
  roll wheels was removed to later be reinstalled
  going thru the new classifier housing.
• Once these items were removed, the top flange of
  the classifier housing was all that needed to be
  unbolted to remove the housing and classifier
  assembly as a whole. Once removed the reject
  chute was lifted up thru the top opening of the
  mill.
• No other internal parts had to be removed and the
  large maintenance doors stayed closed on the
  mill. For safety reasons, the spring tension was
  released for fear of not knowing what forces
  could be on a 40 year old housing after removing
  the top section.

7
Removal Continued

• The removal of the existing classifier and
  housing had some challenges. The existing
  trolley beams could not make the lift due to the
  angle of the pull. Lifting beams had to be
  fabricated to keep a straight pull on the
  structure.
• The tight clearances were the biggest challenge
  to overcome. We had to find dimensionally small
  chain hoist and shackles that were load rated.
  Once found, there was only 4 inches clearance
  between the upper and the lower housing.
• The yoke seal air piping on the pulverizer next
  to this one had to be removed to allow for room
  to lower the old classifier and housing to the
  ground. There was only a few inches on each side
  between the two pulverizers when setting on the
  floor.

8
(No Transcript)
9
Background continued
10
(No Transcript)
11
(No Transcript)
12
Installation

• The new classifier was made in two sections.
  This helped with the installation since the
  overall height of the new classifier was taller.
  The lower housing section included the classifier
  lower cone which hung below the flange line.
  This required this section to be raised high to
  clear the housing on the pulverizer. Loesche
  accounted for these issues in the design and
  engineering and installed lifting lugs such that
  the new lifting beams worked on the two new
  sections.

13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
Installation

• The classifier fit perfectly. No modifications
  were needed.
• The ancillary equipment however did need slight
  modifications. The walkways, platforms and seal
  air pipe had to be modified in the field to make
  them fit. Loesche performed a laser survey of
  the area prior to the project and used it in the
  design all of the equipment. However, there were
  still some modifications necessary. Loesche
  provided quick assistance on correcting these and
  any other problems or issues that would arise.
• Some of the new coal pipe spool pieces were
  already completely fabricated while some required
  to be field fit and the flange welded on.

19
Installation

• Due to limiting seal air available from our
  existing pulverizer seal air system, Loesche
  provided two seal air fans to provide seal air
  for the bearing cavity on the classifier a duty
  and backup fan. These were sized to provide
  enough air for any additional classifiers and
  occupy approximately 6 ft x 10 ft section of
  floor.
• We also had no expansion room in our existing DCS
  system and limited available power in the local
  motor control center. Loesche designed and
  provided a PLC control system, a variable speed
  drive unit for the classifier motor, a
  transformer, a motor control section and a HMI
  for the new classifier. All of these items
  required approximately 6 ft x 15 ft section of
  floor.
• The new MCC section had to be powered from the
  600 VAC switchgear in the 4kv room. Overall
  there was a considerable amount of cable pulls
  that had to be made.

20
Block diagram of cable
21
(No Transcript)
22
(No Transcript)
23
Installation / Removal

• Overall, the installation and removal went well.
  There were minor issues that were always quickly
  resolved.
• The total contractor labor was approximately 6000
  man hours for both the removal and installation.
  This is for all crafts electrical, mechanical,
  insulators (lead abatement) and the scaffold
  contractor.
• The controls were a bit more difficult. The
  company that had built and tested the PLC
  programming logic inadvertently had left
  simulated points in the programming and did not
  get all the point names changed to the correct
  program before shipping. This took several hours
  to decipher before things worked as they should.

24
Installation / Removal continued

• Reasons for high man hours 1) First time
  evolution. 2) This pulverizer has limited access
  from one side. All coal pipes, platforms and
  anything lifted had to be coordinated and made
  from one side. (Any other pulverizer 22 26 has
  access from both sides making coordinating
  lifting easier.)
• Loesche had a service man on site to help oversee
  the installation, startup and testing during the
  project. They provided guidance, answered
  questions and found solutions to any issue. They
  wanted to prove success as much as we wanted it.

25
Future Installs

• For future installs on the remaining 5
  pulverizers, the electrical system would have to
  be copied 5 times over with the exception of the
  PLC. The MCCs could be reduced in foot print
  but would need to be fed from more than one 600
  VAC buss to obtain reliability. This would
  require several square feet of real-estate.
  (This may not be required for everyone)
• While the seal air fans provided by Loesche are
  larger than required for the one classifier,
  consideration should be given to power feeds to
  provide reliability in the event a 600 volt buss
  had to be removed from service.
• Overall project cost would be reduced. Most
  engineering and design should be completed.
  Advanced planning and coordination would reduce
  labor time since this has now been performed.

26
Testing

• No other internal work or maintenance was
  performed on this pulverizer to obtain a true
  test of what the classifier could produce on its
  own.
• Once the unit returned to service, testing was
  performed following the procedure from Loesche.
• Testing was conducted before and after the change
  out to determine what if any improvement was made
  with the new classifier.
• AEPSC, Loesche, REO and plant personnel assisted
  in the test work using the EUCoalsizer laser
  probe to perform the fineness testing.

27
Test continued

• The following are the test results

  Classifier RPM Coal Flow 50 - MESH 200 - MESH HGI
PRE ------------------ 87 99.35 50.8  
POST 70 90 99 53.7 48
  75 105 99.7 60.025 48
  80 110 99.725 61.75 48
  95 90 99.525 62.35 48
  70 105 98.775 55 51
  70 110 99.1 58.375 51
  70 115 99.325 58.475 51
  65 115 98.2 56.15 51
  100 87 99.75 66.95 51
28
Testing Conclusion

• While the goal was to increase thru put without
  effecting fineness, it can be seen in the test
  results this was achieved along with the
  potential for increased fineness.
• This mill was previously limited to approximately
  87,000 90,000 lb/hr coal thru put due to
  excessive dribble and primary air limited. There
  were multiple fires from coal build up both
  internal in the mill and in the pyrites in the
  past.
• With the Loesche Classifier, we were able to
  obtain significantly more PA Differential and
  pass 115,000 lb/hr coal flow with out any dribble
  while maintaining fineness at pre change out
  levels.

29
Conclusion

• When the unit returned to service this mill had
  an original 600 hp motor. While testing at
  115,000 lb/hr coal flow, mill amps continued to
  increase over time. Fearing an overload trip,
  the mill was reduced to 110,000 lb/hr without
  mill amperage increasing. This is over a 20
  increase in thru put on this mill.
• Mill amperage was a limiting factor in the higher
  RPM trials while determining the improvement in
  fineness and this is why the faster speeds could
  not be maintained and allow more thru put.
• During our most recent outage, this pulverizer
  motor has been increased to 700 hp and new
  testing will be conducted to determine maximum
  output on the mill and also an optimum speed for
  the classifier to obtain the best fineness with
  maximum output.

30
Conclusion

• While the Loesche Classifier did what they
  promised, other manufactures also make dynamic
  classifiers.
• Other technologies that predict saving time with
  reduced maintenance, provide increased thru put
  and improve mill performance need to be looked at
  and tested.
• We (AEP) need to continue to focus on improving
  pulverizer performance both in output and reduced
  outages. Remember 286,080 MWHs is a lot of
  curtailments and lost revenue.

31
QUESTIONS??