Back to Publications

Gossman Consulting, Inc.

Preparation for CMS Plan as Part of MACT

 

David L. Constans

Gossman Consulting, Inc.

constans@gcisolutions.com

  

ABSTRACT


This paper discusses the requirements of the HWC MACT CMS Performance Evaluation Plan and provides a model plan.  This evaluation plan must be submitted with the ICPT plan and is to specify the inspection, maintenance and external quality control of the monitoring system components; thermocouples, D-P cells, opacity monitors and CEMs.  Comments from the EPA on submitted plans will be included and discussed if these comments are available prior to the conference paper submittal date.


INTRODUCTION


The HWC MACT rule requires the submittal of a Continuous Monitoring Systems (CMS) Performance Evaluation Test Plan with the submittal of the Initial Comprehensive Performance Test Plan (40CFR 63.1207(e)).  The CMS Performance Evaluation Test Plan requirements are detailed in 40CFR 63.8(e).  Also 40CFR 63.1207(d) requires that the quality control program outlined in 4CFR 63.8(d) also be included in the performance evaluation.  To complicate things somewhat the HWC MACT regulation includes in 40CFR 63.1209(a), (b) and (c) requirements for specific parts of the CMS system.  To complicate things still further the EPA issued a technical amendment to the HWC MACT (Nov. 9, 2000 volume 65, number 218, page 67268-67272).  Part of the amendment was to clarify the difference between a CMS Evaluation Plan and a CMS Evaluation Test Plan and the requirement to submit only the CMS Evaluation Test Plan with the Initial Comprehensive Performance Test Plan.

In the amendment the agency has defined the CMS Evaluation Test Plan to include “testing necessary to demonstrate calibration, minimization of malfunction, and how the CMS will meet the performance specification.”  The CMS Evaluation Plan is to include the quality assurance programs that “specified how a source would maintain calibration...and minimize malfunction.”  It would appear that the CMS Evaluation Test Plan is to exclude QA programs, yet somehow “minimize malfunctions” and “meet...performance specifications.”  Both of which would require such programs.  Also 40CFR 63.8(e) defines CMS Evaluation Test Plan as including the QA programs.  The only resolution that readily appears is to include the QA programs but without specifying the time intervals of the activities.

It had been hoped that one or more of the three CMS Evaluation Test Plans written by GCI for our clients and submitted to the regional EPA, would have been reviewed by that authority, and such review included as an additional guide in preparing this paper.  Unfortunately this has not occurred.  Consequently this paper presents a format for a CMS Evaluation Test Plan that has not been critiqued by the agency.

 

CMS EVALUATION TEST PLAN FORMAT

 

The CMS Evaluation Test Plan has been reduced to two sections.  A master table that provides a listing of the monitored parameters, the device used for such monitoring, the range of measurement, accuracy of measurements (data quality objectives), internal QA and external QA activities, record locations, and comments.  See attachments.\

The second section includes subsections for each monitored parameter.  A subsection includes a table repeating the first four columns of the master table for each of the monitored parameters and text describing the internal and external QA activities and any routine maintenance performed on the device.

Description of the Elements Which Make Up the Table and QA Program Subsections


Parameter Monitored:


These are the parameters that must be continuously monitored as specified in the HWC MACT permit.


Device Description:


This is the manufacturer and model number of the measurement device.


Range and Units of Measurement:


This should be the range the recorder attached to the device and the units of measurement of that recorder.  The actual range of the measurement device may be larger but exceeding the range of the recorder is a violation of the regulation.


Accuracy of Measurement (DQOs):


This is a value expressing the allowable inaccuracy of the measurement device.  In some instances, these values are dictated by the Performance Specification (PS1 (Opacity Monitor), PS4B (Oxygen Monitor) or PS8A (THC Monitor)) in other instances it is based on the inherent capabilities of the instrument and/or the sampling/analysis procedure.  In any event this is the accuracy which the device’s recorder must achieve upon calibration or during an audit.


Table 1.  Master Table

 

Continuous monitoring systems performance evaluation summary

for a typical preheater/precalciner cement production facility

 

Monitored Parameter

Device Description

Range and Units of Measurement

Accuracy of Measurement

(DQOs)

Internal Quality Assurance Activities

External Quality Assurance Activities

Location of Records

Comments

Kiln Feed Weigh Feeder

ABC Company Weigh Feeder

0-250 tons per hour

+/-  X%  of actual weight

Continuous monitoring of function.  Inspection as required.

Calibration

Alternate Fuels Office

 

Pumpable Hazardous Waste Feed

XYZ Mfg. Mass Flow Meter

0-250 pounds per minute

+/-  X%  of actual weight

Continuous monitoring of function.  Inspection as required.

Calibration

Alternate Fuels Office

 

Calciner Coal Feed Weigh Feeder

ABC Company Weigh Feeder

0 - 20 tons per hour

+/-  X%  of actual weight

Continuous monitoring of function.  Inspection as required.

Calibration

Alternate Fuels Office

 

Kiln Coal Feed Weigh Feeder

DEF Co. Rotor Weigh Feeder

0 - 15 tons per hour

+/-  X%  of actual weight

Continuous monitoring of function.  Inspection as required.

Calibration

Alternate Fuels Office

 

Tire Chip Weigh Feeder

GHI Inc. Belt Scale

0-10 tons per hour

+/-  X%  of actual weight

Continuous monitoring of function.  Inspection as required

Calibration

Alternate Fuels Office

 

4th Stage Temperature

Generic Type K thermocouple

1400-2200 °F

+/- X%  of range

Continuous monitoring of function.  Inspection as required

Calibration

Alternate Fuels Office

 

Main Baghouse Inlet Temperature

Generic Type K thermocouple

0-1100 °F

+/-  X%  of range

Continuous monitoring of function.  Inspection as required

Calibration

Alternate Fuels Office

 

By-Pass Baghouse Inlet Temperature

Generic Type K thermocouple

0-800 °F

+/-  X%  of range

Continuous monitoring of function.  Inspection as required

Calibration

Alternate Fuels Office

 

Kiln Hood Pressure

Generic DP Cell Pressure Transmitter

-1.0 to +0.25 inches of water column

+/-  X%  of range

Continuous monitoring of function.  Inspection as required

Calibration

Alternate Fuels Office

 

Kiln Rear Pressure

Generic DP Cell Pressure Transmitter

0 to –3 inches of water column

+/-  X%  of range

Continuous monitoring of function.  Inspection as required

Calibration

Alternate Fuels Office

 

Main Baghouse Differential  Pressure

Generic DP Cell Pressure Transmitter

0 to 20 inches of water column

+/-  X%  of range

Continuous monitoring of function.  Inspection as required

Calibration

Alternate Fuels Office

 

By-Pass Baghouse Differential  Pressure

Generic DP Cell Pressure Transmitter

0 to 20 inches of water column

+/-  X%  of range

Continuous monitoring of function.  Inspection as required

Calibration

Alternate Fuels Office

 

Main Stack Opacity Monitor

JKL Instruments

 Model XXX

0 – 100%

Zero/Upscale drift (24hr) - £ 2% Opacity

Calibration Error  - £ 3% Opacity

Pass 7 day operational test period

Response  time  £ 10 seconds

Data recorder resolution  £ 0.5%

 

Zero calibration 

Upscale calibration,

and for each, a 24 hour drift determination.

 

A Performance Evaluation consisting of the following tests: a calibration error,  a response time determination, 7 day operational  test period with a daily 24 hour zero/upscale calibration drift test.

Daily Calibration Drift records on monitor data terminal.

Annual Performance Evaluation records in the Alternate Fuels Office

 

Total Hydrocarbon Monitor (on By-Pass Duct)

MNO Environmental Instrument Company

 Model ZZ

0-100 ppmv as propane

Zero/Upscale drift (24hr) - ± 3 ppm 

Calibration Error - ± 5 ppm

Pass 7 day operational test period

Response  time  £ 2 minutes

Data recorder resolution  £ 0.5ppm

Permit requires accuracy of:

 ±0.5% of calibrated Range

 

Zero calibration 

Upscale calibration,

and for each, a 24 hour drift determination. 

 

 

A Performance Evaluation consisting of the following tests: a calibration error,  a response time determination, 7 day operational  test period with a daily 24 hour zero/upscale calibration drift test.

All documentation is kept in the CEM room operational record

Two monitors are used for this service.  Both are calibrated each day.

Oxygen Monitor (on the By-Pass Duct)

PQR Ind.

Model YYY

0-25%

Zero/Upscale drift (24hr) - £ 0.5%  O2 

Calibration Error - £ 0.5%  O2  

Pass 7 day operational test period

Response  time  £ 2 minutes

Data recorder resolution  £ 0.5%  O2 

Permit requires accuracy of:

 ±0.5% of calibrated Range

 

Zero calibration 

Upscale calibration,

and for each, a 24 hour drift determination.

 

 

A Performance Evaluation consisting of the following tests: a calibration error,  a response time determination, 7 day operational  test period with a daily 24 hour zero/upscale calibration drift test.

All documentation is kept in the CEM room operational record

Two monitors are used for this service.  Both are calibrated each day.

 

Trace Metal Input Control

 

 

Sample collection methods and sample analysis methods are covered in the Waste Analysis Plan (WAP).  The WAP and the laboratory SOPs detail the QA and QC requirements.   

 


Internal Quality Assurance Activities:


These activities are stated in 40CFR63.8(e)(3)(ii) to include, activities planned by routine operators and analysts to provide an assessment of CMS performance.  All of the parameter monitoring devices send a signal to the central processing unit where the values are monitored for compliance with the permit requirements.  In effect the computer monitors the function of the devices and can, and often is programmed to sound an alarm if the device fails to function, that is fails to send a signal within certain parameters.  Additionally, experienced operators would quickly notice device output readings that were inconsistent with values of other operational parameters.  An alarm or noted inconsistency would lead to inspection of the device and if required maintenance.  The table provides a summary of these activities.  The subsections in the QA section provide more detail.


External Quality Assurance Activities:


The regulation in 40CFR63.8(e)(3)(ii) states that the external QA program include: “system audits that include the opportunity for onsite evaluation by the Administrator of instrument calibration, data validation, sample logging, and documentation of quality control data and field maintenance activities”.  These activities are generally performed at specified intervals, either dictated by a performance specification (such as PS1 for Opacity Monitors, PS4B for the Oxygen monitor or PS8A for the THC monitor) by the device’s manufacturer or by good engineering practice based on site specific conditions.  For the purposes of the submittal of a CMS Evaluation Test Plan these frequencies are not specified.  The table provides a summary of these activities.  The subsections in the QA section provide more details.


Location of Records:


This is self-obvious.  These records may be paper or electronic, the regulation is not specific about this.  Certain records such as a Relative Accuracy Performance Test executed by a contraction would of course be paper files.  Internally generated, such as the one-minute values that are subsequently used to calculate hourly rolling averages will of course be in electronic format.


Comments:


If there are redundant or standby monitors or devices this should be noted here.
 

The Monitored Parameter QA Programs


The monitored parameter QA programs section of the test plan provides additional details on internal, and external QA activities and any routine maintenance activities.  See the Example QA Programs below.  These QA programs will frequently include attachments.

 

Table 2.  Example QA Program 1

Pumpable Hazardous Waste Feed

Monitored Parameter

Device

Description

Range and Units

of Measurement

Accuracy of Measurement (DQOs)

Pumpable Hazardous Waste Feed

ABC Company  Mass Flow Meter

0-250 pounds per minute

+/- X% of actual weight

 

Internal Quality Assurance Program

Coriolis mass flow meters have no moving parts.  Their operation is based on sensing the distortion of the tube versus mass flow rate.  Inaccuracies are the result of electrical malfunction or gross changes in fluid density such as in two phase flow (gas/liquid phases).  Two phase flow is precluded by the inability of the two centrifugal liquid HWF pumps to handle two phase flow.  An electrical malfunction would result in an alarm, either as a “no flow” alarm or a “high flow” alarm.  Either would result in an AWFCO followed by inspection and maintenance if required.

  

Routine Maintenance Program

No routine maintenance required.

 

External Quality Assurance Program

Periodically the accuracy of the device is checked and, if needed, calibrated.  This is done by performing a totalized mass flow versus a change in volume in the supply tank.  An operator obtains an accurate reading of the level in the supply tank at the beginning of the timed period.  At the same time, a totalized reading is taken.  After a predetermined time period, a second reading of the level in the supply tank is taken and a mass calculated using the specific gravity of the fluid in the tank.  Also, at the end of the predetermined time period, a second totalized reading is taken and a mass calculated.  The two sets of data are compared.  A significant difference in mass feed rate would result in a confirmatory test and subsequent examination and calibration of the mass flow meter.


Manufacturer’s literature is attached on the mass meter as well as a blank mass flow rate device accuracy deterermination form.  The records of this test and, if necessary, subsequent calibrations are kept in the Alternate Fuels office.

 

Table 3.  Example QA Program 2

Opacity Monitor

Monitored Parameter

Device Description

Range and Units of Measurement

Accuracy of Measurement (DQOs)

Main Stack Opacity Monitor

JKL Instruments Model XXX

0 – 100%

Zero/Upscale drift (24hr) - £ 2% Opacity

Calibration Error  - £ 3% Opacity

Pass 7 day operational test period

Response  time  £ 10 seconds

Data recorder resolution  £ 0.5%

 

Internal quality Assurance Program

The device has an automatic calibration feature provided by the manufacturer.  This allows the device to conduct its own daily zero and upscale calibration and to perform a 24 hour drift calculation.  Failure to perform the calibration or a drift in excess of 2% opacity would result in an alarm and the operator would initiate an investigation by maintenance.  Diagnosis and repair would follow.  Since the regulation allows ten seconds between samples, this allows the device to execute hourly “dirty window” and calibration corrections as well.  A “Dirty Window” exceedance warning will be issued by the device if a calibration fails this would be followed by inspection and maintenance as required.

Routine Maintenance

The opacity windows are inspected daily and are cleaned as needed.


External Quality Assurance Program

Annually, the device is subjected to inspection and calibration per the requirement of PS-1 as detailed in the Performance Specification Verification Procedure Section 7.  A copy of PS-1 is attached, including blank forms and calculations.  Briefly, the procedure consists of the following:

 

a)      There is a seven-day conditioning period to determine that the device is operating well enough to be tested.  For retesting this is unnecessary.

b)      The device is subjected to a calibration error test over three points in its range: low, midrange, and upscale.  The device must exhibit a £3% opacity difference as calculated in the method.

c)      The device is subjected to the response time test and must exhibit a response time of £10 seconds.

d)      The device must then pass the seven day operational period where no repairs are allowed to the device and the device must successfully calibrate daily and exhibit £2% opacity drift each 24 hours.

e)   The Data Recorder Resolution is documented to be £0.5% opacity.


The daily calibration and calibration drift calculation is recorded in the Data Log in the Opacity Device Terminal.  The Annual Performance Specification Verification documentation is kept in the Alternate Fuels office.  Maintenance repair records are kept in the Electrical Department, such as: manufacturer’s literature on calibration procedures, installation procedures and maintenance procedures or copies of the applicable performance specification.  These attachments may also have site specific procedures written by facility management, such as special calibration procedures or maintenance procedures, this may include special forms for calculating instrument accuracy or for reporting maintenance on the device.


The device subsection in the QA programs section of the test plan provides additional details on internal, and external QA activities and any routine maintenance activities.  These subsections will frequently include attachments such as: manufacturer’s literature on calibration procedures, installation procedures and maintenance procedures or copies of the applicable performance specification.  These attachments may also have site specific procedures written by facility management, such as special calibration procedures or maintenance procedures, this may include special forms for calculating instrument accuracy or for reporting maintenance on the device.

 

NON-CONTINUOUS MONITORING SYSTEMS


The regulation in 40CFR63.2(definitions) defines CMS to include “manual or automatic monitoring that is used for demonstrating compliance with an applicable regulation on a continuous basis.”

Since the regulation requires that trace metal feed rate limits not be exceeded, it is necessary to include trace metals analysis as proof of the CMS Evaluation Test Plan.  The trace metals analysis of the kiln feeds are entered into a data base which is used to compute trace metals feed rates based on the mass input of the various feeds.  The method and QA/QC requirements for these trace metals analyses are already documented in the facilities waste analysis plan (WAP) and it would be redundant to include the WAP with the CMS Evaluation Test Plan, it should however be referenced in the Test Plan.

CONCLUSION


This format for a CMS Evaluation Test Plan should adequately address the requirements.  The extent of the additional details deemed necessary by the agency is currently unknown.  However, this format readily allows insertion of necessary documentation.  As noted earlier the difference between the Evaluation Test Plan and the Evaluation Plan is the lack of QA activity frequency.  This can easily be added to the Internal QA and External QA columns in the master table.