United States
                 Environmental Protection
                 Agency
              Solid Waste and
              Emergency Response
              (5305)
EPA530-F-94-023
    July 1994
vvEPA
Memorandum  on
Trial  Burns
                                     Recycled/Recyclable
                                     Printed on recycled paper that contains at
                                     least 10% post-consumer recycled fiber

-------

-------
              UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                          WASHINGTON, D.C. 20460
                            JUL -5 1994
     OFFICE OF
SOLID WASTE AND EMERGENCY
     RESPONSE
MEMORANDUM

SUBJECT:  Guidance on Trial Burn Failures

FROM:     Michael  Shapiro,  Director/^'
          Office of Solid Waste

TO:       Hazardous Waste Management Division Directors
          Regions  I-X
     The purpose  of  this memorandum is to clarify EPA's policy on
trial burns  for incinerators and boilers and industrial furnaces
(BIFs) under the  Resource Conservation and Recovery Act  (RCRA),  ;
and to address issues  that have recently been raised regarding
trial burn failures.   These issues include: 1) what constitutes a
successful trial  burn;  2)  how to handle invalid data; 3) what
constitutes  an unsuccessful trial burn; 4) how to handle a
request for  a trial  burn retest; and 5) how to restrict
operations after  an  unsuccessful trial burn.

     The policies set  out in this memorandum are not final agency
action, but  are intended solely as guidance.  They are not
intended, nor can they be relied upon, to create any rights
enforceable  by any party in litigation with the United States.
EPA officials may decide to follow .the guidance provided in this
memorandum,  or to act  at variance with the guidance, based on  an
analysis of  specific site circumstances.   The Agency also
reserves the right to  change this guidance at any time without
public notice.


Purpose of a Trial Burn

     A trial burn serves several purposes.  It is used to
determine whether a  facility can meet the required performance
standards for either hazardous waste incinerators (40 CFR
264.343) or  BIFs  (40 CFR Part 266 Subpart H),  and to determine
the operating conditions that should be. set in the permit.  A
trial burn is also used by the permit writer to determine the
need for and establish other limits or requirements on a site-
specific basis under the "omnibus" authority of RCRA Section
3005(c)(3).  This guidance will consider the term "performance
                                                      R«cycl*d/R0cyclabl«
                                                      Printed wtth Soy/dnol* Ink on piper that
                                                      contain* MlMtt SO* recycled fiber

-------
standards" to include both regulatory performance standards and
such site-specific standards imposed through the omnibus
authority.  Until continuous emission monitors  (CEMs) are
available, setting permit operating conditions based on the
results of trial burns is the best method of assuring compliance
with the regulations.

     A trial burn typically consists of a. series of "tests".  A
trial burn test  (or combination of tests) should be done for each
set of operating conditions for which the facility desires to be
permitted.  Three "runs" should be performed for each test.  Each
run of a test should be conducted at the same nominal operating
conditions.  In general, each run of a test should be passed for
the test to be considered successful and for the facility to be
permitted to operate at that set of conditions.

     Facilities will often perform multiple tests during the
trial burn in order to develop all applicable permit operating
conditions.  For example, facilities will usually perform a
minimum and a maximum temperature test, since decreasing
temperatures tend to decrease organics destruction, and
increasing temperatures tend to increase metals emissions due to
an increase in volatility.  These tests, if successful,  will
determine the temperature boundaries between which the facility
can operate in compliance with the destruction and removal
efficiency (DRE)  and metal emissions standards.

     During a trial burn, a facility's general strategy is
typically to operate at conditions that will give it a broad
range of permit operating conditions.  The permit writer should
take great care in reviewing the trial burn plan to assure that
the test conditions meet the regulatory requirements.   According
to 40 CFR 270.62(b)(5)  for incinerators and 40 CFR 270.66(d)(2)
for BIFs, the trial burn plan can only be approved if 1)  it is
likely to determine if the performance standards can be met,  2)
it does not present an imminent hazard to human health or the
environment,  and 3)  it will help to determine the necessary
operating requirements.  In determining if the performance
standards can be met in the trial burn, permit writers should use
their experience and best engineering judgement to make  sure that
the trial burn represents "good operating practices".   EPA
believes that a trial burn plan that allows or incorporates sub-
standard operating practices is less likely to demonstrate
compliance with required performance standards than a plan based
on a well-operated unit.  The Combustion Emissions Technical
Resource Document (CETRED),  which helps to define best operating
practices for various categories of hazardous waste combustors,
can assist in determining good operating practices.   Engineering
judgement and generally accepted industry' practices for achieving
good mixing,  adequate temperatures and residence times,  adequate
oxygen,  steady-state operation,  and minimization of fugitive
emissions can also be used in this- evaluation.   Additionally,  in

-------
 reviewing and approving  a  trial burn plan,  the permit  writer may
 find  it useful  to  examine  the  facility's  compliance  history and
 past  operating  history when  applicable.


 What  Constitutes a Successful  Trial  Burn

      A trial burn  is  successful only if enough tests are passed
 so that the permit writer  can  establish a complete set of
 operating conditions  in  the  permit to assure compliance with
 applicable performance standards.  A successful trial  burn  test
 generally consists of passing  three  separate runs at the same
 nominal operating  conditions.  If a  test  is successful, the
 facility would  be  allowed  to operate under the tested  conditions.
 In general, failing any  performance  standard in any  one of  the
 three runs constitutes a failure of  that  test.  If a test fails,
 the facility should not  be permitted to operate under  the failed
 conditions.

      A facility may fail an  individual test (or several tests)  at
 particular operating  conditions during the trial burn; however,
 if sufficient tests are  passed such  that  applicable  permit
 operating conditions  can be  established from the successful
 tests, then the trial burn is still  considered successful.   For
 example, for a  facility  where maximum and minimum temperature
 limits are necessary, the  facility would  typically have to pass
 both  a minimum  temperature test and  a maximum  temperature test,
 along with any  other necessary tests, for the  trial burn to be
 successful.

      Facilities can receive  final permit  conditions,for only
 those conditions that they passed in the  trial burn or that'are
 set independent of the trial burn (e.g.,   Tier  I metal limits/
 which are discussed later  in this document).  Thus,  in a case  .
 where a facility passed  some tests and failed others, it is
 important to be able to distinguish  the difference between the  '
 successful and unsuccessful conditions.    Final permit conditions
 should be written  to allow the facility to operate at the
 successful conditions while excluding the unsuccessful ones.
Additionally,  the permit writer should be sure to set monitoring
 and recording requirements in the permit  to assure that operating
 conditions are being met.

      Final permit  conditions will directly reflect the successful
 operating conditions from the trial burn.   Due to unforeseen
 circumstances that may arise during trial  burns, the  trial burn
 conditions may deviate somewhat from the conditions  specified in
 the trial burn plan.   If this situation occurs, and  the trial
burn was successful,  the operating conditions  in the  permit
 should be the conditions demonstrated during the trial burn, not
 the conditions from the trial burn plan.   In other words,  for
 conditions that are set based on the trial burn, a facility will

-------
 be permitted to operate only at those conditions that have been
 demonstrated successfully during the trial burn.

      Facilities may perform'several tests during a trial burn in
 an attempt to have different sets of operating conditions for
 different sets of wastes (i.e.,  "campaign burning").   If a
 facility fails a particular test,  it may still be permitted to
 operate on those waste streams and at those conditions that were
 successfully demonstrated,  provided that sufficient data are
 available from the passed tests  to set all necessary  permit
 operating conditions.   If trial  burn results do not provide
 Suf icient data to enable the Agency to set permit conditions
 which assure compliance with the performance standards,  then the
 trial burn would not be considered successful.
How to Handle  Invalid Dafca                .

     In limited situations,  the Agency believes  it may be
appropriate to use data from two successful runs as the basis to
determine that a trial burn  test was successful when
circumstances  beyond the owner/operator's control caused the
invalidation of a
-------
      d)
     e)
There should be no reason to believe  (based on
operating data, observation of stack  emissions, etc  )
that the invalid run was less likely  to be in
compliance than the other two runs.   Immediate
reporting by the facility of an incident which might
invalidate a run  (e.g., QA/QC outside of control
limits) lends more credence to the claim of invalidity
than_if the facility waits until all  analytical results
are in and emission calculations have been made.

A detailed written description of the circumstances
resulting in the invalidation of data related to any
test should be submitted to, and reviewed by, the
Agency.
     Generally,  two valid  runs  should not be accepted as a
successful trial burn  test when the  owner/operator had direct
control over the situation that caused  the  third  run to be
invalidated.  The trial burn  test should be considered
unsuccessful if neglect and/or  carelessness of either the
owner/operator or those conducting the  testing/analysis caused
the invalidation of a  run.
What Constitutes an Unsuccessful Trial Burn

     A trial burn is unsuccessful either because it showed a
failure to. meet the performance standards, or it was
inconclusive.  A trial burn is considered a failure when enough
tests have; failed (i.e., show a failure to meet performance
standards) such that a .full set of operating conditions
representing compliance cannot be set in the permit.

     A trial burn failure is different from failure of a trial
burn test.  A test failure shows nonconformance with the
standards at one set of operating conditions; however, a facility
may still be permitted to operate if it passes one or more trial
burn tests at other operating conditions.  A trial burn failure
occurs when enough tests have failed such that a full set of
operating conditions representing compliance cannot be set in the
permit.  The results of a failed trial burn should not be used to
establish final permit operating conditions.  Following a failed
trial burn,  the permitting authority should take one or more of
the following actions, as appropriate: 1) take steps to restrict
operations (as discussed later in this document); 2)  begin
processing a denial of the facility's permit application (for an *
interim status facility); 3)  initiate proceedings to terminate
the facility's permit (for a new facility);  4)  authorize a trial
burn retest -(also discussed later in this document).

     An entire trial burn (like a trial burn test)  may be
considered inconclusive.  An inconclusive trial burn occurs when

-------
 data  problems  have arisen such that neither conformance nor
 nonconformance with the performance standards can be shown.  The
 results  of  an  inconclusive trial burn may not be used to
 establish final permit  operating conditions.   Following an
 inconclusive trial burn,  the permitting authority should take one
 or more  of  the following actions,  as appropriate:  1)  take steps
 to restrict operations  (as discussed later in this document);  2)
 begin processing a denial of the facility's permit application
 (for  an  interim status  facility);  3)  initiate proceedings to
 terminate the  facility's permit (for a new facility);  4)
 authorize a trial burn  retest (also discussed later in this
 document).

      Facilities may choose not to  test for certain parameters  and
 be permitted at the Tier I or Adjusted Tier I feed rate screening
 limits established in the BIF rule (56 FR 7134,  February 21,
 1991), if appropriate.   These parameters  include metal emissions
 (40 CFR  266.106),  and hydrogen chloride (HC1)  and  chlorine gas
 (C12)  emissions  (40 CFR 266.107).  The Tier I and Adjusted Tier I
 feed  rate screening limits are based on the assumption that all
 metals,  HC1, or C12  (depending on the parameter)  fed into the
 system are  emitted (i.e.,  no partitioning into the bottom ash,
 and no removal  by any air pollution control device).   This  case
 is the most conservative  scenario  possible  and produces  the most:
 stringent feed  limits in  the permit.   The Adjusted Tier  I  feed
 rate  screening  limits also allow for site-specific dispersion
 modeling.  Although directly applicable only  to  BIFs,  these
 provisions are  generally  applied to incinerators as well  through
 the Agency's omnibus permitting authority,  where necessary  to
 protect  human health and  the environment.

      Facilities  that test  for these parameters and fail, or show
 inconclusive results, should not be permitted to operate under
 the tested  conditions.  Instead, a  permit for the  facility  (if
 one is issued)   should limit  the  facility  to the Tier I or
 Adjusted Tier I  feed rate  screening limits.,   For example, a
 permit for a facility that does not meet  the  HC1 or Cl, standard
 when  tested under higher  chlorine  feed  rates  should limit the
 chlorine  and chloride input  to  the  equivalent of 4 Ibs HCl/hr,
 the Tier I limit, or the Adjusted Tier  I  limit, as applicable.

      Similarly, a permit  for a  facility thstt does not meet the
metals emissions standards during high temperature testing should
 limit the metals input into  the system to the Tier I or Adjusted
 Tier  I feed rate screening limits  (see 56 FR  7171, February 21,
 1991).                                                            .

      It  should also be noted  that, where the trial burn did not
 demonstrate compliance with  the HC1, C12,  or metal  emissions
 standards, the permit may specify allowable chlorine or metals
 feed  rates that are more restrictive than the Tier I or Adjusted
Tier  I limits,  based on a site-specific risk assessment which

-------
considers both direct and indirect exposure pathways to a wide
range of pollutants.• In this case, the same assumption
concerning stack emissions should be applied  (that is, the
assumption of no partitioning,or removal).
How to Handle a Request For a Trial Burn Retest

     Facilities that fail or conduct an inconclusive trial burn
test or tests may request a retest and submit a revised trial
burn plan.  The permitting authority would review and approve or
deny such a request.  For a permitted incinerator or BIF  (new or
renewal), this request would be processed through the permit
modification procedures in accordance with 40 CFR 270.42.  The
revised trial burn plan can only be approved if 1) it is likely
to determine if the performance standards can be met, 2) it does
not present an imminent hazard to human health or the
environment, and 3) it will help to determine the necessary
operating requirements  (see 40 CFR 270.62(b)(5) for incinerators
and 40 CFR 270.66(d)(2) for BIFs).  In the case of a request for
a trial burn retest following a trial burn test failure, the'
applicant should conduct an investigation into the reason for the
failure, and make substantive changes in its proposed trial burn
plan which would be expected to prevent failure from reoccurring..
A facility should not be allowed to retest unless it has made
changes to its process  (i.e., design and/or operating
conditions), that are likely to correct the problems encountered
in the failed trial burn test.  A facility should not be allowed
just to "take its chances" on passing a retest under the same
conditions.  The first failed test indicates that, at best, the
unit would not be in compliance some of the time when operated at
those conditions, and that those conditions should therefore not
be incorporated into a permit.

     As opposed to a trial burn test failure, an inconclusive
test would not necessarily require changes to be made to the
process prior to allowing a retest.  The test could be repeated,
under the same conditions as the previous test, but with special
attention paid to the situation that caused the original test to
be inconclusive.  During the retest, all attempts should be made
to prevent that situation from reoccurring.

     There is no set limit on the number of retests allowed under
EPA regulations, so long as after each unsuccessful test the
above criteria are met and the trial burn plan is revised and
approved  (through a permit modification for a new incinerator or
BIF) prior to any retesting.  The same criteria recommended for
the design and conduct of initial trial burns are also
recommended for all retests  (i.e., three runs for each trial burn
test, etc.).

-------
      Facilities that wish to conduct a trial burn retest after an
 unsuccessful test should expeditiously submit a comprehensive
 request consistent with the guidance discussed above.  If a
 complete request is not promptly submitted,  it is appropriate for
 the_Agency to_start permit denial proceedings.  The Agency's
 decision to discontinue or delay permit denial proceedings will
 be highly dependent on the adequacy of any retest request and the
 Agency's ability assure compliance with applicable regulations
 during the interim period.

      For facilities that fail a trial burn test for only the HC1,
 C12/ particulate, or metal emissions  standards, EPA believes  it
 may be appropriate in some cases to authorize a retest for these
 failed performance standards without simultaneous ORE testing.
 This  decision would depend on the nature of  the design or
 operating modifications made for the retest.   If the
 modifications would not adversely impact DRE (e.g.,  addition of
 pollution control equipment),  then HC1,  particulate,  and/or metal
 tests  are sufficient.   In this case,  operating conditions should
 be identical to those of the original trial  burn test for all
 parameters other than those  related to the modifications which
 were made.   In contrast,  if  the design or operating  modifications
 madefy the facility in order to retest for  the HC1,  C12,
 particulate,  or metals emissions standards have the  potential to
 affect DRE,  then DRE should  be retested along with the standards'
 that were not demonstrated.

     The permit writer should ensure  that operating  conditions
 during a trial  burn retest are consistent with the overall  scheme
 of the trial burn plan so that all  successful  tests  can be  used
 in conjunction  to establish  final  operating  conditions.
How to Restrict Operations After an Unsuccessful Trial Burn

     Permitting authorities should move expeditiously, in
appropriate cases, to restrict operations  (to the extent that
regulatory and statutory authorities allow) after receiving
information that a facility conducted an unsuccessful trial burn
(i.e., a trial burn failure or an inconclusive trial burn).

     Permits for new incinerators and BIFs should be written with
a provision that would restrict post-trial burn operations if a
facility conducts an unsuccessful trial burn.  The Agency
recommends that such permits contain the following conditions- l)
the permittee must notify the Regional Administrator within 24
hours of making a determination that the incinerator or BIF
failed to achieve any of the performance standards in any run of
any test, and 2) upon the request of the Regional Administrator,
the permittee shall feed waste and operate the incinerator or BIF
only under restricted conditions as specified by the Regional
Administrator.  (A similar condition is recommended in the

                                8

-------
 incinerator module  of  the  model  permit,  except the second portion
 of  the  condition provides  that,  upon the request of the Regional
 Administrator,  the  permittee shall  cease feeding hazardous waste
 to  the  incinerator.  The new recommended language covers the  case
 where a complete shutdown  is required, while  providing clearer
 authority  in  cases  where some, but  not all, tests were
 successful.)  The permittee  then has the option of applying for a
 permit  modification pursuant to  40  CFR 270.42 to conduct a new
 trial burn pursuant to 40  CFR 270.62(b)  for incinerators or 40
 CFR 270.66 for  BlFs.   If an  already-issued permit does not have
 such a  provision in it, and  the  trial burn is unsuccessful, then
 EPA may still be able  to modify  the permit to restrict operations
 based on 40 CFR 270.41(a)(2)  or  40  CFR 270.41(b)(1), or terminate
 the permit based on 40 CFR 270.43(a)(3).  The appropriate
 authorities should  be  invoked to assure  that  operations during
 the post-trial  burn period will  achieve  compliance with the
 performance standards.

     With  respect to interim status; BIFs, EPA regulations
 establish  certain performance standards  that  must be met  at all
 times when there  is hazardous waste in, the unit  (40 CER
 266.103(c)(l)).  Standards for carbon monoxide,  total
 hydrocarbons, particulate matter, metals  emissions, and hydrogen
 chloride and chlorine  gas emissions are  included in the "
 regulations.  If trial burn  data from an  interim status BIF
 indicate failure to comply with  any of these  standards, then
 under -appropriate circumstances  the permitting agency may be able
 to  restrict operations under RCRA Section 3008 or Section 7003.

     With  respect to interim status  incinerators  that fail their
 trial burns, regulatory agencies should either move as quickly as
 possible to cause the  incinerators  to cease operations by denying
 their permits (or, if  appropriate,   through RCRA Section 7003
 actions),  or, if appropriate, authorize trial burn retests.  This
 guidance also applies  to interim status BIFs  that  fail their DRE
 standard during the trial burn,   since the DRE standard generally
 does not apply to BIFs during interim status.

     EPA has recently proposed a rule which would provide
 explicit authority to  restrict operations at interim status
 facilities after a failed or  inconclusive trial burn (59 FR
 28680, June 2, 1994).   During the post-trial burn period, interim
 status facilities would only be able to operate under conditions
 that passed and were demonstrated to meet the applicable
performance standards,  and only if the successful trial burn data
are sufficient to set all applicable operating conditions.  If
 finalized as proposed,  this regulation would provide additional
authority to restrict operations at  interim status facilities
 following a failed or inconclusive trial  burn.

-------
     For more background on issues such as permit conditions,
trial burn measurements, and validity of data, permit writers may
consult the following guidance documents.

     Guidance on Setting Permit Conditions and Reporting Trial
     Burn Results; January 1989.

     Hazardous Waste Incineration Measurement Guidance Manual;
     June 1989.

     Quality Assurance/Quality Control (QA/QC) Procedures for
     Hazardous Waste Incineration, January 1990.

     If your staff have any questions on this trial burn failure
guidance or how to obtain other guidance materials, they may call
Andy O'Palko at (703) 308-8646, or Sonya Sasseville at (703) 308-
8648.


cc:  Waste Combustion Permit Writers Workgroup
     Dev Barnes
     Matt Hale
     Matt Straus
     Fred Chanania
     Susan Bromm
     Susan O'Keefe
     Office of Regional Council RCRA Branch Chiefs,  Regions I-X
     Brian Grant,  OGC
                               10

-------