may not be appropriate. It is not necessarily adequate to simply incorporate a novel feature or design concept into the test sample to verify its acceptability. In some cases, larger scale testing will be required to support development of special conditions. In other cases, the test burner might be acceptable, but the test stand might require modification (for example, substitution of aluminum frames for the steel frames) in order to produce valid results. This paragraph provides other acceptable means for showing compliance. The discussion of the test fixture modifications and burner orientations is very specific and will provide acceptable results. However, there may be other methods of achieving the same objective, and the discussion below is not meant to imply that only the modifications shown are acceptable.
a. Other Material Types. As previously noted, this AC assumes one of three methods of providing a fire barrier with respect to substantiation of installation details. Other methods, such as foam blocks or spray-on applications, have not been investigated to the same extent; and reliable substantiation methods for installation have not been developed. Conduct realistic testing on these types of materials to establish guidance for their installation. The general principles for avoiding discontinuities and penetrations are expected to be valid, regardless of the type of insulation employed. Some specifics, for example the amount of overlap, are likely to be different.
b. Other Means of Attachment. Means of attachment that vary significantly from those described in this AC will require substantiation with more representative installation fixturing.
For example, a hook and loop type attachment would require substantiation by test, but could probably be accomplished using the test burner, with appropriate modification to the frames and stringers.
c. Modification of the Test Fixture.
(1) If the test fixture needs to be modified in order to address material and/or installation schemes not anticipated by the rule, the existing vertical steel frame is replaced with an aluminum frame. Similarly, two of the steel horizontal stringers are replaced with aluminum stringers (see figure 12). This methodology allows the aluminum members to melt and fail with the realism of an actual aircraft fuselage during a post-crash fire scenario. Under these conditions, not only are the blanket materials being tested, but the ability of the insulation system for preventing flame penetration is examined. Such a test also assesses details of the system used to attach the insulation to the frame, including clips, tape, hook and loop, etc. Since there are numerous combinations of frame geometry, material thickness etc., the applicant should propose a critical case for substantiation of use on the airplane, if this method of testing is necessary.
These elements are replaced with aluminum elements Figure 12. Modified Test Fixture Incorporating Aluminum Components 14
Figure 13. Apparatus Configuration for Testing Overlap
(2) To evaluate an overlap arrangement using less than 6 inches of overlap, both the test stand and its relationship to the burner need to be changed. Figure 13 illustrates the arrangement, which involves moving the burner (or stand) so that the burner flame impinges directly between two frames. In addition, the third stringer from the bottom (i.e., in line with the center line of the burner) is removed and the exposed seam of the overlap is positioned at this point (figure 14). For this configuration, only physical burnthrough is assessed (no heat flux measurement is required). This is because the geometry no longer represents the standard, and the heat flux measured on the back side would not be comparable to the standard. In addition, the basic material will be qualified in the standard configuration, and that will include assessment of the back side heat flux. It is acceptable to either leave the other frame bay empty, or install a complying material in the standard manner.
Figure 14. Test Specimen Configuration for Testing Overlap 16
Figure 15. Modified Apparatus for Testing Overlap
9. INSTALLATIONS THAT DO NOT REQUIRE COMPLIANCE WITH § 25.856(b). As noted in paragraph 5e of this AC, the FAA may determine that certain installations will not contribute to fire penetration resistance, even if they complied with § 25.856(b). The following installations have been assessed and determined to fall into that category because of inherent characteristics of the installation, the location, or both.
a. Lower lobe cargo doors: Lower lobe cargo doors leading into class C cargo compartments, and having a complete (recognizing that there will be cutouts for the hinges and possibly the operating handle) liner on the door meeting the requirements of the ‘ceiling’ portion of Appendix F, part III, do not require modification to the insulation inside the door.
b. Passenger doors: If less than 12” of the door is in the lower half of the fuselage, the insulation on the door does not need to comply with § 25.856(b). If 12” or more of the door is in 17
the lower half, and insulation is held in place mechanically, the insulation material should meet the requirements of § 25.856(b), but the attachment method does not need to be tested. If the insulation is not held in place mechanically, a test of the actual attachment configuration is required. However, actual door structure should not be necessary, as long as the attachment method is represented in a test such as discussed in paragraph 8.
c. Wing box: The wing box itself does not require improved insulation (assuming it is insulated). Note that the insulation installed on the outer skin in the fuselage above the wing box does require improved burnthrough protection for the portion that is in the lower half of the fuselage. See figure 16.
Wing box insulation does not require compliance.
Insulation on skin above wing does require compliance.
Figure 16. Wingbox Area 18
d. Window line: Some allowance may be possible if the half-way point (between the upper and lower half of the fuselage) intersects the passenger windows. That is, adding insulation between closely spaced windows will not contribute to burnthrough protection in some cases.
Because each installation is different, any proposals that involves noncompliant insulation between windows that are in the lower half of the fuselage should be coordinated with the FAA.
See figure 17.
Insulation Half-way line Windows Lower half Upper half Figure 17. Window Line
e. Flightdeck. The flightdeck floor is often very close to the half-way point. Because of the changing geometry at the nose of the airplane, the half-way point can transition from above the flightdeck floor to below the flightdeck floor. In some cases, it may be acceptable to limit the burnthrough protection to the flightdeck floor. However, this should be coordinated with the FAA.
10. TEST CONDITION DETAILS. 14 CFR 25, Appendi x F, part VII, specifies a “modified, gun-type” burner such as a Park Model DPL3400 to obtain consistent test results. The FAA has also developed an alternative burner that does not rely on a motor driven fan and fuel pump.
This burner is an acceptable “modified gun-type” and is discussed in more detail in Appendix 2 of this AC.
a. Research has shown that laboratory test conditions can have an influence on test results.
In particular, room temperature can affect the calibration, which will, in turn, influence the test results in certain cases. Maintaining consistent environmental conditions, especially between calibration and testing, improves the consistency and reliability of the test results.
b. In addition to the calibration procedures described in Appendix F, part VII, it is useful to periodically “map” the heat flux of the burner over a larger area than is typically encompassed by the calibration measurement. Since each burner will have its own signature heat flux map, the important consideration is consistency; that is, a given burner should maintain approximately the same heat flux map over time.
Figure 18. Heat Flux Mapping Fixture 20
Lab F, New Mapping Procedure w/Intake Duct, 2150 Ft/min S3 S2 S1 Figure 19. Example of Heat Flux Map 1 2 3 4 5 6 7
16.83-16.92
16.75-16.83
16.66-16.75
16.58-16.66
16.49-16.58
16.41-16.49
16.32-16.41
16.24-16.32
16.15-16.24
16.07-16.15
15.98-16.07
15.90-15.98
15.81-15.90
15.73-15.81
15.64-15.73
15.56-15.64 21
c. The insulation film on each test blanket should have two small slits cut on the back side (away from the flame) to allow combustion gases to escape. This prevents the test blankets from “ballooning,” which can alter test results. It is recommended that the 2-inch slits be cut into each blanket, far enough away from the center vertical frame so as not to influence test results. The ballooning phenomenon is not an issue in an actual airplane, because the airplane contains continuous structures and heat transfer mechanisms not present in the test fixture. Therefore, use of slits in the actual installation is not required.
d. Fuel Nozzle. Appendix F, part VII, states that a 80°PL (hollow cone) fuel nozzle manufactured by Monarch delivers a proper spray pattern, when used in a Park Model DPL3400 burner. Due to manufacturing changes, this specific nozzle may not be available. In fact, the nozzles used in the Park Model DPL3400 burner evolved over the years. Since the nozzle is a very important element in achieving proper performance, the details of the nozzle design will have to be examined, to confirm that the correct configuration is used. The basic configurations are all similar and consist of a nozzle body and two internal components. Inside the threaded nozzle-body there is a slightly concave swirl disc, which is held tightly against the underside of the concave-tipped nozzle-body, using a threaded backing plug. Only a nozzle with a “slotted” back is recommended. Unless the nozzle contains a "slotted" backing plug, it may not perform acceptably. The original, slotted-backing-plug nozzles were designated as “F-80.” Later versions of that nozzle did not contain this designation but can be identified by a hexagonal backing plug. Experience has shown that a 6.5 gallons per hour (gph) 80 degree, PL nozzle (with a slotted backing plug), with the fuel pressure adjusted to produce the required 6 gph (approximately 85 pounds per square inch), will produce satisfactory results. The actual flow rate should be verified any time the nozzle is changed, even if the nozzles are the same model.
Note that although the nozzle is intended to have a symmetrical, hollow-cone spray pattern, the spray pattern may vary from unit to unit. Therefore, the rotational position of the nozzle, regardless of the stator position, can influence the heat flux measurement and potentially influence the test results. To facilitate calibration it is useful to document the optimum nozzle position.
e. Burner casting. There are two main types of burner castings commonly in use. These are known as ‘socket’ and ‘flange,’ descriptors for the way the draft tube fits into the housing.
Experience has shown that the socket type burners tend to produce somewhat higher velocity exit airflow for the same calibration settings. This can result in conservative test results.
f. Airflow through the burner is of critical importance. Ideally, air should only enter and leave the burner through the air inlet and outlet respectively. The burner housing and any other potential sources of air leaks should be sealed. In addition to measuring the airflow into the burner, it may also be useful to measure the airflow out of the burner. This helps confirm consistency of performance and accuracy of the measurements. Variations in airflow can greatly influence the test results, with all other parameters being equal. For this reason, regularly calibrate the air velocity meter. As with heat flux, it may be useful to periodically map the airflow out of the burner cone to monitor consistency of performance.
g. Stators.
(1) The burner should have a stator at the exit of the draft tube. A Monarch F-124 stator is recommended to produce the most consistent results.
(2) Appendix F, part VII, calls for an internal stator inside the draft tube. A Monarch H215 stator is recommended to produce the most consistent results. However, there are slight variations in the casting for that stator that may make it necessary to modify the stator in order to achieve calibration. Such modifications consist of surface treatment to the stator vanes and have the effect of altering the airflow so that the heat flux measurement can be achieved at the specified location. These modifications do not change the intensity of the burner flame, but are a calibration aide.
h. Igniters. The length of the igniters is not specified in Appendix F. Experience has shown, however, that the igniter’s overall length should be as shown in figure 18.
Figure 20. Igniter Geometry
i. Test specimen mounting frame. The center vertical frame can become distorted from
repeated exposure to the test burner. Deviations from true of more than +/- 0.25” should be corrected to avoid affecting test results. Note that the gauge of the center vertical frame is heavier than the frames on the sides.
j. Fuel and air temperature. The fuel and air temperature controls discussed in Appendix 2, paragraph 7 may also prove valuable when calibrating and testing with the standard burner.
Signed by Ali Bahrami Ali Bahrami Manager, Transport Airplane Directorate Airplane Certification Service 23
Appendix 1
Appendix 1 Acceptable Installation Approaches
1. Batting Systems Figures 1-1 through 1-3
2. Barrier Systems Figures 1-4 through 1-8
3. Encapsulating Systems Figures 1-9 through 1-10 Figure 1-1. Conventional Replacement Batting System A1-1