1 15 1 https://www.nal.usda.gov/exhibits/speccoll/files/original/75817586d378a65a30c088018f1d1e61.pdf f10d5688d35900b5266cb9c5935e91cc PDF Text Text Item ID Number 00376 Author Harrington, John J. Defense Technology Laboratories, FMC Corporation ROpOPt/APtiClO TitlO Spray Tank Unit, Aircraft, PAU-8/A Journal/Book Title Year 1™ Month/Day A ril Color IJ Number of Images 145 DOSCrlpton Notes Avn P ' ' *-• Young had this item filed under the category "Equipment - How Developed, How Used"; contract no. F08635-68-C-0090, task no. 07, work unit no. 00 Monday, January 29, 2001 Page 376 of 382 �Harrington, J. J., 1971 ED/UNLIMITED Spray^Tank Unit Aircraft, PAU-8/A, AD 892*614 v " "' * ^ x?¥ ." -. Technical Report distributed by Defense Technical Information Center DEFENSE LOGISTICS AGENCY Cameron Station .Alexandria, Virginia 22314 ^ROMEDICAL LIBRAE JAN 11 1980 UNCLASSIFIED/UNLIMITED �THIS REPORT HAS BEEN DELIMITED AND CLEARED FOR PUBLIC RELEASE UNDER DOD DIRECTIVE 5200,20 AND NO RESTRICTIONS ARE IMPOSED UPON ITS USE AND DISCLOSURE, DISTRIBUTION STATEMENT A APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED, �I o f\ = : ^ I I.I n- I2'5 2.2 ^ 2 0 - 1.8 11.25 11.4 11.6 MICROCOPY RESOLUTION TEST CHART �AFATL-TR-71-46 SPRAY TANK UNIT, AIRCRAFT, PAU-8/A DEFENSE TECHNOLOGY LABORATORIES FMC CORPORATION TECHNICAL REPORT AFATL-TR-71-46 APRIL 1971 Distribution limited to U. S. Government agencies only; kOtNMif^^mM^AtM^faJuuuiuaa^ distribution limitation applied April 1971. Other requests for this document must be referred to the Air Force Armament Laboratory (DLIF) , Eg1in Air Force Base, Florida 32542. AIR FORCE ARMAMENT LABORATORY AIR FORCt SYSTIMS COMMAND • UNITID STATCS AIR FORCI IGLIN AIR FORCE BASE, FLORIDA �Spray Tank Unit, Aircraft, PAU-8/A John J. Harrington Distribution limited to U. S. Government agencies only; M-hfaw«*BiMUiitt^iUMMM*>JRfp4wMW4 distribution limitation applied April 1971. Other requests for this document must be referred to the Air Force Armament Laboratory (DLIF), Eglin Air Force Base, Florida 32542. �FOREWORD This report documents work performed by the Defense Technology Laboratories (DTL) of the FMC Corporation, San Jose, California, under Contract F08635-68-C-0090, with the Air Force Arnament Laboratory, Eqlin Air Force Base, Florida. Program monitors for the Armament Laboratory were Mr. Michael E. Flynn and Captain Harold L. Hebert (DLIF) . This design, development, fabrication and testinq of the Spray Tank Unit, Aircraft, PAU-8/A, was conducted from 17 May 1968 through 28 February 1971 by DTL under the direction of Mr. A. H. Bussey, Proqram Manager and Mr. John J. Harrington, Project Enqineer. Technical personnel assianed to the program vere Messrs. L. R. Ramsauer, F. A. Hettinger, D. N. Singletary, ?.. ".:. "riebel, D. E. Knutsen, G. A. Powell and A. R. Janes. This report has been reviewed and is approved. . ^ ___. Franklin* "cl. DavitesT'colonel, USAF Chief, Flame, Incendiary and Explosives Division 11 �ABSTRACT A modular spray system for anticrop chemicals was designed, developed, fabricated and tested. The system is capable of external carriage on high and low performance aircraft in four possible configurations using either one, two, three, or four modules. Each of the 50-gallon modules is completely interchangeable and can spray at rates from 15 to 150 gallons per minute. The modules use a compressed-air/gas reservoir to pressurize the agent reservoir and force the agent out the nczzle. Support equipment,designed and delivered with the dispenser, included the loading and handling adapter kit for the MJ-1 and MHU-83E bomb lift trucks, the checkout unit,and the anticontamination kit for use with the F-4 aircraft. Nozzle tests were conducted from aircraft at 198 to 504 knots. Droplet sizes of 105 to 555 micron mmd were obtained with the single nodule configuration at air speeds of 214 to 354 knots. Full scale flow model tests of the agent tank lead to the development of a module which expels 99 percent of the agent from the module at a flow rate of 150 gallons per minute. Scale wind tunnel and jettison flight tests were conducted to support the desicn of a stable two-module configuration. rg*r * JEV»t Distribution limited to U. S. Government agencies only; ijuiuiL, faJJil'LaLlun and distribution limitation aoclied April 1971. Other requests for this document must be referre to the Air Force Armament Laboratory (DLIF) , Eglin Air Fcr'cs Base, Florida 32542. iii (The reverse of this page is blank) �FRSCjSDINO PAGE BUNK-NOT FILMED TABLE OF CONTENTS Section I II Title Page 1 2 2 2 INTRODUCTION SYSTEM DESCRIPTION 2.1 2.2 2.3 Physical Data and 6 2.4 III Detailed Description of 6 23 MODULE DEVELOPMENT 3.1 IV V 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 MODULE ADAPTER DEVELOPMENT 4.1 4.2 4.3 4.4 4.5 4.6 4.7 Testing and Modification DISPENSER TEST UNIT DEVELOPMENT 5.1 5.2 5.3 .. . . , , , 23 25 30 39 41 52 57 57 59 65 65 65 66 70 71 71 72 73 73 73 73 �TABLE OF CONTENTS (CONCLUDED) Section Title VI LOADING AND HANDLING ADAPTER DEVELOPMENT 6.1 Requirements Page 78 78 6.2 78 6.3 VII Design Objectives Design 78 Design Objectives Design 86 86 CONTAMINATION HARDWARE DEVELOPMENT 89 &. 1 8.2 IX 86 86 7.2 7.3 VIII SHIPPING CONTAINER DEVELOPMENT 7.1 Requirements 89 89 Requirements Design TESTING . 91 9.1 Two-Module Wind Tunnel Tests 9.2 Two-Module Captive Flight and Jettison Test 9.3 9.4 X Aircraft Physical Compatibility Tests Spray Droplet Size and Dispenser Airworthinesss Test 91 100 105 105 112 10.1 10.2 XI MAINTAINABILITY AND RELIABILITY 112 117 Maintainability Reliability SUMMARY 126 vi �LIST OF FIGURES Figures 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 ,23 24 25 ,26 27 Title Aircraft Spray Tank Unit, PAU-8/A Aircraft Spray Tank Unit, PAU-8/A (Four Configurations) Electrical Test Unit Loading and Handling Adapter Assembly (Modification Kit for MJ-1 and MHU-83/E Bomb Lift Trucks) Turning Vane Kit „ Tank Assembly Agent Transfer System . . . . .'. PAU-8/A System Schematic — Agent Dispensing System Shipping Container Nitrogen Storage and Control System Test Nozzle No. 1 GFE Nozzle , Test Nozzle No. 2 Test Nozzle No. 3 Prototype Nozzle No. 1 Prototype Nozzle No. 2 Prototype Nozzle No. 3 Production Nozzle , Tank Assembly (Agent Tank) Nylon Cap and Nipple Flow Model of GFE Design Module With Flapper Plate Module With Standpipe Bulkhead Flow Model With Standpipe Bulkhead Flow Module With Central Settling Chamber . Central Settling Chamber vii Page 3 5 8 9 10 11 13 14 15 21 28 33 34 35 36 37 37 38 40 42 48 5: 31 53 54 55 56 �LIST OF FIGURES Figures (Continued) Title Testing Fixture ......... . •- • Page ............. 60 29 30 Module Adapter Designs ........... . ......... 67 PAU-8/A Multiple Module Configurations ............... , ............. 68 31 32 Module Adapter....................... ...... 72 MJ-1 Table and MHU-83/E Fork Adapter ....... V* 33 Two-Module PAU-8/A on Loading and Handling Adapter................. — . ....... 80 34 Four-Module PAU-8/A on Loading and Handling Adapter................. .......... 81 35 Three-Module PAU-8/A on Loading and Handling Adapter . . .......................... 82 Loading and Handling Adapter frTith Tilt-Adjusting Blocks for Use With A-l Aircraft ............................... 84 Loading and Handling Adapter With Tilt-Adjusting Blocks for Use With F-4 Outboard Stations .......... . ........... 85 36 37 38 Shipping Container with Cover Removed 39 Contamination Hardware (Turning Vane Kit).................... .............. 90 Configuration No. 1 - Basic Two-Module Dispenser ....................... 92 40 ...... 87 41 Configuration No. 2 - Four Short Fins.......................... ....... 93 42 Configuration No. 3 - Combination (Multi-Surface) Stabilizer ... .............. 93 43 Configuration No. 4 - Vertical Fin ........ . 44 Configuration No. 5 - Drag Plates 45 46 Wind Tunnel Test Results ...... . ..... ....... 95 Wind Tunnel Configuration Comparisons ................. . . ...... ...... 96 . Longitudinal Stability of PAU-8/A Two-Module Configuration ....... . . ....... 9.7 ... 47 viii .......... 94 94 �LIST OF FIGURES (Continued) Figures 48 49 50 51 52 53 54 55 56 57 58 Title Lateral Stability of PAU-8/A Two-Module Configuration Axial Force Characteristics of PAU-8/A Two-Module Configuration Two-Module PAU-8/A on F-86 Aircraft (Sixty-Five Percent Scale) Captive Flight Flow Patterns PAU-8/A Drop Zone Layout Production Nozzle Functional Level Troubleshooting Guide . . System Schematic Mission Profile for Sequential Dissemination Reliability Model , Line Checkout and Preparation Procedures ix Page 98 99 101 104 107 Ill 114 115 119 121 122 �LIST OF TABLES Table I II III IV V VI VII VIII IX X XI XII XIII XIV Title Physical Data and Operational Characteristics Sequence of Operation Environmental Tests on the Transfer System Agent Absorption of Nylon 6/6 40 Percent Glass Filled Mechanical Properties of Rubber Material Immersed in Herbicide Agents for Seventy-Two Hours at Ambient Temperature ... PAU-8/A Maximum Loadings Two-Module Wind Tunnel Configurations Tested Flight Maneuvers for Two-Module Captive Flight Tests Fit Test Compatibility Summary PAU-8/A Spray Tests Results Nozzle Descriptions Maintainability Data Success Probabilities for Subsystems Reliability Data Page 7 20 31 61 63 69 92 102 106 ^ 108 v -^10 118 120 124 �I INTRODUCTION This f inal report describes the work performed in the desian, development, fabrication and testing of an Aircraft Spray Tank Unit, PAU-8/A. (Prior to 29 June 1970, the PAU-8/A was known as the TMU-G6/A, Chemical Anticrop Dispenser, and is referred to by that name in some portions o* this report.) The program documented herein was conducted for the purpose of improving the design of a previously developed sprav tank. The desian improvement was for a chemical anticrop spray tank of modular design, consisting of nose cone, identical liquid container modules, module mating assembly, tail cone, liquid transfer and power unit, and dissemination apparatus. The system is capable of operating in four configurations, using either one, two, three, or four modules, which can be operated either simultaneously or in sequence. The four-rodule configuration has a capacity of 200 gallons (50 gallons per module). All nodules are equipped with both 14- and 30-inch lug suspensions and all are completely interchangeable. The tank is suitable for external carriage on high and low performance ground support aircraft employed in counterinsurgency (COIN) and tactical operations. The system provides the capability of attackina both large and small crop-growing areas. Under tnis proaram a dispenser was desianed; subsystems, r.odels and full scale prototypes were fabricated and tested for performance; and test items were delivered to the Air Force for R&D Engineering Tests. Subjects discussed in this report include development of the module, module adapter, dispenser test unit, loading and handling adapter, shipping container, contamination, subsystems testing, and system demonstration testing. �SECTION II SYSTEM 2 •1 DESCRIPTION PURPOSE OF THC EQUIPMENT The Aircraft Spray Tank Unit, PAU-8/A, is used for d«. .ivering chemical anticrop agents. The system is designed for external carriage on high and low performance ground support aircraft employed in counterinsurgency (COIN) and tactical operations and provides the capability of attacking both laroe and snail crop-growing areas. 2.2 GENERAL DESCRIPTION The PAU-8/A (Figure 1 and Figure 2) is of modular construetier.. Each module is capable of being used as a spray tank 'since each contains the necessary controls, valves, pylon attachments, nozzles, etc. for disseminating anticrop agents. The system can also be used in a two-, three-, or four-module configuration. Thus, the PAU-8/A can be arranged to match the load carryincr ability of the aircraft pylon selec .-d for use. An adapter is provided to assemble the modules whenever multiple-module configurations are required. The assembly requires an appropriate set of fins and electrical interconnections. Picture 1 depicts fin arrangements and principal dimensions. The spray tank is designed for use on the F-4, F-100, F-105, F-lll, A-1E, A-7, and A-26 aircraft, and may be used on other aircraft where adequate clearance and suitable pylons exist. Aircraft electrical provisions are required for operatina the tank. Each module consists of a fifty-gallon agent tank, a transfer system, an electrical control system, and a dissemination valve and nozzle. The agent tank is 13 inches in diameter, 106 inches long, and is fabricated from aluminum forging, castings, and rolled sheet aluminum. The interior of the tank has a protective coating to prevent the corrosive action of the anticrop agents from attacking the tank interior. The transfer system consists of a compressed-air/gas pressure reservoir, an arming valve, two check valves, a pressure switch, pressure regulator, low-pressure relief valve, charging valve, filter, pressure gage, high-pressure relief valve, and two bleed valves. These items, along with the electrical control system, are located at the forward end of the modulo and are housed in the forward fairino inclosure. A pneumatically �W tTCM *, SUSPENSION LUC. MAY ftC USCD m PL 30-mCM, FOA-ICONFICUR«rtON 01 Figure X. Aircraft Spray �fiOLT-UACHINE AIRCRAFT WARNC3S A&SY, I N?E_RJMOOULE __ CMfM'ICAI AM^t icAo^ 41 N6i« JS.99JJI* ; FIN , SHORT 12 13 14 AOAPTER, MODULE CAP, FIN HOLDER «*. SUSPENSION LUC. MAV sr USED IN FH.*£C OP ^TEM 12, LJG " - ' i SO-IMCH, FOR -lCOMPlftUKATiON CNLY AS SHOWN IN VIEW B^B 6 LUG, SUSPENSION 17 ' 3O INC WIRC.LOCK WASHER RAD.^WAY BRACE. IHNER PAD." SWAY INNER FIUI9TCR HD PIN.QUICK RELEASE ! 1. Aircraft Spray Tank Unit, PAU-8/A (The reverse of this page is blank.) �.Ji'raagaCTjfKCi«*iA.i.i.* - Figure 2. Aircraft Spray Tank Unit, PAU-8/A (Four Configurations) �operated valve provides the on-off control for dissemination, and an adjustable nozzle is mounted at the aft end of the module. An aft aluminum rairing inclosure is provided for the valve and nozzle. Operation of the system requires two actions by the pilot: (1) Closinc? a cockpi'. arming switch which'arms the system and releases stored high-pressure air/gas into the agent tank.at a rerulatecl pressure, and (2) depressing the "pickle button" on z'.'.e pilot control stick which opens the dissemination valve and ?.lLcv:s the aaent under pressure to be directed to the nozzle ?.sser.cly. Dissemination continues until the pilot releases the button which closes the dissemination valve. Nozzle adjustment is set prior to flight to vary the flow rate (from 15 to 150 gallons per minute per module). The modules in the multimodule configuration can be operated simultaneously or in sequence. 2.3 PHYSICAL DATA AND OPERATIONAL CHARACTERISTICS Physical data and operational characteristics of the system are shown in Table I. 2.4 DETAILED DESCRIPTION OF SPRAY SYSTEM Equipment furnished for the PAU-8/A system includes the spray tank unit (.Multi-Module) , an' electrical test unit (Figure 3), loading and handling adapter assembly (Figure 4), and turning vane kit (Figure 5). 2.4.1 2.4.1.1 Spray Tank Construction - General Each of the four modules of the PAU-8/A contains the necessary controls, valves, pressure reservoir, pylon attachments, nozzles, etc. for disseminating chemical anticrop agents. The nodule consists of a tank assembly which stores the.agent, a pneumatic system which pressurizes the agent tank, and the agent-dispensing system for control of agent, dispersal. Interconnection of the aircraft control system and the PAU-S/A system, is provided through an electrical connector rcur.ted in a well aft of the mounting lugs. The assembly of a cv:c-, three-, or four-module configuration requires a module �TABLE I. PHYSICAL DATA AND OPERATIONAL CHARACTERISTICS PHYSICAL DATA SINGLE MODULE DISPENSER Store (overall) Length Maximum Diameter Weight Modules Small Fins Large Fins Mating Assembly Straps Weifht Total (Empty) Agent (i.fl Specific Gravity) Weight F i l l e d (1.0 Specific Gravity) Agent (1.2 Specific Gravity) Weight Filled (1.2 Specific Gravity) Agent (1.4 Specific Gravity) Weight Filled (!.4 Specific Gravity) Agent Capaci ty Lugs (Per Hil-Std-8591) TWO-MODULE DISPENSER THREE-MODULE DISPENSER FOUR-MODULE DISPEKSER 135.51nches 32.5lnches 135.5lnches 13 Inches 135.5!nches 32. 5 Inches 214 Pounds 12 Pounds 429 12 8 31 15 Pounds Pounds Pounds Pounds Pounds 644 6 16 31 22 Pounds Pounds Pounds Pounds Pounds 858 12 16 31 30 Pounds Pounds Pounds Pounds Pounds 495 872 1367 1048 1543 1220 1715 Pounds Pounds Pounds Pounds Pounds Pounds Pounds 719 1303 202? 1572 2291 1830 2549 Pouncs Ps'jr.ds Pounds Pounds Pounds Pounds Pounds 947 1744 2631 2096 3043 244Q 3387 Pounds Pounds Pounds Pounds Pounds pounds Pounds 226 435 8B2 524 750 610 836 Pounds Pounds Pounds Pounds Pounds Pounds Pounds 50 Gallons 14 & 30 Inches 135.5lnches 32.5tnches 100 Gallons 150 Gallons 2C9 Gallons 3Q Inches 30 Inches 30 Inches STA. 65.9 STA. 69.2 4 Small 2 Large STA. 65.5 STA. 69.0 2 Smal 1 4 Large STA. 64.9 STA. 68.8 4 Small 4 Large STA, 64.8 STA. 68.8 Agent Flow Rate 15 -150GPM 15-3QQGPM 55 -45QGFM 15 -6QOGPM Electtical Data 28 VOC 1 Amp 28 VOC 2 Amp 28 VOC 3 Amp 28 VOC 4 Anp 151 3/4 "Long 33 1/4" Wide 41 3<8"High No. of Fins Required Center of Gravity Position Center of Gravi ty Empty Center of Gravity Full (1.0 Specific Gravity) 4 Small Shipping Container None None None �00 Fiaure 3. Electrical Test Unit �ADJUSTING 3LCT,KS FOR A-1 AND F - 4 1RCP.AFT VO :i HOLD-DOBN BOLTS • cv.-^* .••• •^•^.' "X:-- •*.-^:-: '-*•••'. ".'::--.>.;;'.'1^ ^^:"&^^\ ''•^tti-Z:'*'---??;?-'-* .. ^ : S| •v^-y->.vy?;« •.'>.''-: .;'>-.VulS ^•"'••^•yra ^V:^V^:^.. Vl - ->v^3 . . . Figure 4. Loading and Handling Adapter Assembly (Modification Kit for MJ-1 and MHU-83/E Bomb Lift Trucks) ; .^>s' 'V-J»< �Figure 5. Turning Vane Kit adapter. Two high-strength stainless steel straps are used to attach each module to the adapter. 2.4.1.2 Tank Assembly The tank assembly (Figure 6} of the module is 13 inches in diameter and approximately 106 inches long, fabricated from aluminum forging, castings, and rolled sheet metal. The hardback is an aluminum forging containing provisions for 14-^inch and 30-inch lugs for attaching the dispenser to the aircraft pylon. Two hand-support bulkheads are welded to the hardback forgihgs to support the tank when assembled to the module adapter. The inner tank or settling chamber, with a capacity of approximately seven gallons, is located between these two bulkheads. 10 �FIU PORT tHHER TMK STROHGBACK PICK-UP STARDPIPE BULKHEAD Figure 6. Tank Assembly �The skin is a 1/8-inch aluminum rolled sheet fitting around the band-support bulkheads and butting against the hardback. The skin is welded to the bulkheads and hardback. The tank is closed at the aft end by an aluminum casting v/elded to the skin. The forward end is closed with a drav/n aluminum bulkhead welded to the skin. A pick-up tube runs from the inner tank to the aft bulkhead and is welded to the rear band support sulkhead and the aft bulkhead. The end bulkheads contain provisions for mounting the pressure reservoir with attached valves, electrical control box, and nozzles. The tank also contains fore and aft filler caps and fin attachment provisions. The inside is coated to protect the aluminum from the corrosive action of the agent. 2.4.1.3 Agent Transfer System (Pneumatic System) The pneumatic system (Figure 7) consists of a 3,000 psi pressure reservoir, an arming valve, two check valves, a filler valve, a pressure gage, a pressure switch, pressure reaulatcr valve, low-pressure relief valve, and a high-pressure relief valve. These items, along with the electrical junction box, are located at the forward end of the module and are housed in the forward fairing. Two bleed valves are provided to vent the residual pressure in the agent tank prior to servicing the module. The fluid-transport-system schematic (Figure 8) shows the arrangement of the equipment. • The pressure reservoir is attached to the front bulkhead of the tank by six screws. A single bolt retains the forward fairing. The aft end of the forward fairing rests against the skin of the tank. The arming valve, pressure regulator and interconnecting piping are attached to a mounting plate which is bolted to the pressure reservoir. 2.4.1.4 Agent Dispensing The dispensing system (Figure 9) consists of a dissemination control valve (shut-off valve) and the adjustable nozzle located at the aft end of the agent tank. The dissemination control valve, housed in the aft fairing, is operated bv pneumatic pressure piped from the pressure reservoir. The nozzle and dissemination valve are attached to the aft I'ulk'i.cad with four bolts. The aft fairinn engages the rear i".tlkho.ui and is attached to the nozzle by four screws. 12 �Fioure 7. Ayent Transfer System 13 �H I G H PRESSURE GAS IS P U M P E D I M J S»ST£» HERE |~' »IRCIUFT~I COCKPIT C H » « : i N C V A L V E -»* ., ' Jj -^X MlCH-PR£S$!jl»E 6AS F I L T E K — • v3ISS£"l««»t <m£> S«ITCN CAUGE HIGH-PRESSURE G»S S T O t l C E CONTAINER (3:00 pi.i) —NOTE: COMPONENTS INSIOE C 3 T T E O L I N E ADE A T T A C H E D 10 T H E S V S ' E V M A N I F O L D L O C A T E D I N IhE NOSE O F T H E PRESSURE S*ITCH A:E«T CHECK VALVE GAS-PRESSURE «GtlATOR_ . I ' 10*-»R{SSURE BLEE3 VALVES I.. «.PBESSURE OEUEF VALVE • •• Fill PORTS •UM STRAINERS ALtAYS HIGH PRESSURE 000 H I G H PRESSURE tHEN 10 1 PRESSURE IHEN AMiEO D I S S E M I N A T I O N rilOT VALVE . D I S S E M I N A T I O N VAIVE ACIUAIOR A • INNER.: "•' TANK ;:j TO ATHOSPHERE ENT INTO AID3SPHERE ACENT PROTECTIVELY AOlUSTAfLE COATED AGENT TANK O I S S E B I N A I I O N BALI VALVE ANT I-SLOSH CHAMBER Figure 8. AGENT P I C K - U I P TUBE PAU-8/A System Schematic 14 �Figure 9. Agent Dispensing System �2.4.1.5 Module Adapter Assembly The module adapter consists of two castings joined together with a tube. The forward casting contains four locator pins to rate with the modules. The straps for nolding the modules are attached to each casting of the adapter by ball-lok pins. 2.4.2 2.4.2.1 Principles of Operation Basic Principles The operating principle of the system consist's of pressurizing the agent tank with filtered air or nitrogen which forces the agent to flow from the tank through the adjustable spray nozzle. A> Fill Ports - The tank can be filled through either the forward or the aft fill port. The fill ports are equipped with strainers to prevent the filler hoses from entering and damaging the inner coating of the tank and to prevent large foreign particles from entering the tank. If the agent tank is pressurized above five psi, the filler caps cannot be removed until the tank is bled to about two psi. The bleed valves must be left open during agent filling to allow cr.e air to bleed out of the inner tank; otherwise, the tank will r.ct fill. B. Tank Pressurization and Regulation - The pressure reservoir is filled through the charging valve located on the large r.anifold mounted on the reservoir. The gas going through the charging valve passes through a 10-micron filter before entering the pressure reservoir. The reservoir has a volume of 650 cubic inches and is charged from 1,200 psi to 3,000 psi, decendino upon the operating temperature. A high-pressure gage mounted, on the reservoir reads the pressure in the gas reservoir. The pressure reservoir has enough volume to go through a temperature cycle from +160°F to -65? F and still have enough pressure remaining to completely empty the agent tank at the low temperature. The hicrh-pressure relief valve, mounted on the reservoir next to the pressure gage, prevents the reservoir from being over-pressurized and allows gas to be relieved at high temperatures. The valve opens at 3,400 psi and reseats at 3,100 psi. 16 �A pressure switch, mounted on the large manifold of the pressure reservoir, performs a switching function when the tank is empty. The switch opens at.250 ± 50 psi and remains open until the pressure in the reservoir drops to 200 + 10 psi, at which time it closes. The solenoid-operated arming valve is located in the large manifold and is controlled by the arming switch in the aircraft cockpit. Actuation of the arming valve allows the high pressure gas to flow to the pressure regulator located in the large manifold. The ourlet operating-pressure of the regulator is 55 ± 5 psig; the nonflow or lock-up pressure outlet of the regulator is 71 psig. There are two outlets from the regulator: one to the dissemination valve and the other to the agent tank. After the gas leaves the regulator on the way to the agent tank, it first passes through a check valve and then by a low-pressure relief valve. This low-pressure relief valve prevents over-pressurization of the tank in case of failure of the regulator. The valve cracks (opens) at 85 psig and reseats at 75 psig. Before entering the agent tank, the reaulated eras passes through a second check valve. A section of nylon tubing conriectincr this second check valve to the tank acts as a trap to reduce the amount of acrent in contact with the seat on the check valve to prevent build-up of crystallized agent on the seat.The two bleed valves mounted on the forward bulkhead are for bleeding the pressure from the agent tank before the filler caps can be renoved and for venting the internal settling chamber during filling. C. Dissemination Control.Valve and Nozzle - The dissemination control (shut-off) valve and nozzle are located at the aft end of the agent tank. This is a pneumatically operated ball valve whose operation is controlled by a solenoid-operated pilot valve through a piston, rack, and gear. The pilot valve controls regulated air piped from the regulator to the piston. Actuation of the pilot valve is controlled bv the bomb switch (pickle button) located on the control stick of the aircraft. When the dissemination control valve is open, the pressurized air in the agent tank forces the agent into the settling chamber through the stand-pipe and out through the agent pick-up tube. The settling chamber allows the air mixed with the agent to separate from the aqent before entering the agent pick-up tube, thus insuring that no air passes out of the nozzle until the agent tank is empty. This inner tank is the last to empty 17 �during dissemination and prevents the uncovering of the aaent pick-up tube and subsequent loss of tank pressurization during the nose-up or -down flight. The nozzle assembly, located aft of the dissemination con~r:l -.-.live consists'of a housing, diaphragm, and nozzle adjustr = r.-; «l=e*.-e. The dianhraqm is made of a flexible rubber over s-iffsr.ers and is threaded to the housing at the aft end. The nozzle adjustment sleeve is threaded to the housing and fits over the diaphragm. The nozzle is adjusted by screwing the sleeve in and out; two setting labels provide for adjustments from 0 to 16 and from 1 to 15. When pressurized agent is flowing through the nozzle, it forces the flexible diaphragm to move back until the sti'feners come to rest against the nozzle sleeve. When the nozzle sleeve is screwed all the wav in, the orifice in the diaphragm is small; screwing the sloeve out enlarges the orifice and adjusts the flow rate. The nozzle sleeve is held at the desired setting by a ball detent. Two drain ports are provided in the agent tank; one is located in the bottom of the aft bulkhead for draining the main tank; the other is located in the side of the rear band-support bulkhead for draining the inner settling chamber. A third port is located in the upper rear of the aft bulkhead. This port is for a pressure gage to be used during calibration of the tank* D. Electrical Control - The electrical circuitry (Figure 8) in the PAL'-8/A consists of control circuitry for operating the arming and dispensing control valve. The control circuitry provides the means for arming the module and opening and closing the dissemination control valve. Operation is from the controls located in -che aircraft cockpit. The circuitry includes a selector switch mounted in each module for selecting the mode_ of operating the multi-module system (simultaneous or sequential) . The system utilizes 28 VDC supplied from the aircraft system. Two control functions are required for arming and operation of the rodules. The arning switch applies power to the arming circuit. When this circuit is enerqized, power is applied to the coll of the armino relay and to the solenoid of the orminq "nlvc which- opens the valve and nllows the aqont tank to be pressurisod. �Dissemination is obtained by operating the bomb release (pickle button) on the pilot's control column. This circuit, when energized, applies power through the arming relay to the dissemination solenoid control valve. When the bomb release switch is released, power is removed and the control solenoid closes, shutting off the flow of agent to the nozzles. A selector mode switch is installed on a junction box assembly mounted in the forward end of each module. The selector switch provides four positions: Position No. 1 --sequential rrodule usaoe; Position No. 2 — simultaneous usage of two modules, followed by the remaining two modules^. Position Mo. 3 -simultaneous usage of three modules, followed by the remaining module, and Position No. 4 -- simultaneous usage of all four modules. Although there is a selector switch in aach module, only the top switch needs to be set to get the desired sequence. The remaining switches can be in any position without affecting the sequence. Sequencing of the modules is achieved through the use of the pressure switch. When' module pressure drops below 200 psi (agent tank empty), the switch moves to the NC position (normally closed), diverting power through the selector switch to the next module, or modules, in sequence. The sequence of operation for two-, three-, and fourmodule dispensers is shown in Table II. A safety switch (arming switch) is used to prevent aircraft power from inadvertently entering the system while on the ground. The safety switch is located on the forward bulkhead just aft of the nose fairing and is actuated by a quick release ball-lok pin, which is inserted through the top side of the module. A red "REMOVE BEFORE FLIGHT" streamer is attached to the switch actuation pin. 2.4.3 Shipping Container The shipping container (Figure 10) is a wooden, wirebound container fabricated from plywood and timber consisting of a base, side assembly, saddle and saddle retainer, and cover assembly. The side and ends are held together with wire and, when assembled, interlock into the base with a cleat at the bottom of the sides. The retainers are held in place on the base between blocking pieces. The saddles interlock in the retainers and are held to the dispenser by two 0.75 in by 0.023 inch straps. The fins are stowed in two boxes mounted to 19 �TABLE II. SWITCH POSITION TYPE OF DISPENSING Sequential 1 SEQUENCE OF OPERATION TWO-MODULE CONFIGURATION THREE-MODULE CONFIGURATION I .». t A ® o". 0 ««. to o 2 • Simul taneous (Two Modules) 3 Sitnul tnneous (Three Modules) 4 Simultaneous (Four Modules) FOUR-MODULE CONFIGURATION «2» ^ »2« 0 ©Q �^^^ti^^^-^f^^ Figure 10. Shipping Container �the base and the covers to the fin storage compartments arc held in place fcv one strap on each compartment. The container cover is secured by four straps which qo around the entire container. The fcur-r.odule configuration is rotated 45 degrees from its r.rrr.al upricrht position to reduce the storage and shipping volure rf the container. 2.4.4 Dispenser Test Unit A test unit (Figure 3) is used to aid in monitoring and functionally checking the system as well as for functional checks of the aircraft and pylon control circuitry. 2.4.5 Loading and Handling Adapter Since the two-, three-, and four-module configurations will not fit on the standard adapters furnished with the MJ-1 and MKU-83/E bomb lift trucks, a special adapter is furnished with the system so that these two bomb lift trucks can be used to load any configuration on the aircraft. This loading and handling adapter allows the two-bomb lift trucks to load on the outboard stations of the F-4 (which have a 7-1/3-degree cant angle) and on the Al-E (which has an incidence angle of 30-1/2 degrees), as well as on other aircraft pylons. 2.4.6 Contamination Control Adapter Kit (Turning Vane) This hit is supplied for use with the F-4 aircraft to reduce the spray contamination of the under side of the wing. The kit is desierned for use with the three-module configuration. 22 �SECTION III MODULE DEVELOPMENT 3.1 MODULE REQUIREMENTS • Deliver chemical anticrop aqonts, future chemical agents, and plant grov/th hormones. • Be suitable for external carriage on high-and low-performance, ground-support aircraft employed in counterinsurgency and tactical operations. • Be of modular design. • Have identical liquid containers. • Have a dissemination apparatus with off-on capability. • Capat-le of operating in various configurations, using single modules and combinations of ttvo, three, and four modules. • Operate modules simultaneously or in sequence. • Agent container modules shall be completely interchangeable. • Four-module configuration shall have a capacity of at least 200 gallons. • Modules shall be equipped with 14-inch and 30-inch lug suspensions. • The number and combinations of modules shall be determined by the weight capability of the given station. • Modules shall be readily and quickly filled and serviced while suspended on tne aircraft. • Be compatible with armament circuitry of each aircraft. • Cause no appreciable degradation to the aircraft operational performance. • Be capable of delivering agent with a specific gravity of 0.80 to 1.40 and pH within the range of 3.0 to 8.5". • Capable of storage of agents for a minimum of six months, 23 �Capable of five-year storage with no loss of perforrar.ce. Ar aircraft operational speeds of 200, 325, and 500 KIAS, be capable of disseminating a liquid agent with l.U specific gravity so that the spray will be mostly in the particle-size range of 200 to 400 ricrons ness median diameter (mud) and with less than 10 percent by volume of the agent being outside the range of 100 to 500 microns raid. Arming and dissemination shall be two separate and distinct actions. Dissemination shall continue until bomb switch is released. i All configurations shall be capable of carriage to speeds of 1.5 Mach at 35,000 feet and up to Mach 0.90 at sea level. Function at speeds of 150 to 600 KIAS within lead factor envelopes of MII.-A-8591C. Safe ejection from any aircraft station with minimum limitations to ejection speed of the aircraft. Capable of delivering agents at 25, 75, and 150 gallons per minute at speeds from 200 to 500 KIAS. All configurations shall be capable of simultaneous operation with minimum requirements for electrical power from the aircraft. The empty weight of one nose cone, one tail cone, one agent module and associated control and piping equ'1^ment shall not exceed 250 pounds. Minimum of three man-hour mean time to repair four ules. Dd- The failure rate of the item shall be no greater than 10 percent. Minimum of parts are to be for one time use. 24 �3. 2 AGENT TSANSFTR gysrp! The purpose of the 'anent transfer svsten is to force the 50 qallons of aaent through the dissemination nozzle and out the rcdulo into the atmosphere. In addition, the agent transfer system is also designed to: • Occupy minimum space. • Be lioht woioht. r • ur.etion with hiah reliability under a vurietv of environmental conditions. • 'eouire minimum maintenance. • Be easily repaired. • Provide a constant flow rate throuahout a aiver. settir.r. • Control the flow of agent. • Be safe to operate and service. 3.2.1 Fluid Control Three basic methods of movina the aaent v«ere investigated; • Pur.pma the aoor.t out of the tank bv an electrically or pneuratically driven punp. • Contair.ina the aaent in a bladder and forcing it out by pneuratic or nechanical means. • Pressurizina the aaent in its tank bv a pneumatic svster and control lino the flew bv the control valve. An invest ioation of these methoas indicated that the and power consurption of a pump ssysten would be prohibitive. ?,~c that the cost of an effective bladder s"ster. would be rue'.' hir>.« er than that of a pressurized tank system. The desior.ec: .".sentransfer svsten stores er.ouoh compressed eras' oneroy to fcrce tr.5 a{?ent out" of a tank and to operate the aaent flow control valve, To accomplish this function, the system must contain the relieving components: 25 �• High pressure gas storage reservoir. • G-is charging valve and filter. • >?as pressure gage. • High pressure relief valve. • Pressure activated electrical switch. • Electrically operated preurtatic control valve. • Pr.our.atic pressure regulator. • Lev-pressure check valve. • Lev-pressure relief valve. • Low-pressure bleed valve. These components nay be joined into a system in two ways. They r.ay be connected together by pneumatic lines and fittings, cr i-y a r.anifcld which would connect and support the components. If there is r.o need to separate the individual components to fulfill the systor. function, a manifold offers the following advantages over pneumatic connecting lines: (1) higher reliability due to decreased susceptibility to x'ibration and shock and fewer external joints to develop leaks, (2) greater maintainability lecause of easier access to individual system eler.ents, and (3) lever overall cost due to the reduction in the nurrber of pneuratic fittings required, the greater reliability, and the re.'.-.-.c^d assembly and maintenance. Because it was both desirable -.r.^ feasible to locate most of the components in the nose of the r;dule, a r.ar.ifold system was employed. Tests were conducted on four GFE modules for the purpose of .•eterrir.ir.c! their operational qualities. These tests supported i.-.e selections of the above components. The tests indicated ir.at: • Failure of the burst plug used to prevent the tank from over-pressurization was due to fatigue. • The fittings used to connect the system's components leaked. • Failure of the arming valve allowed the agent tank to be pressurized before arr.ing occurred. 26 �• Failure of the check valve located between the arming solenoid and regulator allowed fluid to enter the arming valve and the dissemination pilot valve in the rear of the nodule. The energy source of the agent transfer system is 3,000 psi nitrogen or air stored in the high-pressure container. The system is controlled by electrical signals initiated by the pilot. Fellowinq is a breakdown of agent pneumatic transfer syster. crrpcr.ents (Figure 11), explaining the function, perfcrr.ance requirements, and special features of each: '. 1) Tre hicTh-pressure gas storage reservoir is a hollow , sphere, the two halves or which are drawn irom shatter-proof steel and welded together. Lugs for bolting the sphere directly to the forward bulkhead of the module and bosses for attaching two r.anifolds are welded to the outside of the sphere. Each sphere is tested for weld integrity, heat treated, and then subjected tc a serios of pressure tests before it is accepted for use ir. ~he syster.. The sphere is designed to withstand 2-1/2 tires the actual system pressure of 3,000 psig. {2} The electrically—operated pneumatic control valve (arr.ir.a valve) controls the flow of high" pressure" etas from the reservoir to the rest of the system. It is normally closed (to subject as few parts as possible to high pressure gas) except when the syster. is functioning. This reduces the hazard of dar.agir.g the aircraft if the module is danaced in flight, ar.i reduces the nur.ber of potential leak points. V.'hen the ar~ switch is en, the arming valve opens, allowing pressurized gas to flcv through the system into the agent tank. The system can be disarr.ed by turning the arming switch off. The valve is a spring return type and automatically closes when the solenoid is de-energized. This type of aruing valve ailovcs the pilot to disarm the system after it has been armed in the event the mission is discontinued, and, in the event of a power failure, the syster a-utomatically returns to an unarmed configuration. The tar.k re-ains pressurized after the arming valve is closed. (3) The gas charging valve is a stainless steel device threaded, into the system manifold which allows the pressure sphere to be filled*with gas while the module is on the ground. A' ten-ricron filter is incorporated in the charging valve to keep contaminants from entering the systcro. �LOW-PRESSURE CHECK V A L V E ( 7 REGULATOR OUTLET PORT PNEUMATIC-/PRESSURE ( 8 REGULATORv 4 J5>STEW MiNlFOLD HIGH-PRESSURE RELIEF VALVE REGUL4TE!) OOTLET TO D I S S E M I N A T E ULVE GAS , , PRESSURE(10J GAUGE V FILTER HIGH-PRESSURE GAS STORAGE ELECTRICALLY-OPERATED PNEuHATIC CONTROL VALVE Figure 11. Nitrogen Storage and Control System 28 �(4) Manifolds are attached to the pressure sphere to replace mounting brackets and pneumatic lines. They are machined blocks ported in such a manner as to properly connect the transfer system components attached. (5) The low-prassure relief valve is necessary to guarantee that the"pressure"In the agent tank will not exceed 92 psig even if the pressure regulator fails. In the event cf regulator failure, the valve will vent the pressurized gas into the atmosphere at a sufficient flow rate to prevent the agent tank from being over-pressurized. When the tank pressure drops to 75 psig, the valve reseats itself. (6) The pressure-activated electrical switch is located ir. the manifold between the reservoir and the arming valve. Its pressure sensing components are set to throw an electrical switch from one pole to another when the pressure in the sphere falls below 200 psig. This action interfaces with the system logic in such a manner as tot (a) allow the system to be armed only if the sphere pressure is greater than 200 psig; (b) automatically turn the dissemination valve off when the sphers pressure reaches 200 psig; (c) automatically initiate the next programmed step in the dissemination sequence. (7) The low-pressure check valves permit fluid or gas flow ir. cr.e direction only I Their purpose is to prevent the flow of ager.*: hack up the pneumatic lines from the agent tank to the resz cf the aaent transfer system pneumatic components. Two c'r.eck valves are used to insure that the agent does not back up into the agent transfer system; thus, one check valve could fail without any harmful effects to the system. (The second valve was added after failure of the check valves on the GFE modules.) (8) The pneumatic pressure regulator is designed to maintain a constant flowing pressure of approximately 55 psig in the agent tank. No-flow pressure is 71 psig. (9) and (10) The jiigh-pressure relief valve anc the gas pressure gage are both riounted on a small rr.anifolc! which attaches to the pressure sphere to provide a good view of the gas pressure gage. This gage is designed with a color coded scale to simplify the charging procedure and to add to the safety cf the system by providing accurate indication of the gas pressure. The high pressure relief valve increases the safety of the syster. by negating the possibility of over-charging the pressure sphere. This valve opens to allow gas to escape when the reservoir gas pressure reaches 3,400 psig and remains open until the gas pressure in the reservoir has fallen to 3,100 psig. At 29 �3,100 p'si'g the valve reseats itself. The low-pressure bleed valves located on the forward bulkhead cf the module allow the low pressure part of the system tc ce bled to atr.cspheric pressure in order to remove the tank fill caps. The valves have been designed and located in such a way that the module nose cone cannot be attached to the module until these valves are closed, thus assuring that the module cannot be flown with the valves open. 3.2." Component Testing Two agent transfer system units (Figure 11) underwent environmental tests to insure the design would meet specifications. These tests are listed in Table III. During low temperature (-65°F) tests, the solenoid arming valve stuck and leaked, and the low pressure relief valve leaked. These problems were solved by better surface finish on the base of the solenoid valve, changing the 0-rings to silicone rubber, and better alignment of the poppet with the base and solenoid. This reduced the amount of pull required to operate the valve. The seat of the relief valve was changed to silicone rubber. Atter these changes were made, the units were retested and found acceotable. 3.3 DISSEMINATION SYSTEM The purpose of the dissemination system is to control the start and stop of the agent flow, the flow rate, and tc disser.ir.ate defoliant agents in such a manner that they reach the crcur.d vegetation in 20C-to 400-micron diameter particles, after ejection, from high speed aircraft at an altitude of 100 feet. There are two basic methods of controlling the start and stop of the flow. • • Electromagnetically, or Electropneumatically. An investigation of the power requirements to operate the valve in less than 75 milliseconds indicates that the available electrical power was inadequate -o do anything other than perform a control function. Because of the lack of electrical power, the electropneur.atic systen was chosen so that the 30 �TABU-! III. ENVIRONMENTAL TKSTS ON TliK TRANSFER SYSTEM TEST UNIT •• I..1NIT : S y s t e n Leak And P e r f o m a n c p X X Lo» P f e s s u i e Per M I L - S T D - 3 1 0 8 . Method 5GO I - Piocptline M X High T e m p e r a t u r e Per HIL-SID-810B. Method 531.1 - Pioceduie 1 X Lo* T e m p e r a t u r e Per MIL-STD-3108. Method 502. 1 - Pioceduie 1 X Tenpeialiire Shock Per MIL-STD-8ICB. Method 533.1 - Pioceduie I X Tempera t u r e - A I ti iude Per HiL-$TD-8IOB, Method 5 0 4 . 1 - Piocedure 1 X Hunidity Pei MIL-STO-8IOB, Metliod 537.1 - Piocedure 1 X Fungus Per MIL-STD-8IOB. Method 503 1 - Piocedure 1 X Sr.lt Fog P 5 f M I L - S T O - 8 I 3 B . M e t h o d 503 1 - P i o c e d u r e 1 X Sand And Ous.t Per MIL-STO-8IQB. Method 510.1 - Procedure 1 X txplosive Atmosphere Per HIL-STO-610B. Method 511. 1 - Pioceduie 1 X Acceleration Per MIL-STD-610B. Method 5 1 3 . 1 - P r o c e d u r e 1 X Vibration Per MIL-GTO-BinB. Method 514.1 - Procedure 1. Equipment Class 1 Mounting A, Figure 5 1 4 . 1 . C u r v e 0 X Acoustical Noise Pei MIL-STD-8IOB. Method 515 - Pioceduie 1 X S^ten A c c e p t a n c e ( S y s t e m Leak And P e r f o r m a n c e ) X 31 X �ccr.pressed gas within the module could perform the power operaticr.s. To accomplish the desired function of dissemination, the following items are required: • Dissemination pilot valve (solenoid valve). • Dissemination valve pneumatic actuator. • Dissemination valve. • Nozzle. All components are attached to the aft bulkhead of the module to reduce the opening time of the dissemination valve. (The pilot valve, the pneumatic actuator and the dissemination valve comprise one modular assembly.) The first two components are required to operate the dissemination valve through a signal from the cockpit of the aircraft. They are inclosed in a stainless steel housing attached to the top of the valve. Each of these components has been designed to withstand agent contamination, both structurally and functionally. When the pilot gives the arm command to the module, regulated 60 psig gas is introduced into the inlet port of the pilot valve. When the dissemination signal activates the pilot valve solenoid, the regulated gas is allowed to enter the cr.eur.atic actuator piston cylinder. The gas pressure drives the piston to the other enc of its cylinder. This opens the dissemination ball valve chrough a rack and pinion gear. When the solenoid pilot valve is deactivated, the air from the actuator piston is bled off and the piston is spring returned. This closes the ball valve. This type of valve assembly was selected because it offers the highest reliability, smallest size, lightest weight, lowest pressure drop through the valve, and the lowest cost due to its design. The valve will open completely within 75 milliseconds of the pilot's command. To develop a nozzle, a series of static and flight tests (Section X) was conducted. The results of these developmental tests indicated that: • The orifice size had small effect on the size of the particles ultimately striking the ground. • 7hi> rol.T*:iV'.' vc-' -cii_y of air striking the ejected part iolo:'. find .1 major of foot on tlu-ir size. �• 'i'hc primary effect of varying agent flow rate on droplet size was due to the resulting changes in slip stream* particle relative velocity and not to internal turbulence. Nozzle desiqn evolved in the following manner: To generate data, a nozzle was designed, fabricated and tested. This first nozzle (Test Nozzle No. 1, Figure 12) was designed so that agent flow rate, orifice size, nnd agent shear direction could be varied. Using Test Nozzle No. 1 and the GFE nozzle, (Figure 13) static and flight tests were conducted. As a result of these tests, more simplified test nozzles were designed to further study nozzle configurations under dynamic conditior.r . '"ORIFICE PLATE Figure 12. Test Nozzle No. 1 Test Nozzle No. 2 used one- and two-orifice plates with deflectors (Figure 14). Test Nozzle No. 3 (Figure 15) used conical nozzles with different size restricting orifices up-stream from the exit nozzle to regulate flow. From aircraft flight tests with these nozzles, it was determined that a variable size orifice could be used to get the proper droplet size as well as to control the flow rate. Prototype Nozzle No. 1 (Figure 16) using an elastic diaphragm backed up with metal stiffeners bonded on the diaphragm was then fabricated. This first prototype was used in static tests to demonstrate the design. Prototype Nozzle Ho. 2 (Figure 17) consisted of a stainless steel housing with a molded diaphragm and with stiffeners -elded into the diaphragm. When the nozzle underwent static tests, the molded stiffeners prevented the elastic diaphragm frc~ stretching, causing it to tear. A flat free-floating stiffer.er was designed to overcome the tearing problem. Prototype No. 3 (Figure 18) underwent successful static cycling and flight testing. After these tests, the nozzle housing was redesigned for weight reduction and to simplify fabrication and production. Polypropylene was selected to mold the nozzle 33 �Figure 13. 34 GFE Nozzle �Figure 14. Test Nozzle No. 2 sleeve and body because, in addition to being easy to mold, it is also resistant to the agents used. The nozzle allows the agent to pass through the circular orifice, which can be varied from 3/8-inch diar.eter to 1-1/4inch diameter by rotating the adjustment sleeve surrounding the orifice. The body of molded polypropylene connects the nozzle orifice to the dissemination valve, functions as an agent accumulator in order to reduce fluid turbulence, and -supports the nozzle orifice diaphragm, adjustment sleeve, and tail cone. The adjustment sleeve, also of molded polypropylene, controls the orifice size and protects the nozzle diaphracr. from damage during loading and storage. The nozzle diaphrasr. (molded fluorosilicone rubber with a 3/8-inch diameter orifice in its center) is molded to a threaded stainless steel ir.sert. Tec flat stainless steel inserts are held to the orifice by retainers molded in the rubber. They run radially fror. the ^orifice to the outer perimeter of the diaphragm. When the dissemination valve is opened, agent flows into the nozzle, * fills it, and pressurizes the inside face of the nozzle dia.phragra. The diaphragm deforms under the pressure with an out- 35 �Figure 15. Test Nozzle No. 3 36 �&M ADJUSTMENT SCt£»ED IN TO DIAPHRAGM THIS »UO*S 3tWH»*uM TO Mill MUM EJECTION VELOCITY. 3UT. OuT«»80, INCKEASINu 9<IFlC£ SUE. EJECTION VELOCITY AND MAINTAIN MINIMUM 9u»H«AaM EIPANSION AND. THUS. o FLOW. Figure 16. rrotctype Nozzle No. 1 VALVE ZLE 0«IFICE SITTING LASEL N02ZLE ADJUSTMENT SLEEVE N022LE AMVSTXNT DETENT ACCUNULATO* TUBE ORIFICE Figure Prototype Nozzle So. 2 37 �Figure IS. Prototype Kozzle No. 3 �ward buKTo which enlaraes the orifice. Ordinarily, the diarhraam would tend to form a hyperboloid when pressurized; however, 'the flat inserts cause it to.fern a cone. This allows the orifice diameter to be precisely controlled by the adjustment sleeve, which has a hole in it larger than the laraest orifice diameter desired, but smaller than the diaphragm's outer diameter. This annular end cap is prepositioned to the desired distances outside the diaphragm by screwing the adjustment sleeve in or out (Fioure 19), Thus, when the pressurized agent deforms the diaphragm, the adjustment sleeve end cap determines the central angle of the cone formed and, consequently, the orifice diameter. The nozzle orifice diameter adjustment is easily race external to the module. The tail cone need not be removed to reach the adjustment sleeve. 3.4 ELECTRICAL CONTROL CIRCUITRY The system is operated and controlled electrically fror the aircraft. The two circuits available to operate the system are: (1) the arming circuit and (2) the fire circuit, pickle circuit, or dissemination circuit (in the PAU-8/A it will be referred to as the dissemination circuit). The system must be capable of being armed and disseminated through electrical commands from the pilot. The electrical c'.ntrols intorfaco with tho pneumatic system through the arming sol'j.'Vi! •:, 'iissomination solenoid, and pressure switch. The sys'.or rust also Le capnblo of several different configurations cf sequential dissemination. To provide these configurations, a pressure switch and a mode selector switch are required. To provide safety, an arming switch (safety switch) and an arming relay are required. The system operates on 25 VDC with a current drain of 1.06 arps per module when armed and disseminating. The power breakuowr. is as follows: Dissemination Solenoid Arr.ing Solenoid Arr.ir.o Relay Coil 0.5 amps 0.5 amps 0.06 ar.ps If all four modules are disseminating simultaneously, the total current drain is 4,23 amps. The longest drain with all four modules disseminating at orce is approximately two ninutes (25 sal/min for 50 gallons). Therefore, the largest drain on the aircraft power would be 0.141 ampere-hours per frur-rodule configuration. The 4.24 amps per four-r.oc.ule configuration is 39 �NOZZLE SETTING NO. ! X -Nf , p 1 \ "& ' „__ I 1 \ NOZZLE SETTING NO. Figure 19. Production Nozzle belcv the 5 ar.ps maximum allowable on some aircraft stations. A rode selector switch is mounted on each junction box. Hcvever, when r.ore than one module is used, only the mode svitc;-. in V.o.iule No. 1 is operable; the other rode switches have r.c controlling function. Mode position Tio. 1 disseminates rcaules sequentially; top, left, right, then hotter, . o e "d ccsiticr. \ 6 2 disseminates modules tcpand rioht simultaneously, '. t.-en .eft and bettors simultaneously. Mode position No. 3 disser.ir.ates modules top, right and left simultaneously, then botton. Mode position \ o •! disseminates all modules simultan'. eously. Table II sl.ows other modes of operation for the twoand three-module configurations. The electrical system operates in the following manner: When' the red-flagged* arning pin (REMOVE BEFORE FLIGHT) is rerovec, it allows th. two poles on the arming switch to close. One pole permits ar: ing, the other dissemination. Actuation �of the arr.ir.g switch by the pilot provides 28 VDC power through the arminq switch (safety switch) and the pressure switch'to the arming solenoid, which operates the arming valve. It also energizes the arming relay coil. Activating tre arming relay coil closes a set of contacts, making dissemination possible. Before the circuit can be armed, the pressure switch r.ust be in the high position which only occurs when the pressure in the high-pressure gas container is greater than 200 psi. When the pilot depresses the dissemination butcon, 28 VDC is supplied to the dissemination solenoid through the arninc switch and the arming relav contacts. V.'hen the tank is fully disseminated (pressure below 200 psi), the pressure switch supplies power to the next module for dissemination. The pilot ray start and stop dissemination by releasing and depressing the pickle button. Three plugs are located on the junction box. One connects the module to the circuit, another connects other modules together electrically, and the last is a test plug. Also mounted in the junction box are two zener diodes. These diodes allow parallel operation of the arming valve and relay at normal operating voltage of 28 VDC, but will isolate these components for continuity testing at 6 VDC. 3.5 CENTER SECTION (AGENT TANK) The center section (Figure 20), basically a 13-inch OD tube connecting the fore and aft bulkheads, provides the following r.ajcr functions: * Attaches to the bomb racks * Is the major strength component * Contains the agent * Contains two fill ports * Attaches tc the module adapter * Accepts up to four stabilizing fins * Provides mounting for all other hardware (pneumatic sy?tem, disseminate mechanism, etc,) Paragraph 3.5.1 gives the design philosophy for each subcomponent in the center section and paragraph 3.5.2 summarizes 41 �FILL PORT FILL CAP AFT BULKHEAD INNER TANK STRONGBACK BAND SUPPORT BULKHEADS DOUBLES ELECTRICAL CONDUIT FORWARD BULKHEAD Figure 20. Tank Assembly (Agent Tank) �tho flow tests which wore conduct ml to optimise ."Mont flow itynarics witliin Lite cor.tor sfetion of the modulo. Sootior. IX discusses the fit tests and wind tunnel tests which al#o ir.fluer.ceti the design. 3.5.1 Component Parts The center section (agent tank assembly) consists of the following major components (Figure 20): • Strongback (forged) • Skin (13-inch OD tube) (rolled sheet) • Forward bulkhead (drawn sheet) • Aft bulkhead (cast) • Bank support bulkheads (cast) • Inner tank (formed sheet) • Fill ports and caps (wrought aluminum; Nylon 6/6, 40 percent glass; stainless steel) • Pick-up tube (formed aluminum tubing) • Stainless steel pipe • Electrical conduit The design philosophy for each of these components is explained in the following paragraphs. All components except parts of the fill ports and caps are made fron aluminum to keep hardware weight down commensurate with strength requirements. Alloy 5083 is used exclusively for all forged and wrought components; 5083 provides an excellent corJbination of strength, impact and fatigue resistance, we3 debility, and resistance to corrosion and stress cracki". Cast components (aft bulkhead, bank support bulkheads) are made fron either 356 or 357 aluminum casting alloy to provide high strenqth and light weight. 3.5.1.L Strongback The nodule Strongback incorporates both 14-inch and 30-inch 43 �lugs, has an electrical cavity aft of the rear 30-ir.ch lug (to .accept pylon unbilical cord), and provides bearing area for the sway.ijrace pads of the following bomb racks: • MK51 • - F-100/Type I • F-100 'Type III • !!AU-9/A and MAU-9A/A • MAU-12B/A • F-105/Multi-Weapon Adapter • F-105/Universal B/D Pylon Tiie hardware delivered is compatible with these racks. The sway Jbrace pads vv'ere riot made compatible with MIL-STD-8591C until after design approval; therefore, the hardware is compatible with the racks only. The strongback is also able to withstand the high-lug and sway-brace forces which are generated with the systen in the four-module configuration. The stronqback could have been either forged, cast, or rolled and machined from thick plate; however, rolling and machining is costly for any production quantity. While both casting and forging provide excellent design flexibility, forging produces a more integral part which requires considerably less inspection and, therefore, reduces cost. Because of this, forging was the fabrication method selected. Threads are cut directly in the strongback to mount the 14-inch and 30-inch lugs. Rigorous testing shewed that this aaethod of lug mounting provided excessive strength while keeping costs and weight at a minimum. Separate external highstrength steel pads are bolted to recesses in the strongback to provide adequate sway-brace pad-bearing resistance and to alter the basic 13-inch OD store configuration to allow mounting cf the store to the seven specified bomb racks. 3512 ... Skin The skin is a 13-inch OD tube which mates with the strongback and circumscribes th£ fore, aft and band-support bulkheads. 44 �It must withstand 55 psig internal operating pressure and all imposed environmental loadings during handling and flight. The skin is made from rolled and welded 5083-H323 aluminum, 1/8inch thick. Rolling and welding was selected because it is a far less expensive method of forming a tube with irregular cutouts. 3.5.1.3 Bottom Doubler The bottom dcubler strengthens the underside of the store, providing a cradling area for handling. MIL-STD-8591C dictates dcubler size and strength. The doubler is adequate to support a single nodule on forklifts, but the loading and handling fixture must be used for the two-, three-, or four-module configuration. The doubler is made from simple rolled rectangular plates, fillet-welded to the module skin. Three plates are used to provide locating recesses for the mating adapter bands. 3.5.1.4 Forward Bulkhead The forward bulkhead closes the front end of the center section and provides mounting support for the pressure sphere and all pneumatic hardware, electrical tubes, bleed tubes and nose cone. The bulkhead can either be cast, forged, machined from stock, or drawn from sheet. Castina and forging provide excellent high-strength components; however, both require secondary machining operations and the resulting parts are too heavy. Machining from stock is excessively costly for large components such as the forward bulkhead. Drawing the front bulkhead fron sheet aluminum provides an inexpensive part (requires little machining) which will meet all strength requirements. 3.5.1.5 Aft Bulkhead The aft bulkhead closes the rear of the center section and provides support for the fins, aft fairing, dissemination valve and electrical tube, and incorporates a pressure gage port (for checking tank pressure), and drain plug. Because of the high fin loading, a bulkhead drawn from sheet aluminum cannot be used. Forging and casting are the major alternatives, both of which require many machine operations. Forging requires more costly tooling for small quantities and long lead time on the first order, but provides a part which requires little inspection. Casting provides a part with less strength but since it 45 �is adequate for the aft bulkhead and less expensive than forging, the aft bulkhead was cast from 357 aluminum. Bani-Support Bulkheads 7;-.e bar.i-support bulkheads are located inside the center section skin directly under the mating adapter band strars. The bulkheads are fillet-welded to the strongback, the assembly fitted inside the skin, and the bulkheads plug-welded to the skin. The band-support bulkheads must be able to withstand the band and mating adapter loads. These bulkheads also form the ends as well as house the drain plug for the inner tank. Casting was selected as providing the least expensive, most applicable fabrication process because of the thin dip webs and complicated shape. 3.5.1.7 Inner Tank The inner tank is essentially an elliptical tube welded between the band-support bulkheads. This inner tank provides a settling chamber and is used to optimize agent flow dynamics. Agent is transferred to the inner tank from between the bandsupport bulkheads of the main tank, picked up (pick-up tube) inside the inner tank, and fed to the disseminating mechanism at the aft end of the store. Aluminum sheet (5083) is ferried and welded to form the tank. The transfer tube and the pickup tube are formed from standard tubing and welded in place. 3.5.1.8 Pick-up Tube The pick-up tube passes through the center of the module and runs from the inner tank to the aft bulkhead» This tube carries the agent from the inner tank to the dissemination valve at the aft bulkhead. The tube is welded to the aft bandsupport bulkhead and to the aft bulkhead. This tube is made from aluminum tubing bent at one end and swaged at the other. The tube was swaged to match the inside diameter of the dissemination valve. 3.5.1.9 Standpipe The standpipe, made from rectangular aluminum tubing, and weldod to the bottom of the inner tank, carries the agent from the main tank to the inner tank. 46 �3.5.1.10 Electrical Conduit The electrical conduit is a small aluminum tube which passes fron the front bulkhead to the umbilical in the strongback to the aft bulkhead. This tube is used to run the electrical control wires and pneumatic lines. 3.5.1.11 Fill Ports and Caps Fill ports are recessed into the skin near the fore ard aft bulkheads to accept the filler caps and nipples (Figure 21). The nipples are of the flange type, utilizing a port-type seal (o-ring) and are made entirely from Nylon 6/6, 40 percent glass-filled. This nylon is compatible with the various agents and enables the nipples to be injection molded for low production costs. The quantity produced under this contract did not warrant the cost of tooling to have them injection molded and were, therefore, machined from extrusions. Screens are incorporated in the nipples to limit the size of foreign matter which may enter the module. The screens can be removed from the nipples by simply removing a snap ring. The cap is a ball-lok type with a pressure locking device, preventing cap unlocking when internal tank pressure is over five psig. The cap is a custom design and uses Nylon 6/6, (40 percent glass) for most components. Where required, 300 series stainless steel is used (ball bearings, etc.). The nylon caps ana nipples are much lighter than metal caps and nipples. Color ceding on the cap indicates when the cap is locked or un-> locked and a note on the cap explains the color coding. 3.5.2 Coating Material The severe corrosion of aluminum requires that it be protected (through coating) from contact with the herbicide agents, particularly agent Blue. Coatings must also withstand the flexing of the dispenser under operational conditions. Those possessing physical properties which permit elor.ga-irr. in excess of 100 percent are able to withstand the fiexir.g. The more brittle coatings, such as the phenolics, crack under these conditions, and, while polyurethane possesses suitable physical properties, agent Orange causes polyurethane material to swell. Silicone coat:.ngs exhibit excellent physical properties and the herbicides do not cause the fluorosilicone dispersion coatings to swell. Several fluorosilicone dispersion coating materials were considered but none had all of the 47 �.Si****cr«A^jf "si^^^SiS'CC1" "<* jS&*"'k* ***'***^' *V'*tl"' *5»>*sV: .iFr*C.^' ;5'.iiv •fuxvr.. tT"\!r"*iJitt' 1 w ) ^< Figure 21. Nylon Cap and Nipple �properties desired. Most of the materials had the desired physical properties but were not suitable for coating the irside of the tank since they required the use of a primer which must be applied in a thin even coat. This could not be done on the inside of the closed tank. Dow Corning Corporation developed a fluorosilicone in conjunction with agent material compatability studies conducted during this program. To insure proper coating, a 13-inch diameter by 10-inch long module section with bulkheads was made with removable ends. This section was coated with the fluorosilicone material and inspected to determine methods of applying the material. From these coating tests, it was determined that rotation about two axis would be required to get a complete coating of the inside of the module-. 3.5.3 Agent Flow During the first part of the program, ground flow tests were conducted with the GFE TMU-66/A modules. These tests re-4 vealed that the pick-up tube system began to suck air around three-fifths empty when the tank was continuously discharged at flow rates of 150 gpm. The cause could not be determined from the modules and, to further study this phenomena, a full* size plastic flow model was built (Figure 22) around the design of the GFE. This design used a tube near the aft end to pick up the agent and pipe it to the dissemination mechanism. A bulkhead with flapper plate, slightly forward of the pick-up tube (Figure 23) was employed to keep the agent near the pickup tube (for a short time) during aircraft deceleration or a nose-down condition. Although the flapper plate performed its intended function, it also caused the system to exhaust its air supply through the pick-up tube while considerable agent was still in the store. This is cnaracteristic of a flapper plate since a differential fluid head across the plate is required to cause agent flow through the plate opening (equal air pressure exists on both sides of the plate). When the aft (or lower) fluid level drops near the pick-up tube opening, air at 55 psig flows freely out the pick-up tube, exhausting the internal air supply with substantial agent remaining in the store. Figure 22 shows this differential head in the flow model at a flow rate of approximately 90 gpm. The differential head at 150 gpm is approximately twice that shown. One additional drawback to the flapper plate design is that it employs a moving component, and the highly corrosive nature of the agents would cause sticking of the flapper after short periods of use. 49 �o Figure �FLAPPER PLATE SLOPE DUE TO OPEN CHANNEL FLOW 55 psig BULKHEAD WITH FLAPPER PLATE AFT FORE FLUID FLOW Figure 23. PICK-UP TUBEModule With Flapper Plate 51 �The first alternate system investigated employed a sealed bulkhead and used a standpipe type transfer tube as shown in Figuras 24 and 25. The standpipe design works on the principle that any agent (or air) which passes through the picJ.-up tube ~ust first pass through the standpipe. Thus, the aft chamber (around the pick-up tube) remains full until the fore chamber is nearly empty. In addition, as, the aft chamber drains, the sr.ar.dpipe continues to scavenge the forward section of the r.cdule, almost draining it of agent. To allow complete filling, both the fore and aft chambers must be vented during the filling operation. Figure 25 shows the aft chamber full while the fore chamber is only about half full. The major problem encountered with the standpipe bulkhead concept was eg shift. Testing indicated that approximately seve.i gallons capacity in the aft chamber would be required. This will cause a eg shift of about 15 inches when the aft chamber is full and the fore chamber empty. Since only a ±3-inch c;; shift can be tolerated (MIL-STD-8591C) , further design improvements were investigated. A seven-gallon separate chamber was designed to fit between the band support bulkheads. This general design was used for extensive testing before the final design (Figure 26 and 27) was arrived at. The first standpipe used was round, which allowed the formation of vortices and passed excessive air into the settling'chamber. The rectangular tube standpipe eliminated the vortices . Another problem which occurred in the settling chamber was turbulence, which mixed the air in the settling chamber with the fluid and then forced the mixture out the pick-up tube. To reduce the turbulence in the area of the pick-up tube, the anti-turbulence bulkhead was placed in the settling chamber. This keeps the turbulence in the front of the settling chamber and away from the pick-up tube. The tests performed on the final flow model incorporating the seven-gallon inner tank showed total fluid left after dissemination is under one pint at all fj.ow rates (25, 75, 150 gpm) . Estimated fluid in the entire model when the disseminate valve first passes air is about one quart. 'Die central-tank settling chamber performs well, providing controlled dissemination for about 99.75 percent of the store's agent capacity. 3.6 NOSE CONE The nose cone acts as a wind screen covering the forward v>iu ur.v.t ic- vlovicvs and mates with the forward bulkhead. It is �STANDPIPE BULKHEAD SLOPE DUE TO OPEN CHANNEL FLOW — PtCK-UP TUBE FORE =3 AFT FLUID OR AIR FLOW- Figure 24. Module With Standpipe Bulkhead 53 �I'THl Figure 25. Flow Model With Standpipe Bulkhead �Ul Figure 26. Flow Module V7ith Central Settling Chamber �BAND-SUPPORT BULKHEADS (STRONGBACK NOT SHOWN FOR CLARITY) TUBE FOR FILLING ANT I TURBULENCE BULKHEAD ..CENTRAL SEVENGALLON TANK TO DISSEMINATOR ui SLOPE DUE TO OPEN CHANNEL FLOW AFT FORE Figure 27. Central Settling Chamber �held in place by a single bolt at the apex of the cone which preloads it against the forward bulkhead/skin assembly. This single bolt technique is used to provide an easy method of removal, since the nose cone will be removed and replaced numerous times during the module life. The nose cone is spun from 6061-0 aluminum alloy and heat treated to the T6 condition for scrength. Metal spinning is the least expensive and most proven method of fairing fabrication for the quantity required by this program and a metal fairing meets all strength and functionability criteria. Other processes such as fiberglass layup, filament winding, etc., are far too expensive for the small weight advantage. A plate is welded to the apex for the nose bolt. A bolt rather than a stud with a nut was used so that there would be no exposed threads which could become damaged. 3.7 TAIL CONE The tail cone is a hemisphere of spun aluminum used to cover the dissemination valve and nozzle assembly. It is attached to the nozzle assembly by four screws and serves as support for the nozzle assembly. The forward edge of the tail cone engages the aft bulkhead. 3.8 FINS The purpose of the fins is to provide stability in the event of store jettison. Two fin sizes (a 10-1/2-inch span and a 16-inch span) are required for compatibility with the four different module configurations and all carrier aircraft. Fin chordwise profile was minimized, commensurate wit:-, strength considerations. Aerodynamic characteristics of the 5tores. w j e j t i j ; ; J i er_^cr^^L'^ "vTous program. Wind tunnel tests and jettison tests (Section IX) on the two-module configuration were conducted during this program to design fins and fin configuration for a stable two-module configuration. The two-module configuration was considered to be unstable under the previous program. Fit tests (Section IX) on the aircraft involved in this program were conducted to verify the paper study on the clearance of the fins. Numerous manufacturing techniques have beer, considered for both metal and plastic fins. 57 �• Metal forming and welding (various configurations) • Die casting and welding (metal) • Injection molding and ultrasonic welding (plastic) • Rotational molding - foam filling (plastic) • Fiberglass layup - foam filled The type of manufacturing process is obviously dependent upon the type of fin material. Metal fins are strong and can be made with proven techniques but are excessively heavy and costly. Plastic fins are relatively new but offer the following advantages: • Very low cost • Extremely lightweight • Xonconductive and resistant to defoliant agents • Will readily grind off without sparking if they contact the runway during hard landings. Because of these advantages, considerable effort was made to develop a plastic fin. Rotational molding was selected as the meth<"d of fabrication because of the low cost of tooling and parts for prototype and limited production. Ir, addition, rotational molding produces an integral fin without seams. Nylon, glass filled nylon, glass filled celcon, and polyethylene were used to fabricate sample parts. The nylon was considered too brittle for handling in case it was dropped on a corner. Glass filled nylon had flow problems in the mold and did not produce satisfactory parts. The glass filled celcon exhibited porosity and low strength because of the porosity. The polyethylene fins filled properly and exhibited strength which was considered adequate for the design loads. Polyethylene fins were fabricated and delivered to the Air Force for flight testing. The polyethylene fins failed when tested on the F-4 aircraft at 550 knots. Information obtained after the flight test indicated that the design loads were too low because of the irregular flow under the wing stations of the F-4. a Davis, Ronald E., Flow Field Characteristics Beneath the F-4C Aircraft at Mach Numbers 0.50 and 0.85. Arnold Engineering Development Center, Arnold Air Force Station, Tennessee, AEDC-TR-70-8, February 1970, Unclassified. 58 �Because of the lack of. time to further carry out the development of the low cost plastic fins, the effort was discontinued. A fiberglass-foam filled fin with the same external, configuration was designed, fabricated, and tested. Figure 28 shows the load being applied with a foam pad to simulate aerodynamic load. The fiberglass-foam filled fin was about five times as strong as the polyethylene fin. 3.9 MATERIAL SELECTION The process of material selection for this system was difficult, not only because it must be designed to withstand che action of more than one chemical agent, but also because these agents (Agents Orange, White/ and Blue) belong to different chemical families. Since one agent is a mixture of an inorganic salt and an inorganic acid, the materials, particularly the metals, must be selected to resist attack from this acid-salt combination. The plastics and rubbers must not only resist this acid but also the solvent, swelling action of the other agents which are organic compounds. Many plastics and rubbers will meet the first of these requirements but will not meet the second. Therefore, it was necessary to test each material for its resistance to all three agents. 3.9.1 Aluminum The aluminum alloys 5083 and 5086 were considered for use in the PAU-8/A module. The 5086 alloy is less corrosive resistant than the 5083. Both alloys were subjected to t:-.e ccrrosive effects of the three agents at ambient temperature ar.r. at 130°F for several weeks. Agents White and Orange have lit_l= corrosive effect on aluminum; however, it is heavily attacked by Agent Blue. Agent Blue attacks unprotected 5083 alloy at the rate of one mil per week and 5086 alloy at three mils per week at 130°F. The rate of corrosion at ambient (77°F) is about 10 percent of the rate at the elevated temperature. These results indicate that 5083 alloy should be chosen over 5086 and that it must be protected with a corrosion and solvent resistant coating when exposed to the agents. 3.9.2 Stainless Steels Grades 304 and 316 stainless steel resisted the corrosive effect of the three agents at ambient temperature and at 130°F for several weeks without visible corrosion. Microscopic examination of the metal surfaces indicated that no corrosion and 59 �: • • • % i I 1 £•«•*.::• :> • ^2:^ ; 'itli fesf' !• -m^;-€^ v*V«i- i<s";;" :.•«•?''«*'i'-i^O Kii;'> -.*.. ' Figure 28. Fin Testing Fixture 60 �very little film formation had occurred. These steels should withstand tine corrosive action of the herbicide agents for extended service. 3.9.3 Plastic Nylon. Nylon resists a wide range of organic and inorganic substances. It is not affected by, nor does it affect, lubricating oils and greases, aliphatic and aromatic hydrocarbons (including the conventional fuels), or thr common esters, ketones, ethers and amides. It resists most inorganic reagents, and unlike many metals, it is not affected by electrolytic corrosion. Zytel 38 has the highest acid resistance of the nylon series. Since Nylon 6/6, 40 percent glass filled, resists solvent swelling and acid attack by Agents Orange, White, and Blue, it was selected for use in the construction of the caps for the fill ports. Table IV summarizes solvent retention of the nylon. rASLE IV. AGEIJT ABSORPTION OF NYLON 6/6 40 PERCENT GLASS FILLED PERCENT SOLVENT RETENTION AGENT WEIGHT PERCENT Deionized Water 0.30 Agent W h i t e 0.79 Agent Blue 0.34 Agent Orange 0.25 61 �Polypropylene. Samples of the polypropylene 20 percent glass filled underwent testing in contact with Agents Orange, V.'hite, and Blue without any effects of swelling or acid attack by the agents. 3.9.4 Rubber Gasket and O-ring Selection. Neoprene and Buna rubbers normally used as gasket and O-ring materials are severely swollen by Agent Orange with ultimate loss of the mechanical properties. Fluorocarbon, Viton1, and ethylene-propylene rubbers were testud for loss of mechanical properties and swelling in the three agents. Tensile sample bars and volume swelling samples wer« immersed in each agent for 72 hours at ambient te.'perature. Ethylene propylene is the only rubber which could withstand the solvent swellinci of Agent Orange. Viton® and acid resistant Viton®both swelled to about 20 percent volume increase. Agent Blue blisters the latter two rubbers but does not. in general, alter their mechanical properties. Ethylenepropylene rubber was selected for gasket and O-ring application because of its resistance to solvent swelling. Table V summarizes the test results of these materials. Nozzle Diaphragm. This application requires a high performance rubber with high elongation (>500 percent)« Neoprene can be obtained in elongations above 500 percent but tests indicate that it is decomposed by contact with Agent Orange. Silicone rubbers have a high resistance to solvents. Fluorosilicone rubbers undergo solvent swelling to only about 10 percent increase in volume, whereas tne silicone rubbers swell to over 150 percent. A fluorosilicone rubber (Dow Corning LS-2332V) was selected for use due to its superior resistance and high (500+ percent) elongation. 3.9.5 Coating Material Fluorosilicone was chosen for the coating material for the inside of the agent tank because of its elongation and its ability to withstand swelling when exposed to herbicides. Fluorosilicone coated aluminum samples underwent long term explosure to the three agents at temperatures of 130°F without any change noted. Bond peel tests were conducted with the fluorosilicone coating. Samples were prepared as follows: • 1 Solvent cleaned and degreased Trademark 62 �TABLE V. SAVPLE MATERIAL MECHANICAL PROPERTIES OF RUBBER MATERIAL IMMERSED IN HERBICIDE AGENTS FOR SEVENTY-TWO HOURS AT AMBIENT TEMPERATURE TENSILE (psi) AGENT ELONGATION (PERCENT) HARDNESS (OURO) "A" VOLUME SWELL (PERCENT) As R e c e i ved 2,354 225 73 Orange 2.262 275 62 22.94 Blue 2,174 250 66 10.24 White 1.750 250 70 7.70 E t h y 1 ene As R e c e ived 2.345 200 76 -- P r o p y 1 ene Orange 2.267 225 75 3.92 Blue 2,191 200 77 0.22 2J53 200 76 1.00 V i 1 0n ^ U) White i | Acid- As Received 2,205 350 62 -- Res i s t a n t Orange 2,130 275 56 18 90 Viton® Blue Win te 2.059 300 62 4.75 1.945 400 61 5.64 �• Sand blasted • Solvent cleaned, degreased, .and primed • Sand blasted and primed (50:50 mixture Dow Corninq 1200 primer and Naptha) . The mixture was used because of "chalking" of the primer at high concentrations. The sample preparation list is given in increased peel strength. The solvent cleaned and degreased surface Xi?as the lowest strength but was considered to be strong enough. 64 �SECTION IV MODULE ADAPTER DEVELOPMENT 4.1 MODULE ADAPTER REQUIREMENTS The module adapter must: • • Be capable of carriage in configurations using two, three, and four modules; the number and combinations of modules to be carried on each station on each aircraft to be determined by the weight capability of that station; any combination of modules to be readily and quickly filled and serviced on the ground or when suspended on the aircraft. • Afford maximum usage of payload capacity for each aircraft. • 4.2 Be capable of carriage on the F-4, F-100, F-105, and A-l aircraft with consideration also given to the F-lll A-7, and A-26 aircraft. Be capable of carriage at speeds of up to 1.5 Mach at 35,000 feet and up to Mach 0.9 at sea level. ANALYSIS OF THE PROBLEM For the previous effort, the module adapter had two primary functions. First, the adapter provided a method of attaching the TMU-66/A modules in the required three- or four-module configurations (a two-module configuration was not a requirement at that time) which allowed up to four modules to be mounted on d single bomb rack. Second, the adapter provided housing for an electrical junction box which controlled the module dissemination sequence. During design analysis, determination was made that the individual module was a more suitable place for the central electronics, and the second adapter function was dropped. Thus, the objectives of the mating adapter design became: • A low-cost, functional, and reliable method of attaching the PAU-8/A modules in the required configurations, • Structurally capable of supporting the modules in all configurations, and • Lightweight. 65 �4.3 .SYSTEMS ANALYZED s The module (mating) adapter consists of two main parts, the supporting structure and the attaching mechanism. Various designs were considered in this program. To keep weight and corrosion at a minimum, aluminum was selected for all supporting structures. The five supporting structura designs considered (Figure 29) were as follows: (1) A simple (continuous) extrusion (2) Three castings or forging with exterior plates riveted or welded in place (3) A large central square mechanical tube with formed supports (4) Three main castings or forgings welded to a section of internal mechanical tubing (5) Two main castings or forging^ welded to a section of internal mechanical tubing. The methods of attachment considered were (1) modules bolted directly to the support structure,(2) modules connected by latches or cams, and (3) modules secured to the support structure by straps which passed around each module. A compressed vertical-spacing module-adapter configuration was also investigated. A comparison of the normal spacing and the cor.pressed vertical spacing is shown in Figure 30. By cor.pressing the vertical spacing so that only one inch separates the upper and lower modules, the overall depth of the configuration was reduced some 5.5 inches while the width was increased by 7.5 inches. Since the primary purpose for considering the compressed configuration was to increase the number of modules that could be carried on certain aircraft, as well as easing loading problems, a comparison was made of the maximum module-loading capacity of the specified aircraft. Table VI indicates the maximum number of modules that could be carried on these aircraft, utilizing wing stations only. Physical and weight compatibility were controlling parameters. Multiple pylon loadings were evaluated to check store-to-store clearance in determining maximum aircraft loadings. The only aircraft affected by store-tostore clearance considerations was the A-7; however, some aircraft store jettison would have to be accomplished in a set pattern to avoid store-to-store collision during separation. 66 �(I) CONTINUOUS EXTRUSION (2) CASTINGS WITH EXTERIOR PLATES [' (3) LARGE SQUARE TUBE WITH [I IK FORMED SUPPORTS (4) THREE BULKHEADS WITH CENTRAL TUBE Figure 29. (S) TWO BULKHEADS WITH CENTRAL TUBE Module Adapter Designs 67 �NORMAL SPACING o\ CO COMPRESSED VERTICAL SPACING Figure 30. PAU-8/A Multiple Module Configurations �TABLE VI. AIRCRAFT PAU-8/A MAXIMUM LOADINGS MAXIMUM NUMBER OF MODULES PER AIRCRAFT NORMAL SPACING | COMPRESSED VERTICAL SPACING F-100 6 io a 6 F-4 12 IB 3 F-111 (26° Sweep) 32 3 32 3 F-111 (72.5° Sweep) 123 123 A-1 2 8b A-7 20 LOSS 6 F-105 GAIN 4 4 vo A-26 a 4a Requires Reduced Agent Loading (Slight O v e r l o a d ) Requires Reduced Agent Loading, Plus Acceptance Of 0.5 Inch Ground C l e a r a n c e ( W o r s t Case) 6 16 4a 4 . �Using compressed vertical-spacing would cause both gains and losses in maximum loading capabilities, depending upon the aircraft involved. Other advantages and disadvantages of the compressed configuration have been determined, and the overall evaluation, based on present studies, is as follows: Advantages of Compressed Vertical Spacing: • • • Increases F-4C/D maximum loading by four modules per aircraft Eases loading problems on the F-r4C/D Increases A-IE maximum loading by six modules, if 1/2inch ground clearance (worst case) is acceptable. Disadvantages of Compressed Vertical Spacing: • Reduces F-105 maximum loading by four modules • Reduces A-7D maximum loading by four modules • Increases difficulty of filling lower module (two-and four-module configurations) • Negates multiple-module wind tunnel data generated during previous effort. Since the disadvantages outweigh the advantages, normal spacing was retained for development. 4.4 DESIGN CONSIDERATIONS The five support structure designs were compared on the. basis of the mating adapter design objectives. The results are summarized below: (1) The single extrusion method would have fewer parts but would be extremely heavy compared with the other designs. An extrusion with multiple hollows would reduce weight, but tooling costs would be extremely high. (2) Casting or forging three bulkheads and attaching exterior plates would result in a heavy part, and fabrication would be more costly than the other designs. (3) The large central square mechanical tube with formed supports would be lightweight, but fabrication costs would be almost as high as in the second design. 70 �In addition, the square tubing would have to be custom excruded which would increase the cost above Design No. 2. (4) Three cast or forged bulkheads welded to a section of internal mechanical tubing would combine low weight and low cost, but the design itself would not be as satisfactory as other designs. (5) Two cast bulkheads welded to a section of internal mechanical tubing would be the lightest and cheapest of all designs analyzed. Cast bulkheads would be slightly less expensive than forged bulkheads and would still possess the required strength. The round central tube-would provide sufficient bulkhead support with minimal cost and weight. Of the three attaching mechanisms analyzed, the strapping method was the most satisfactory. Since the latch or cam-type attachment would require module and mating adapter reinforcements to decrease localized stresses caused by this type attaching mechanism, the overall cost of the strap would not exceed that of latches or cams. In addition, the strap with its accessible take-up bolt would provide the easiest type of attachment. 4.5 SYSTEM SELECTED The selected design consists of two cast bulkheads ccr.r.sc.ed with a round central tube (Figure 31). A strap surrounds each mounted module and is secured to the bulkhead with quickrelease ball-lok pins. A take-up bolt is used to tighten the bands. The forward bulkhead contains four locator pins which support forward and aft module loadings while simultaneously locating each module when mounting. Aluminum-silicon-magnesium alloy 357 provides high strength cast bulkheads with good capability. The tube is standard 6061 aluminum stock, a low cost and weldable material for easy adapter fabrication. The high strength stainless steel straps, trunnions, and ball-lok pins in the support assembly provide a functional, reliable method of module attachment. The steel take-up bolt has electrolytic nickel-coating to prevent corrosion and to minimize wear. 4.6 PRODUCIBILITY ANALYSIS Fabrication can be accomplished through established 71 �BULKHEAD CENTRAL TUBE REMOVABLE STRAP Figure 31. Module Adapter manufacturing procedures. The two sand-cast bulkheads are identical except that the forward bulkhead has one additional machine operation in which the holes for the dowel pins are machined.. The dowel pins are press fitted to the bulkhead and each bulkhead is then fillet welded to the central tube. The band, formed from standard size strap and spot welded around machined trunnions, is designed for easy production. The balllok pin and take-up bolt aro both stock items. 4.7 TESTING AND MODIFICATION During acceleration tests conducted at Picatinny Arsenal, Dover, New Jersey, under the direction of Armament Laboratory personnel, the module adapter failed at 9 g. The failure occurred in the forward bulkhead casting at the top dowel pin and mode of failure was tearing of the edge. To overcome this weakness, a one-inch thick 5083 aluminum plate was welded to the forward face of the forward bulkhead and a longer dowel pin vis used in the bulkhead to increase the bearing area of the cir.. Later structural tests conducted at Wright-Patterson Air Fcrc-e Base, Ohio, indicated that the reinforced bulkhead was r.ore than adequate. 72 �SlsOTlON V DISPENSER TKST UNIT DEVELOPMENT 5.1 REQUIREMENTS The dispenser test unit provides a functional check of the r.odules and aircraft arming system prior to takeoff, 5.2 DESIGN OBJECTIVES The design objectives for the unit were as follows: • Capable of checking the major circuits in the system and the connections between thess circuits. • Capable of checking the electrical power of the aircraft at the;pylon. • Capable of arming the system and operating the dissemination circuit both before and after the system is loaded on the aircraft, and after the system is connected to the aircraft circuitry. • Capable of checking one, two, three, or four modules sequentially during any one test. • Easily operated, compact, portable and rugged. • Reliable and relatively maintenance-free. 5.3 DESIGN' The test unit is self-contained. Batteries are utilized to supply 28VDC for module functioning and a 6-volt battery is used for continuity checking. This provides the capability for complete evaluation of a fully-charged tactically-ready PAU8/A. Rechargeable lead-dioxide batteries were selected for use because of their long shelf life, wide operating temperature range, high cycle life, and low cost. The test unit requires five inputs from the items being checked: one from each module and one from the aircraft. These must be plugged in before the start of each test. These inputs are provided by a 15-foot cable connecting the test unit to the dispenser and the aircraft. The red-flagged arming pin must be pulled from each module before testing. After pulling the arming pin, all of the indicator lights on the test unit must be tested to ensure proper functioning. The lights are of the press-to-test type. 73 �The test unit has ten test modes for checking the continuity of the control, arming, dissemination, and ground circuitry and for checking switch contacts, arming and dissemination functions, the aircraft electrical system, battery voltages of both 6- and 28-"olt batteries, and battery charging requirements. The main selector switch is used to select the mode desired. "ode Or.e; Continuity The main selector switch is turned to the "continuity check" position. The continuity section consists of a locking toggle switch, a momentary switch, and a rotary switch. The rotary switch makes possible the testing of one, two, three, or four modules sequentially. After selection of the module to be tested, the "ready" switch is actuated to light six red indicators marked as follows: (1) Arming Switch Pole #1; (2) Arming Switch Pole =2; (3) Arming Solenoid; (4) Dissemination Solenoid: (5) Arming Relay Coil; and (6) Arming Relay Contacts. The "step" switch is then actuated. A motor steps this switch through eight positions in less than a second. In each of the eight positions, a continuity check is marie on a major portion of the control circuitry. If the circuit checked is continuous, the appropriate red indicator light goes out. Positions No. 1 and No. 2 check the continuity of the pressure switch circuitry. One of the amber lights marked "Pressure Switch Low" or "Pressure Switch High" must go on, indicating which position the pressure switch is in. Simultaneously, the remainder of the checkout circuitry is then automatically programmed for checkout in tne appropriate mode: low or high pressure. Position No. 3 checks the continuity of Arming Switch Pole No. 1, and Position No. 4 checks the continuity of Arming Switch Pole No. 2. If either of these two lights stays on, there is a discontinuity in that portion of the circuit. If both of these lights stay on, there is a strong possibility that the red-flagged arming pin has not been pulled. In this case, the pin must be pulled before continuing wich the rest of the test. Position No. 5 checks the continuity of the arming solenoid valve. If the circuit is continuous, the appropriate red indicator light will go out. 74 �Position No. 6 checks the continuity of the dissemination solenoid valve. If the circuit is continuous, the appropriate red indicator light will go out. t"-*.*^ Position No. 7 checks the continuity of the arming relay coil. If the circuit is continuous, the appropriate red indicator light will go out. Position No. 8 checks the contacts of the arr.ing relay for the possibility of a short. If no shorts exist, the appropriate red indicator light will go out. If any of.the lights stay on, the ready switch may be reset and the check may be made again by actuating the step switch. If the light or lights stay on again, the corresponding section of the circuitry must be repaired. The red light will stay on until the system is repaired, replaced, or until the test unit is switched to another mode. Before switching to another mode, the ready switch must be returned to the reset position. Mode Two; Arm Check The main selector switch is turned to the "arm check" position. All four green indicator lights in the arm check section should light sequentially as the arm check selector switch is ro.ated from 1 to 4. This check is an intermodular continuity test of the arming circuitry. The light numbers are directly related to the module numbers; for example, a continuity check of the arming circuitry traversing the interconnection cable and mating plugs and receptacles to module No. 1 will involve light No. 1, a check of module No. 2 will involve light No. 2, etc. If one or more of these lights fail to light, there is a discontinuity in the indicated circuitry. Mode Throe; Dissemination Check The main selector switch is turned to the "dissemination check" position. All four green indicator lights in the dissemination check section should light sequentially as the dissemination check selector switch is rotated from 1 to 4. This check is an intermodular continuity test of the dissemination circuitry. The light numbers are directly related to the module numbers; for example, a continuity check of the dissemination circuitry traversing the interconnection cable and mating plugs and receptacles at module No. 1 will involve light No. 1, a check of module No. 2 will involve light No. 2, etc. If one or more of these lights do not light, there is a discontinuity in the indicated circuitry. 75 �Mode Four; Ground Check The main selector switch is turned to the "ground check" pcsicicr.. The ;r>omentary switch in the ground check section is -;-.er. r.cvec. t<. the "check" position. All four green indicator lights should go on. This check is an intermodular continuity test of the ground circuitry. The light numbers are directly related to the module numbers. If one or more of these lights fail to go on, there is a discontinuity in the indicated circuitry. Mode Five; Switch Check This is a test of the continuity of the switch contacts and it also indicates which position the junction box mode selector switch is in. The main selector switch of the test unit is turned to the "switch check" position and the momentary swjtch in the switch check section is moved to the "ch-eck" position. There are four green indicator lights in chis section. If the junction box switch is in Mode No. 1, only light No. 1 will light; if in Mode No. 2, lights No. 1 and No. 2 should go on; if in Mode No. 3, lights No. 1, No. 2, and No. 3 should go on, and if in Mode No. 4, all four lights should go on. If any of the lights that should light do not, a discontinuity exists in the switch contacts of switch circuitry. Mode Six; Function In this mode it is possible to arm the PAU-8/A and to disseminate the agents in any of the usual configurations. The main selector switch is turned to the "function" position. The system may then be armed and the dissemination function activated by the toggle switches located in the function section of the test panel. Mode Seven; Aircraft Check Mode No. 7 is a check of the electrical system of the aircraft. The main selector switch is turned to the "aircraft check" position. When the pilot actuates the arm switch in the aircraft, the red arm light in the function section should light. When the pilot depresses the pickle button in the aircraft, the red disseminate light in the tunction section should light. If either or both lights fail to operate, a failure is indicated in the electrical system of the aircraft. 76 �Mode Eiaht and Nine; Battery Check Modes No. 8 and No. 9 are voltage checks of the 28-volt and 6-volt batteries, respectively. After the main selector switch is turned to the proper position, the voltmeter is checked for a reading to determine whether the battery needs charging. At full discharge, the 28-volt battery will read 26.25 volts and the 6-volt battery will read 5.25 volts. Mode Ten; Battery Charge If the meter indicates that either battery needs charging, the main selector switch should be changed to the "battery charge" rtode. In this mode, both batteries are put on charge until they register the proper voltage. When a battery is fully charged, the charging circuit automatically cuts off and puts the battery on a trickle charge. Two lights for each battery, located in the charging section, indicate which battery is charged or being charged. If any of the tests in modes one through six should give unsatisfactory results, the system shall be considered inoperable until repaired. 77 �SECTION VI LOADING AND HANDLING ADAPTER DEVELOPMENT 6.1 REQUIREMENTS A loading and handling adapter for the MJ-1 and MHU-83/E bomb lift trucks was required to load the PAU-8/A multi-module configurations on the aircraft. 6.2 DESIGN OBJECTIVES The design objectives for the adapter were as follows: • Must be compatible with the MJ-1 and MHU-83/E bomb lift trucks. • Must be compatible with all configurations of the dispenser (1, 2, 3, or 4 modules). • Must be capable of loading any of the required configurations on any station of the F-4, F-100, F-105, F-lll, A-l and A-7. Must aid in assembling the modules into 2, 3, and 4 module configurations. • 6.3 DESIGN The loading and handling adapter was made from aluminum for light weight during the ground handling operation of placing it on the bomb lift trucks. The adapter has a base plate (Figure 4) which can be mounted to the table of the MJ-1 bomb lift truck and to the standard fork-type adapter accessory of the MHU-83/E bomb lift truck (Figure 32). The height of the table of the MHU-83/E with a four-module configuration was in excess of an acceptable limit for the F-4 aircraft. As a result, it was necessary to use the bomb lift truck fork adapter in conjunction with the PAU-8/A loading and handling adapter base plate. Cradles are mounted to the base plate to support the modules (Figure 4). The centerline cradles support the one-, two-, and four-module configurations (Figure 33 and 34) and the outer' cradles support the three-module configuration (Figure 35). The bearing surfaces of the cradles are covered with rubber to cushion the modules and to prevent paint damage during handling. _ _ _ j ' • 78 �a. MJ-l Figure 32. b. MHU-83/E MJ-l Table and MHU-83/E Fork Adapter 79 �KHU-83/E TIE-DOVN STRAPS CENTERLINE CRADLE oo o HOLD-DOWN BOLT LOAD!KG AND HANDLING ADAPTER Figure 33. Two-Module PAU-8/A on Loading and Handling Adapter �LOADING AND HANDLING ADAPTER HOLD-DOM BOLT Fiyure 34. Four-Module PAU-8/A on Loading and Handling Adapter �OLTER CRADLES MHU-83/C TIE-OOIN STRAPS LOADING AND HANDLING ADAPTER oo ro MHU-83/E FORKS HOLD-DOKN BOLT Figure 35. Three-Module PAU-8/A on Loading and Handling Adapter �Since the MJ-1 and MUU-83/K tables arc lirdted in the amount they can be tilted, and the outboard station of the F-4 has a cant anule of 7-1/2 degrees and the A-1E has an incidence angle of 10-1/2 degrees, adapter blocks had to be provided for tiitina the modules on the adapter to accommodate these stations. Figures 36 and 37 show the adapter with blocks, in the A-IE and the P-4 loadina configurations, respectively. A mock-up of the loading and handling adapter was used v:iclv the MJ-1 bomb lift truck for the fit tests of"the F-4, F-100, F-105, and F-ill (Section IX). 83 �00 Figure 36. Loading and Handling Adapter With Tilt-Adjusting blocks for Use With A-l Aircraft �CO en i^m^m^'^mK'^ Figure 37. Loading and Handling Adapter With Tilt-Adjusting Blocks for Use With F-4 Outboard Stations �^SB- VII SHIPPING CONTAINER DEVELOPMENT 7.1 REQUIREMENTS ' • Shipping containers for the PAU-8/A must meet the following rejuirer.ents: • Capable of shipping the PAU-8/A in the four-module configuration. • Acceptable by common carrier for safe transportation at the lowest rate to the point of delivery. • Capable of withstanding storage, handling and reshipment with no degradation of the spray system. 7.2 DESIGN OBJECTIVES T'r.e shipping containers must be lightweight, with low cubic vclur.e, and must be producible at low cost. 7.3 DESIGN The following were considered in determining the most effective design for the four-module shipping container: • Whether systems should be shipped with fins attached or removed, • Whether systems should be shipped with the vertical centerline of the module assembly in the normal vertical orientation, or • Whether the systems should be shipped with the module assembly rotated 45 degrees from orientation. Shipping the PAU-8/A with the fins attached would require a very large and very heavy container; therefore, since the fins are easy to remove from and easy to attach to the modules, the container was designed for shipment of the system with fins rerr.oved but inclosed in the container. The 45-degree orientation was chosen (Figure 38) for weight, volume, and cost savings. By rotating the four-module configuration 45 degrees from the normal orientation, the volume of the container was 20 percent less than that required for the system shipped in the upright orientation. Since the structural members in such a container are of smaller cross section. 86 �Figure 38. Shipping Container with Cover Removed 87 �the weight savings would be approximately 75 percent. The 45degree rotation causes no problem in loading and unloading the system since the four-module assembly can be picked up from the lugs in a side module. This rotates the four-module configuration to about the 45-degree position for loading. The modules are supported in the. container by txvo saddles at the band-support bulkheads. The container is completely inclosed and the fins, the fin attachment bolts, and the lugs are in separate compartments within the container. The container is of the wire bond type and can be shipped and stored in a completely knocked down condition. To assemble the container, only the wire loops need to be connected together and eight steel straps applied, each around the two fin containers, one each around the two saddles, and four each 5.rr--r..i the too sides and bottom of the container. �SECTION VIII CONTAMINATION HARDWARE DEVELOPMENT 8.1 REQUIREMENTS Specific requirements for contamination hardware development were as follows: • • 8.2 Must be compatible with the three-module configuratior.. Must reduce aircraft contamination when used on the F-4 aircraft,. DESIGN At the time of the compatibility fit tests, the spray contamination on the inboard wing station of the F-4 was considered to be a problem. When flight tests were conducted at Eglin Air Force Base with single modules on the inboard wing stations of the F-4, excessive contamination occurred on the ving with the spray going up into the wheel well. Films taken of the spraying indicated that the air flow around the aft end of the r.odule was causing the spray to go up the aft of the pylon and into the wheel well. To prevent the air flow around the aft end of the module from being forced up, an air scoop was designed to pull air in from around the side of the module to the nozzle to reduce the influence of the upward air flow at the nozzle (Figure 39) . The prototype hardware was designed to fit the three-module configuration since this is the configuration most likely to be flown on the F-4. 89 �a. Side View b. Air Inlet Figure 39. Contamination Hardv.'are (Turning Vane Kit) 90 �SECTION IX TESTING • - .-Numerous static and dynamic tests were conducted cvirir.c . -the •.'.-programto evaluate and verify designs. The follov:ir.^ paragraphs summarize the major tests. 9.1 TWO-MODULE WIND TUNNEL TESTS Wind tunr.el tests using 16 percent scale rodels v.ere conducted to determine configuration modifications necessary to ensure stability of the PAU-8/A two-module configuration at Kach 0.5. (Previous wind tunnel tests on this conficuratic: resulted in the determination that the stability r.argir. v:as such as to r,ake aircraft-store separation unsafe and that minimum stability occurred at Mach 0.5.)" All tests were conducted in the four-foot Trisonic Wind Tunnel at Douglas Aerophysics Laboratory, El Segundo, California. The five configurations tested are described i.n Table VII and shown in •Figures 40 through 44. A total of 21 good runs, including a repeatability run, were made. On the basis of producibility and drag considerations, as well as stability effects, configuration No. 2 was considered best. The aerodynamic force and moment coefficient slopes of interest at lev; angles of attack or yaw are shown for the various configurations in Figure 45. Figure 46 is a closer look at stability margins and drag effects (a negative stability margin indicates the number of module diameters aft of the center of gravity where the center of pressure is located). Longitudinal stability is significantly affected only by configuration No. 3. However, longitudinal stability for the basic configuration is adequate and improvement in this direction is not as important as improvement in lateral stability. Lateral stability is significantly improved by any of the modifications. Configuration 5, however, produces a large increase in drag, which is detrimental to other flight characteristics. The aerodynamic data for the chosen configuration (No. 2) is presented in Figures 47, 48, anc. 49. b Air Force Armament Laboratory Technical Report AFATL-TR-69-65, Chemical Anticrop Dispenser Development, May 1969, UNCLASSIFIED 91 �TABLE VII. TWO-MODULE WIND TUNNEL CONFIGURATIONS TESTED CONFIGURATION NUMBER MODIFICATION DESCRIPTION Basic Two-Module Configuration 2 Four Short Fins 3 Combination Stabilizer (Figure 40) (Figure 41) (Multi-Surface) (Figure 42) 4 Vertical Fin (Figure 43) 5 Drag Plates (Figure 44) Figure 40. Configuration No, 1 - Basic Two-Module Dispenser 92 �Figure 41. Configuration No. 2 Four Short Fins Figure 42, Configuration No. 3 Combination (MultiSurface) Stabilizer 93 �Figure 43. Configuration No. 4 - Vertical Fin Figure 44. Configuration No. 5 - Drag Plates 94 �MACH 0.5 SMALL ANGLES PITCHING MOMENT NORMAL FORCE -0.8T C.. 0-H D 0-J 1 2 3 1 4 CONFIGURATION 2 3 4 CONFIGURATION vo YAWING MOMENT SIDE FORCE 0,8 T -0.4- J OJ 1 2 3 4 1 5 2 3 CONFIGURATION CONFIGURATION Figure 45. Wind Tunnel Tost Results 4 5 �0= 0° O f = 0-10° LONGITUDINAL , C T A f t 1 1 1 TV O 1 **D 1 L 1 1 T MARGIN. \ r •> SEFERENCE -| . I \ ««/ 0 1 2 3 «t COJiFIGUKATION cc= 0° -2- p- 0-10° — LATERAL ^, _ |— STABILITY ~" CONFIGURATION I REFERENCE MARGIN. (&) V. I 2 3 CONFIGURATION = /9= Q c ZERO-LIFT DRAG. CCKFIGtHATIOK REFERENCE 0 -i 2 3 CONFIGURATION Figure 46. Kind Tunnel Configuration Comparisons 96 �k»CH 0.5. /?= 0 (CONFIGURATION 2) y o -u- i -3, -2- - I- 05 10 15 ANGLE OF ATTACK, a (DEGREE) Figure 47. Longitudinal Stability of PAU-8/A Configuration 97 �Cn HACK 0.5. « =0 (CONFIGURATION 2} -7- -e- LL. U_ UJ 5S O -5- o UJ -H- O UJ t_) cc o 3 -3- -2- -I- 5 10 15 ANGLE OF YAW, ft (DEGREE) Figure 48. Lateral Stability of PAU-8/A Two-Module Configuration 9S 20 �HACK = 0.5 ( C O N F I G U R A T I O N 2) CJ a: i-o z —I C£ O — — u_ x u. 4 UJ O 5 10 15 ANGLE OF ATTACK, « (DEGREE) 2i = 0 UJ <_> cc t~ o = U. Ul —I «_> X cX I- U. UJ o S 10 IS ANGLE OF YAW, a(DEGREE) Figure 49, Axial Force Characteristics of Two-Module Configuration 99 20 �Kind tunnel data presents a confident picture of stability for the configuration chosen. Data scatter appeared small and repeatability was good, and Reynolds number effects were checked and found negligible. One run was performed at "ach 0.7 and showed only a slight increase in drag coefficient, indicating the lack of Mach number effects in this range. 9.2 TV:O-:-:ODULE CAPTIVE FLIGHT AND JETTISON TEST A series of flight tests was conducted to determine the qualitative captive flight and jettison characteristics of a two-module PAU-8/A dispenser. The test store was flown and ejected from Aero-7A bonb ejector rack on an F-86 aircraft. 9.2.1 TEST STORE The store used in the flight tests was a boiler plate replica of the two-module configuration. Due to limited ground clearance on the F-86 test aircraft, a full-scale two-module configuration could not be used; consequently, a sub-scale (65 percent) version was used (Figure 50). This scale was based oh ground clearance considerations for safe flight. The aerodynamic characteristics of the scale store were the same as the full-size item; the magnitude of the forces and moments was reduced due to the difference in scalo. To achieve dynamic similarity between the test store and the full-scale iteri, the mass and moment of inertia were scaled according to the methods of reference.c The "heavy model" technique, which results in an accurate portrayal of the store separation trajectory, was chosen. The test store represented the two-module configuration in the full condition. This was chosen because a full system results in the nir.ir.um stability margin. The scaled weight cf the test store was 655 pounds, with a mass moment of ir.ertia cf SO slug-ft- in both pitch and yaw. Flow tufts were applied to the aft end of the store in an effort to determine general flow patterns and flow separation points under various flight conditions. c XACA Report NACATIX 3907. similitude Relations for FreeMcv.ol vrir.a-Tur.nel Studies of Store-Dropping Problems, January 100 ��?.2.2 Captive Flight Tests ~'.-.e captive flight phase of the test consisted of two scrties ir. vhich air speeds up to 475 KIAS were achieved. Flow pa—err.s alor.g the tufted aft end of the store were determined through the use of the on-board camera. The primary flight conditions and maneuvers performed during the captive flight phases are shown in Table VIII. The straight and level flight phase was performed at a pressure altitude of 10,000 feet; maneuvers were performed at 5,000 feet. TABLE VIII. FLIGHT MANEUVERS FOR TWO-MODULE CAPTIVE FLIGHT TESTS MANEUVER AIRSPEED S t r a i g h t & Level 250 Straight & Level 300 Straight & Level 350 Straight & Level 400 Straight & Level 430 Straight & Level 450 Straight & Level 475 Left Yaw 325 R i g h t Yaw 325 Wings-Level P u l l o u t (2g) 325 Left R o l l 350 Right RolI 350 Landing Configuration 150 Landing Configuration 115 102 (KIAS) �Left wing down-trim was required at 350 KIAS; increases in speed beyond this point required less significant trim changes. Buffet onset for the F-86 with the two-module store was 430 KIAS, with the buffet increasing in severity as speed was increased. At 475 KIAS, the high frequency buffet was severe, and high speed tests were halted at this point. Yaw maneuvers indicated that the store greatly increased the lateral (directional) stability of the aircraft. Flow pattern films taken at 64 fran-.es per second indicated reasonable flow conditions during all Of the flight conditions investigated. In general, the captive flight flow followed the stream lines shown in Figure 51 even during maneuvers. Flow separation did not occur until the flow had reached the boattail, and the separation point tended to move forward as speed increased. At no time during the, flight tests did gross flow separation occur forward of the boattail. 9.2.3 f Jettison Test The store was ejected from the F-86 v/hile the aircraft was in level flight at 350 KIAS (367 KTAS) at an altitude of 2,500 feet AGL (approximately 4,800 feet MSL). A T-33. aircraft was used as a chase plane to view and film the jettison. Separation of the store from the F-86 aircraft was clean and positive. A slow roll to the left started shortly after ejection, and the store reached a roll angle of 90 degrees approximately 60 feet below the aircraft. The store was observed to be completely stable throughout its flight. The pilot reported that the ejection reaction on the aircraft was nild and that no handling difficulties were experienced at ejection. . 9.2.4 Conclusions Based on the results of this jettison test, and considering the wind tunnel data and results of the computer simulation of separation characteristics, it has been established that the two-module PAU-8/A configuration is a stable store which reacts to ejection in a normal manner. Its separation characteristics may, therefore, be predicted with the same degree of accuracy as any other high density, stable airborne store. In general, the two-module PAU-8/A configurations may be considered safe to jettison under normal fliaht conditions. 103 �-STREAM LINES SEPARATION POINTS: — 250-350 MAS — tOO KIAS —tSO KIAS -1(75 KIAS Figure 51. Captive Flight Flow Patterns 104 �9.3 AIRCRAFT PHYSICAL COMPATIBILITY TESTS The purpose of these tests was to verify aircraft/store physical compatibility findings from a previous study wherein aircraft compatibility drawings were used. Because certain detail features are sometimes omitted from compatibility drawings and since compatibility drawings are relatively small scale (1/16), fit tests provide a realistic, final check of physical compatibility. 9.3.1 Test Equipment Full-scale PAU-8/A mock-ups of fiberglass shells filled with lov; density foam were used for the fit tests. A mock-up module mating assembly and actual store fins were used to create one-, two-, three-, and four-module configurations as required. 9.3.2 Fit Tests The first series of fit tests was conducted at Nellis Air Force Base, Nevada, on the F-4, F-100, F-105, and F-lll aircraft. In addition, the MJ-1 and MHU-83/E bomb lift trucks were checked for compatibility. The MJ-1 with the proposed (simulated) loading and handling adapter plate was used for the fit tests. The second series of fit tests was conducted at the Naval Air Station, Lemoore, California, on Navy models of the A-l and A-7 aircraft. No storaoe handling gear was available for these tests, and the stores were mounted by hand. However, loading from the side with either the MJ-1 or the MHU-83/E appeared feasible for both aircraft. Table IX shows a sumnary of the fit tests results. 9.4 DROPLET SIZE AND DISPENSER AIRWORTHINESS TESTS Aircraft flight tests were conducted on: • GFE supplied modules and nozzles • GFE supplied modules and test nozzles • GFE supplied modules and prototype nozzles • Final module and nozzle design Six series of tests were conducted to determine the droplet size of the spray and airworthiness of the PAU-8/A. Droplet size samples were obtained by arranging 5x6-1/2 inch Kronekote cards as shown in Figure 52. All tests were conducted at Fallon Naval Auxiliary Air Station, Nevada (altitude —415C fee~ above sea level). All flights were made at 100 feet AGL. Table X sunmarizes test conditions and results. The nozzles used in the tests are described in Table XI. 105 �TABLE IX. WING WING AIRCRAFT STATION SWEEP o tr. F-1 F-i» F-IOO F-IOO F-IOO F-IOS F-IOS F-lll F-Itt F-lll F-llt F-lll F-ll! SINGLE MODULE TWO MODULES THREE MODULES FOUR MODULES 1 Yes 2 1 Yes No" Nod Yesa Yos Nod Nod Yes Yes 2 3 1 See Note b Yes 2 1 26' 26s 2 26" 26- c Yes Nod No No«' * Noc' d Yes NoC>" Yes c Yes Yes Yes No Yes Yes Yes d c SPRAY CONTAMINATION Quest i enable None c No No6' d Noc> d Probable Probable None Yes Possible c d No Yes Yes No ' Yes Yes None Probable Possible Yes Yes Yes None Yes Yes Yes Nod Yes Nod None Probable Yes Yes Nod Probable None Noc- d Probable 3 4 1 72.5" Yes Yes 2 72=5" Yes Yes Yes Yes d No Yes Yes No* Yes Yes Yes Yes Probable Yes Yes Yes Yes None A-l 1 A-7 1 A-7 A-7 2 3 Notes: FIT TEST COMPATIBILITY SUMMARY 3 a Requires Short Fins On Bottom, Long Fins On Top "Station Not Recommended Due To Possible Short Fin/Aileron Interference c Weight Incompatibility ^Physical Incompatibility �T 250 FEET 975 FEET 1 ! • • • * • * • • • • • • • • • • •• • • • • • • • • • • < w ^-SAMPLING CARD (TYP.) 500 FEET ••••••••••«*•••••••%•••••••' >•*•••••• 250 FEET >•*••••••< 25 K£T (TtP. 250 FEET 8 x 8-FOOT HARKER PANEL (TYP.) 2000 fEET 500 FEET PLASTIC PANEL 500 FEET (TOTAL OF 120 SAMPLING CiRDS) FLIGHT LIHE Figure 52. PAU-S/A Drop Zone Layout 107 �TABLE X. -1t *~ I -0 o ~! • — ; ; ! ^. i £ « "' ^- «j Q ; - ; :? - : ;j ---;. :s . - . , -; .. :> • « ! • • » * * *- -^ v 2 ;- * ti---' 2^T 1 •-. -V £S i EG EG EG j 3 i E 7 » ' 3-ji EG I 3 E -5 • 351 EG '-; si "-'C-iJ j T-IO-VS : f-si ! 3 , E : ; rj. P 5" •; : 370 c . j :s s« i F 51 | !o • E j* 7 - 1 ? tS • r 5' ' : r! E - « # * 'C t? '• f J 1 • ' o ' E >* 7 : •'.( si ; F ?i i a ' i •' s* 1 ! x j f * 'c is ' F t ' J ' '' :: : •4 5-6 i S . * 7' «.e .-j '* 5 7 ;j e E ; S, i 3f>3 Tv 0 • 350 . ! , i F- I T l ' '5 ' 364 1 51 ' 1 | - 76 ' 364 f !i ; F - S i ! i j • . T: ; :ro ? s» 1 i j . ; *£ ^ j j s-5 -rs i F 51 ' 'i ; :-s! J6i , :u ! :n p i 1 • : • ,7; i ;u • • ** •:• : ' t 51 ! (t.Ji 1 t j - 'SO : :?1 - ; SC 353 1 i - F 5i i 4 E *: • 3<i9 -J ?54 s; - rco s • -is . P r i • • ; 'o- ts P ss 2 E ' 7 4 ! :05 • • ; «c os ; F ?s 2 E ' 74 : cos • P-ci ; E . 75 - :03 -^3 ! F-fa 53 ! P - ? 5 ri ! F-S5 ? j E ,' 7 J' - C03 'i '. 1C- -sa j : '; ' ^ •: I :C •: 1 | S £3 ta .,5 • r - ? s ! i -. f i .: 7 t 54 i !S8 M S 54 s IS8 0 t 5S ; 504 S} ; 53 ! 504 • f ti t ' 1 1a I' I NS N* • 341 ' i \J \* f • '• \» 114 iu : 3f4 4 * : 5 * i * S •• -. , •: ' t • ' N» : \t-3S4 * ' : \» ' \* :J5 « • ' , I , it ! ^ i r -«; ' i - ,: : • : ! P ' • C - £tt } 22: ; « ' ' •: - . S 3 ! 358 5-5 , f- :•: 3 '• : -, • « ' j t-3 ! P I -4 3 -- ,-..> : 1 : • •: ' P • ' : s 4? . 3^6 f 3 :• ' ; ! f ' i ; 4 \E ^4 * 349 4 SE T« ' 2' 8 2 %E . 5 3 ' T ' 4 : r : NE c: : 3^ 11 « ! , „ . , , F M I . K J . » - Ic Sou Eli E j l i j l f f G!»C.I! 1 M eg . 4 1 r , , ( i 8, 1 C « > SG i - rnirs> . O^ 1 -H « ;•' \ C S - • # . . ' " 7 f !• ; e ' e : . ; 17 t.,-. .i 0 : - a' - 2 1 7 .• i, • i j j E ; 3 ' 35 1 f :• '3 ' ' : " ^i ~- O*t J j; • '>» S3 •it: s- REMARKS .V ** a--~ Si ,>^ " ; C- ' • i i » 1 * < -lic «"5i Ji^5 "*6 1 - 4 S- - 'ice tiinJ 'a; M I J I F o i Goo,' C.^t." £FE l»:«tiile * i! C : ' < » t 9 l s 5 , " • I47J 3- - 4 } .-;t, s; 5 . "i' ES EG ES EG o^^t ^~> -^ n.1 — t' — '*"'5 '."^ 5^" .«S. >v — -X -. > i~,« _ -» 4 if tS sS 6S re I i ii jf.^ '•*• "~^ i A ^ -^ !l '^ -*• -* ~- 7^ —• ft. • O ™ :t..i. .• - ^iw< , ««- a; ' ' S J 8-5 S-£ S-S f 6 PAU-S/A SPSAY TESTS RESULTS J 3i f Si ' 3J 1 S • " «S J 1 ^; m . EC EG ! 8C« « , «%4 ES ; M FO 23 8 ' S5 '• • FO s i 13 : : ->i 21 s i s ? * FO FC Z3 9 i ? ? ' j> ; » C ; *0 • FO FO : l\ »3 - - ? FO FO FO e: '} * $.2 • i; ' •js: • FO 4? '3 : is?7 " G 44 U "' $5 :: FO ! 4S 13 »S9 6 44 (4 ' if$ " FO 44 »4 1 I3J H?» 43 M ; iE3 FO TUP* 44 U S ^5 er£ M e t t l e Aiu* C o r t i o l s 4C t3 <j ^5 FO 2*. tr tu 5 f «» < i EG EG EG EG EG B t t 8 8 8 0 G * C '•3 15 is 75 15J f J *M 1? !J3 »S 150 I* 75 13 T5 »50 f • 53 8 f SC '•* 15* ;* ., " ^C M "k s. W » ' f.-* 4* » | A 1 * *t " - 1 C 1 5 ;| :*: ! ' vv? S^E M o d u l e ( t j t » ) *il!i N t » CfSiji* ' J^5 ' t*"ttais - ' ^t* Besign Woojl* Canplele I - * | j;j \ ^ ^^ ^ ^ : ': ?J3 ' ' x':3 ; -.11 \ti Of si (ii Voatl* C o r p l e l e 8 f l u * 1.34 SG • - Qia«?e 1.28 SC • - V t i t e 1. '4 SG HA - hat A o i l a U t B u u n j t « j l SG v S - ^ - t l i c S i .1 « t O s i f c r i p i >;.• > Q I ,c f i » - n «,::• SG -> t » » r > • » « . < i i n i i f i rc: jr:;^»*t ».••;? i « ' 4. SG , 4 : <,• • • » - t - 1QS �TABLE X. j .Z <* ! a> r^ ' »: ^* x -•I -•; ?' ;-f .-: 4; 4; 4j 44 4^ «J PAU-S/A SPRAY TESTS RESULTS (CONCLUDED) ? *t a , J :« £9 i i-18-69 , J 19-69 ! 3-19-69 ! 3--C-69 i 5-19-69 j 3 13-69 i » 20-03 ; 3 :0-69 • J-20-E9 ! ? .'3-C9 *— iS g oc -t _ ' Uj V i'a" •D 10 0 0 1 1 F-51 F-51 4 4 3 3 0 r 51 U ''I] "*cv ii^i •* *o ~~2- F-«! f-5! f-5i F-51 F-51 > *_ Z ci ^>-~- ^[^ .* 'V' ^ ~^ *i; <m'. _ F-t' F-51 ^" s s -7 »- ^*».-— ":o ""7' tt ^ ^~ -sU 7Q ."i 70 50 SO ,VJ4 NE 54 t«E 54 ft 67 w 6? SE 52 E 52 CO 61 :n o:uj *l i— .-j Ul 0 — «I eo 0 0 II 366 N ."23 *' W 342 223 333 223 3G9 ,\"4 ?4J - * » 0 0 0 0 c 1*1 *.(-—•* "L^ 'lj~. x :s 25 150 130 25 25 75 75 25 25 75 75 Path Njrn To South Eth,le-e Si, to I 1.11 SG ( S p e c i f i c G r a v i t y ) -1 Futl 0 - 1 0 . 8 3 SG G l y c e r c l -SC 1 -) G l y c e r i n e ) 1.25 SG TOP - T e t r a Potissiun Py rooheiphate (40*. S o U t i a n In H a t e O 1.45 SG E6 F3 G - 109 REMARKS _J 1 fs| «^ C* I-" Z-— , J •^ — ^ i 20 20 IS 16 20 20 19 19 20 20 19 19 f * V* < «TSI 4 «« Ol^> ?9i SCO Ne* Design Modul* C o m p l e t e 1 D," 134 ise 1T7 243 14* 163 US 15E «?7 B 0 * M New Design Nodule Complete - B.ue 1.34 SG - Orange 1.28 SG - W h i t e 1.14 SS - Not A v a i l a b l e O g l i n g T t t t �TABLE XI, NOZZLE DESCRIPTIONS ORIFICE '.::::.• a S. 5:< SJME I T e s t N:::ie N o . l 12 0.25 Diameter 0° Aft 12 2 T e s t N o z z l e No. I 12 0.25 Diameter 90° Down 12 3 GFE 32 0.25 Diameter 0° Aft 13 4 T e s t N o z z l e No.l 90° Down 12 5 Test N o z z l e No.l 192 0.062 Dianeter 90° Down 12 6 GFE 0° Aft 13 7 Test Nozzle No.l 192 0.062 Diameter 0° Aft 12 8 T e s t N o z z l e No. 2 2 0.2B1 Dimeter 9 T e s t N o z z l e No. 2 1 0.391 Diaraeter 10 Test N o z z l e No. 2 1 0.391 Dianeter W i t h Deflection 11 510. SIZE OS SETTING ( INCHES) 2 0.25x1.875 Slot 32 0.10 Diameter INJECTION ANGLE W I T H HORIZONTAL 20° Up And Down REFERENCE FIGURE 14A 148 30° Dpwn 14C Test N o z z l e No.2 1 0 . 1 8 7 x 0 . 675 S l o t 90° Down 14D 12 Test N o z z l e No.2 1 0 . 1 8 6 x 1 . 5 0 Slot 90° Down 14E 13 Test N o z z l e No. 3 1 0.75 Diameter 3 0° Aft ISA 14 T e s t N o z z l e No. 3 1 0.50 Diameter 0° Aft 158 15 P r o t o t y p e No. 3 1 No. 2 0° Aft 18 16 P r o t o t y p e No. 3 1 No. 8 0° Aft 18 17 P r o t o t y p e No. 3 1 0° Aft 18 18 Production 1 No. 1 0° Aft 19 & 53 19 Production 1 No. 5 0° Aft 19 & 53 20 3 0° Aft Production 1 No. 13 0° Aft 19 & 53 No. 14 F l o * R i t e C o n t r o l l e d U;i Strenni W i t h An O r i f i c e . 110 �Figure 53. Production Nozzle 111 �SECTION X MAINTAINABILITY AND RELIABILITY 10.1 MAINTAINABILITY The maintainability program for the PAU-8/A Spray Tank consisted of maintainability analysis, troubleshooting analysis, maintainability parameter predictions, and preventative maintenance. These four analyses combine to form a major indicator of system effectiveness. 10.1.1 Maintainability Analysis The first step of the maintainability analysis was to ziefir.e the operating environment and conditions. The syster. has i:eer. designed to operate on a variety of aircraft, but this v:ill not be a maintainability constraint since the operational procedures for the system are independent of the type of aircraft. The varied weather conditions will not impose any maintainability constraints as lone as the agent does not freeze in the tank. A set of r.ainte.-.ance and preflight operational procedures, based on qualitative maintainability recruirerents and constraints, was devised to form a base for the quantitative analysis. These procedures and associated assumptions are as follows: • A preventative maintenance schedule will be followed for all units. • The modules will be configured and prepared for mounting at the maintenance facility. • The system will be mounted and the functional verification test will be performed using the dispenser test unit. • Any failure during the functional test that is due to a component failure will require system removal for bench repair. • Failures due to non-defective components, i.e., improper hook-up, etc., will be corrected on-line and the functional test repeated. 112 �• After successful completion of the functional test and after the pressure reservoir is charged, the agent tank is filled. The maintenance strategy adopted is to make repairs at the component level. This means that if a valve fails, the valve is replaced but not repaired at the field maintenance shop level. If the valve is to be repaired, it will be done at a higher level in the maintenance hier. chy. This strategy -.-.-as adopted after considering the characteristics of the syster., skill level of the maintenance personnel, and available tools and equipment. The only special equipment needed at the field maintenance shop level will be the functional verification-(dispenser test) unit which is supplied. A rack, stand, or loading and handling adapter will be needed to hold the modules for shop maintenance. The personnel skill requirements should easily be net by Air Force mechanical or electrical technicians. The syster. design is such that one person will bo able to carry cut the preventative maintenance steps or correct a malfunction with proper technical data support. 10.1.2 Troubleshooting Guide A troubleshooting guide (Figure 54) was developed to give structure to the maintainability prediction. A schematic cf the syster. (Figure 55) is the major path. If the syster. is operational, there is no deviation from this path. Brar.chir.cr fror the major path is necessary when a failure is encountered. The branching continues until the cause of the failure is deterr.ir.ed. After the fault is found and corrected, the technician is directed back to the start of the functional test to verify system operation. As previously stated, the troubleshooting strategy is to replace components with no attempt to fix them at the field level of the maintenance hierarchy. Since it is likely that some of the malfunctions will be due to clogged lines and valves, all of the failures may not require component replacement. It is possible that cleaning the component which has been isolated by using the troubleshooting guide and the line leading in and out of the component will correct the problem. As a preventative maintenance measure, the lines and 0-rings associated with a component should be cleaned whenever the component is removed for inspection or when it is replaced. 113 �O «:. Figu.ve 54. Functional Level Troubleshooting Guide �PRESSURE, GAUGE TO DISSEMINATION SWITCH REGULATOR HIGH-PRESSURE RELIEF VALVE\ HSSEMINATION PILOT NOZZLE AGENT ( DISSEMINATION VALVE ACTUATOR -— BLEED VALVE 3 M( OK-PRESSURE TO NEXT MODULE CHECK / CHARGING VALVE GAS INPUT Figure 55. System Schematic RELIEF VALVE �10.1.3 Maintainability Prediction The calculated mean-time-to-repair (MTTR) for the PAD 8/A .is 44 r.ir.utes. The maintainability data and the calculations are giver, in paragraph 10.2.5. The task times used for calculating the MTTR are composed of fault location tine, fault correction time, and fix verification time. The fault location tires are based on following the troubleshooting guide; the fault correction times are estimated. Since the MTTR is well, below the three-hour specified requirement, measurements of component removal and replacement times were not necessary to ensure compliance with the maintainability specification. The system design is such that all components except one are readily accessible when the nose cone is removed. This easy access is the reason for the low MTTR. The only component that requires removal of another component for access is the 10-micron gas filter. The charging valve must be removed before the filter can be changed. This good maintainability design is the result of using the manifold in the pneumatic system. The manifold eliminates a considerable amount of tubing and fittings, thereby increasing the reliability and enhancing the maintainability of the system. 10.1.4 Preventative Maintenance The preventative maintenance requirements for the system have been minimized by careful design. Hermetically sealed relays are used to eliminate any periodic maintenance for the electrical system. Mesh filters are used in the fill ports to prevent large particles from getting into the system and clogging pneur.atic lines and valves. It appears that the unit will not require any special periodic maintenance. The one possible exception to this will be the gas filter which n.ay require periodic replacement or clear.ir.g during use. The diaphragm in the nozzle assembly will require periodic inspection during use. Accelerated life tests indicate that the diaphragm should withstand 1200 to 1500 cycles at r.axinur. opening of 150 GPM and more cycles at a lesser flow rate. 116 �10-1-5 Maintainability Data and Calculations The maintainability data is presented in Table XII. This data is used to calculate mean-time-to-repair (MTTR) fron the following formula: ?Aiti MTTR = ZAi >.i and t^ are the failure rate and task time, respectively, for the i^th conponent. When there is more th&>. one of a particular component and the task time is the same for all of them, the failure rate cap be multiplied by the quantity. The summation is overall maintainable components. The MTTR is calculated for a four module assembly: 5305 The n. is included since the repair time for all components will be the sarr.e in each module. En iX ifci ~ 2-30,114 Therefore, MTTR =43.5 Since the task times are estimates, only two significant digits will be retained. The MTTR will, therefore, be 44 minutes. 10.2 RELIABILITY The first step in the reliability analysis is to define the mission and to specify what constitutes a mission failure. Mission failure is specified as "any malfunction that may cause mission degradation". A sample mission profile is shown in Figure 56. For the reliability analysis, the mission can be 117 �TABLE XII. MAINTAINABILITY DATA QUANTITY FAILU3E/ TASK TIME ;' OF TOTAL MAIN. TIME (1) .(Tj_. P;s:«j?e S w i t c h 4 240.3 39 37.700 16.40 t r - n j 'ijlve 4 257.3 37 38.000 16.50 4 277.0 58 64.200 28. C3 c 9 .5 48 d IL I .843 flfl O ,30 COMPONENT .: j:'cr M, PI nXt ml JQ C-.-jctor J2. P2 4 7.2 48 1.380 0.60 Diode 8 o.e 36 230 OJQ R e l a y Kl 4 3.3 45 1,570 C.75 Selector Switch 4 2.0 53 424 0.18 Safety S*itch 4 24.3 32 3.050 1.32 F i l l e r Caps 8 24.3 6 1.105 0.43 Strap 4 G 1 in 244 O.II Bait 4 1.3 1 37 0.02 Charge V a l v e 4 12.2 26 1.270 0.55 B e l i e f V a l v e (H'gh Pressure) 4 20.0 33 2.640 1.15 Filter 4 6.0 26 624 0-27 P r e s s u r e Gauge 4 10.0 33 1.320 0.57 P r e s s u r e Vessel 4 5.4 69 1,495 0.65 Check V.ilve 4 13.1 40 2.100 0.91 Bleed V a l v e 8 12.2 41 4,000 1.74 S e l i e f V a l v e (Low Pressure) 4 20.0 35 2.800 1.22 -•iafjhrag.i 4 43 0 22 4.230 1.84 Sissermiation V a l v e 4 282.0 53 59.750 25.90 118 �(1) (2) (3) (4) (5) (6) (7) (8) (9) TAKEOFF CLIMB CRUISE (OUTBOUND) DESCENT DELIVERY CLIMB CRUISE (INBOUND) DESCENT UNO (T) Figure 56. TIME ( MINUTES) 0.4 20.0 1.4 4.0 (WORST CASE: 4 MODULES SEQUENTIAL) 9.0 20.0 2.2 '//// '//' '//// Mission Profile for Sequential Dissemination 119 �considered complete after segment five. The four-module sequential mode of operation was selected since it is the highest stress mode from a reliability point of view. The reliability model developed for the system is shown in Figure 57. The model is based on a functional partitioning of the system. For this sequential model, the probability of success is the product of the success probability of each subsystem;, i.e. , = P P P P P P P *A*B C D E F G (1) (Pv is the probability of successful mission operation of the X xth subsystem.) Two assumptions that affect the reliability analysis are made. The first assumption is that all required periodic preventative maintenance will be done. The second assumption is that for all missions, the modules will be given the preflight checkout using the dispenser test set. A functional flow diagrar. of the preflight procedure is shown in Figure 58. This assures that th« system is operational just prior to use, thereby allowing the reliability calculations to be made using operational time only. The data and the calculations for the probability of success for each module are given in paragraph 11.2.1. The results of the calculations are presented in Table XIII. TABLE XIII. SUCCESS PROBABILITIES FOP SUBSYSTEMS A = 0.90">87 P£ = 0.99934 B = 0.99999 P_ = 0.99845 r PG - 0.99989 P P P~ = 0.99989 C P D = 0.99989 �A B C 0 E F A. MODULE STRUCTURE E. NOZZLE AND DISSEMINATION SYSTEM B. MATING ASSEMBLY F. ARMIN3 AND REGULATING SYSTEM C. HIGH PRESSURE SYSTEM G. ELECTRICAL CONTROL SYSTEM 10 0. LOW PRESSURE SYSTEM Figure 57. Reliability Model G �SYSTEM OPERATIONAL SET SELECTOR SWITCH AND NOZZLE TO - DESIGNATED POSITION ^ PASS INFORM AIRCRAFT MAINTENANCE CREW CONNECT TEST SET AND PULL SAFETY PIN t FAIL TEST ARM AND FIRE CIRCUITS FAIL PASS AIRCRAFT SCHEDULED FOR MISSION PERFORM CONTINUITY TESTS MAKE PROPER CONNECTIONS CHECK ALL ELECTRICAL CONNECTIONS EXCLUSIVE "OR" NOT OK ( ) INCLUSIVE "OR" ? REPLACE SAFETY PIN AND NOSE CONE OK RETURN UNIT FOR BENCH MAINTENANCE Figure 58. Line Checkout and Preparation Procedure �from the data in Table XIII, the probability of mission success cai be calculated, using Equation (1). The result is PS = 0.997. This would give a failure rate of 0.3 percent, wnich is well under the ten percent specified in the contract. It is at this point that the implications of a development versus a production contract become important. The failure rate calculated above is based on a development contract where all of the units are inspected and tested. On a production basis, the failure rate could be higher since sampling techniques will be used for inspection and testing. In fact, since the inherent reliability is so high, a key factor in selecting sample size for the production testing will be the minimum acceptable failure rate. The inherent reliability of the module structure, i.e., welds and material, is so high that it has a negligible contribution to the probability of failure. This high inherent reliability is due to the large safety factors used iii the design. These high safety factors are a result of strict adherence to the design criteria in MIL-A-8591 and the need to maintain structural integrity under the severe test conditions specified in MIL-STD-810. This high structural reliability is achieved as long as 100 percent inspection of miterial and joints is done. Kith anything less than 100 percent inspection there is the possibility of a module with a bad weld or inferior material being shipped to the field. If this happens the achieved field reliability may be reduced. This distinguishes the field reliability from the design reliability. The reliability figure used in this report is a design reliability where 100 percent inspection was conducted. 10.2.1 Reliability Data and Calculations The reliability data is presented in Table XIV. The data was gathered and developed from the following sources: * MIL-HDBK-217A, Reliability Stress and Failure Rate Data for Electronic Equipment. • Bureau of Naval Weapons Failure Rate Data Handbook (FARADA), Volumes 1A and IB 123 �TABLE XIV. COMPONENT RELIABILITY DATA QUANTITY "i O P E R A T I N G FAILURE RA1E TIME FAIL/IO& HCHRS t • (HOURS] Ai P.M. Functional Unit Module S t r u c t u r e K i l l e r Caps Tank 8 11 0.5 24.3 18.7 : .nctional Unit w?:.Je A d a p t e r Strap Bolt 4 4 0.5 0.5 6.1 1.3 12.2 2.6 14.8 V, 0.5 0.5 0.5 0.5 0.5 0.5 12.2 . 20.0 6.0 10.0 5.4 REMARKS 24.4 40.0 12.0 20.0 10.8 1.0 Functional Unit Hign-Pressure System Charge Valve R e l i e f Valve Gas Fitter Pressure Gauge 0-Rings (Total For System) 0.5 . 97.2 37.4 134.6 Approx. Of Aqent Tank •— 4 4 4 <i 10 0.2 108.2 Functional Unit Low-Pressure System Check Valve Bleed Valve Relief Valve 4 8 4 0.5 0.5 0.5 13.1 12.2 20.0 26.2 48.8 40.0 1(5.0 Functional Unit Nozzle And Disseminator Assembly Diaphragm Dissemination P i l o t Valve 4 4 0.5 0.5 48 282 Functional Unit Arming And Regulating S> sten Pressure Switch Solenoid Valve Regulator 4 4 4 0.5 0.5 0.5 240.9 257.3 277.0 481.8 514.6 554.0 1550.4 Functional Unit Electrical Control System Connector Jl Connector 02 Diodes CR I, 2 Relay Kl Selector Switch S a f e t y Switch 4 4 8 4 4 4 0.5 0.5 0.5 0.5 0.5 0.5 9.i 7.2 0.8 9.3 2.0 24.3 19.2 14,4 3.2 18.6 4.0 48.6 108.0 , I.M 96 564 660 8 Pins Used 6 Pins Used �« Timmerman, P., Fault Data for the Prediction of Reliability^of Electronicancl^ Mechanical Equipment and Systems, Danish Atomic Engergy Commission Technical Report, February 1968. In cases where specific data could not be obtained, the similar equipment procedures as specified in MIL-HDBK-217A were vsed. 125 �SECTION XI SUMMARY The basic modular configuration was defined in the contract. The design evaluation was centered on improving the general design furnished under a previous contract. Irprcver.erts have been obtained in the areas of cost, more ccr.trclled. flov: conditions throughout the dissemination cycle, renter of gravity control, reliability, maintainability and weight reduction. The PAU-8/A Spray Tank consists of four modules, which are constructed to allow the system to be used in one-, two-, three-, or four-module configurations where aircraft pylon characteristics are limited by maximum weight, ground clearance, etc. The design and development effort resulted in a modular system which has an empty weight of 225 pounds per module, has flow rates of 15 to 150 gallons per minute per module, can be externally carried and operated on high and low performance aircraft, contains 50 gallons per module, and is structually sound and aerodynamically stable. Plow models of the internal sections of the agent tanks which simulated the GFE furnished tanks and the proposed design were constructed to study the flow problems within the tanks. Several test and prototype nozzles were fabricated and evaluated during development to ensure nozzle simplicity and reliability. Other equipment designed, developed, fabricated, tested, and delivered to support the PAU-8/A were the loading and handling adapter for use with the MJ-1 and the MHU-83/E bomb lift trucks, dispenser test sets to preflight check the modules and the arm and fire circuit of the aircraft as well as operate the modules for static ground operations, temporary storage ar.d shipping containers, and an adapter kit to reduce spray contamination of the F-4 aircraft. Kind tunnel tests and jettison tests provided data to establish aerodynamic stability and an adequate safety margin for jettison on all configurations. 126 �Aircraft compatibility studies were made with layouts and full scale mock-ups of the PAU-8/A on the F-4, F-100, F-105, F-lll, A-1E, and A-7 aircraft. Flight tests to study nozzle design, air flow, and the complete system were made on F-51 and F-86 aircraft with a single, full-scale module and a 65 percent scale of the twomodule configuration. Material compatibility studies were made to determine what metals could be used in contact with the agents. Studies were also made to determine what materials could be used to coat the metal to protect it from the effects of the agents. Eight complete four-module systems have been fabricated and delivered to the Air Force for P. & D Engineer* no Evaluation. 127 (The reverse of this page is blank) �FRSC£DIN3 PAGE BUNK.NOT FILMED �DOCUMENT CONTROL DATA - R & D '$v*.r,rt tlm\Mtifmt*. . hotly of . l tfputt ON 4 i s » ' i N i i « c r i . i T » 1t ..rpot.l* author J *« ctailitifd) *'"UNCLASSIFIED" Defense Technology Laboratories FMC Corporation San Jcse, California SPRAY TANK UNIT, AIRCRAFT, PAU-8/A t » c » i e * i » E N O T E } fTVp* ottrpetl mnd tnclullvr Oft**) inal Report - 17 Hay 1968 through 28^February 1971 John J. Harrington ?•. T O T A L NO April 1971 OF P A C E S 7b. NO OF R E F S 136 U. C O X T H A C T OR C R A N T M O »•. ORIGINATOR'S RCPQHT NUMB F08635-68-C-0090 6. PROJEC ' NO Task No. 07 •6. OTHER REPORT NO<*> {Any other nwotbcr* thmt may ttf «. Wbrk Unit to.. 00 to U. S. Government agencies _ f ~iil + liirdtation applied April 1971. Other requests for this document must be referred to the Air Force Armament Laboratory (DLIF), Eglin Air Force Base, Florida 325U2. IJ. SPONSORING MIUIT»BV A C T I V I T Y Available in DDC Air Force Armament Laboratory Air Force Systems Command » Eglin Air Fc : Base, Florida 325*42 e A modular spray system for .anticrop chemicals was designed, developed, fabricated and tested. The system is capable of external carriage on high and low performance aircraft in four possible configurations using either one, two, three, or four modules. Each of the 50-gallon modules is completely interchangeable and can spray at rates from 15 to 150 gallons per minute. The modules use a coinpressed-air/gas reservoir to pressurize the agent reservoir and force the agent out the nozzle. Support equipment, designed and delivered with the dispenser, included the loading and handling adapter kit for the MJ-1 and MHU-83Z bomb lift trucks, the checkout unit, and the anticontamination kit for use with the F-4 aircraft. Nozzle tests were conducted from aircraft at 198 to 504 knots. Droplet sizes of 105 to 555 micron rrmd were obtained with the single module configuration at air speeds of 214 to 354 knots. Full scale flow model tests of the agent tank lead to the development of a module which expels 99 percent of the agent fivm tl« nodule at a flow rate of 150 gallons per minute. Scale wind tunnel and jettison tlight tests were conducted to support the design of a stable two-module configuration. DD ,Fr\,1473 UNCLASSIFIED Srrunlv Clarification �*i*» ' j r i f y fll 1 4 I INK * ROCC WT t IN K H ROLL W1 PAU-8/A Aircraft Spray Tank Unit Dispenser Test Unit Anticrop Dispenser Anticrop Chemical Agents MJ-1 Bomb Lift Truck MHU-83/E Bomb Lift Truck UNCLASSIFIED Security Classification I IN* «OL t C wr ��UNCLASSIFIED/UNLIMITED PLEASE DO NOT RETURN THIS DOCUMENT TO DTIC EACH ACTIVITY IS RESPONSIBLE FOR DESTRUCTION OF THIS DOCUMENT ACCORDING TO APPLICABLE REGULATIONS. UNCLASSIFIED/UNLIMITED � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 026 Folder The folder containing the original item. 0376 Series The series number of the original item. Series II Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Harrington, John J. Description An account of the resource <strong>Corporate Author: </strong>Defense Technology Laboratories, FMC Corporation Date A point or period of time associated with an event in the lifecycle of the resource 1971-04-01 Title A name given to the resource Spray Tank Unit, Aircraft, PAU-8/A Subject The topic of the resource spray equipment Ranch Hand aircraft herbicide application