Ad 0225488

8/13/2019 Ad 0225488

1/102

UNCLASSIFIED

AD NUMBER

AD225488

NEW LIMITATION CHANGE

TOApproved for pub l ic re l ease , distribution

un l imi ted

FROMD i s t r i b u t i o n au thor ized to U.S. Gov't .

agencies and their con t rac to rs ;A d m i n i s t r a t i v e / O p e ra t i o n a l Use; Ju l 1959.Other r eques t s s h a l l be r e f e r r e d to WrightAir Development Center, W righ t -Pa t te r sonAFB, OH 45433.

AUTHORITY

AFAL ltr d t d 17 Aug 1979

THIS PAGE IS UNCLASSIFIED

8/13/2019 Ad 0225488

2/102

DEFENSEDOCUMENTATIONCENTERFOR

SCIENTIFICANDTECHNICALINFORMATIONCAMERONSTATION. ALEXANDRIA.VIRGINIA

K

M AShe

8/13/2019 Ad 0225488

3/102

or ~RErTrnItTRNto

WADC TECHNICAL RE T 53-192 - _ .....Part XIEI

r'ATAU.)GEDBYwcos

MECHANISM OF RAIN EROS ONPart Xli. Mechanism Studies on Neoprene Coatings

Olive G. Engel

National Bureau of Standards

JULY 1959

This report is not to be announced or di.ty ez'e>t: itomaticafyto foreign governments (AFR 205-43A, T ah 6d )

WRIGHT AIR DEVELOPMENT C NTER

BESTAVAILABLECOPY

-- -----

8/13/2019 Ad 0225488

4/102

REPRODUCTION QUALITY NOTICE

This document is the best quality available. The copy furnishedto DTIC contained pages that may have the following qualityproblems:

"* Pages smaller or larger than normal.

"* Pages with background color or light colored printing.

" Pages with small type or poor printing; and or

"* Pages with continuous tone material or color

photographs.

Due to various output media available these conditions may or

may not cause poor legibility in the microfiche or hardcopy output

you receive.

7 If this block is checked, the copy furnished to DTIC

contained pages with color printing, that when reproduced inBlack and White, may change detail of the original copy.

8/13/2019 Ad 0225488

5/102

NOTICES

When Government drawings, specifications, or other data are used for any purpose AtherSthan in connectiun with a definitely related Government procurement. operation, the United S~utes

Government, thereby incurs no responsibility nor any obligation whatsoever; and the fa,, hiat

the Government may have formulated, furnished, or in any ;w'ay supplied the said drawings,specifications, or other data, is rot to be regarded by implicattn or othe-wise as in any

manner

licensing the holder or any other person or corporation, or conveying any rightq or perriszion

to manufacture, use, or sell any patented invention that may in any way be related thereto,

Qualified requesters may obtain copies of this report from the Armed Services lechnficalInformation Agency, (ASTIA), Arlington Hall Station, Arlington 12, Virginia,

Copies of WADC Technical Reports and Technical Notes should not be returned O'U WrightAir De VCtwpr ehL Center unless return is required by security considerations, cotj.aeH titll oIhgI -tions, or nlotice on a specific, document,

8/13/2019 Ad 0225488

6/102

WADC TECHNICAL REPORT 53-192

PART XIII

MECHANISM OF RAIN -EROSIONPOOtXII ch.,iism Studies o" N..p~wn Coatings

Olive G. Engel

National Bureau of Standards

JULY 19659

MateriaLq Laborator-

Contra.-t No. AF .13(616) -12

Project No. 7340

WRIGHT AIR DEVEI.OPMEN (CENTERI It RESEARCH AND I)EVELOPMENT COMMAN4D

UNITED STATFS AIR FORCE

WRIGHT-PATTERSON AIR FORCE BASE. OHIO

BESTAVAILABLECOPY BESTAVAILABLECOPY

8/13/2019 Ad 0225488

7/102

POREWOfD

This report was prepared by the National Bureau ofStandards under USAY Contract No. AP(61 6 ) -5 .42 . The contractwas initiated under Proje~t No. 7540*, "Rul'ber, Plastic andQolpoeite Materials", Task No, T3'00, "Structur l Plastics".The projeot was admainistered under the direction of theMateriaus Laboratory, Directorate of Laboratories, WrightAir Development Center, with Mr, Oeorse P, Pateprii actingas project engineer.

The eperimental work that is reported Was accomplishedwith the ooopetation of the MinnesOta Xiang and ManufaeturrniCompany, the Cornell Aeronautical Laboratory, the OatesEngineering Company, and the Goodyear Tire and RubberCompany,.

Thisreport covers the period of work from abuut Oktober

1954 to June 1958.

WA1NC TE 53-192 Pt XIII I

re

8/13/2019 Ad 0225488

8/102

ABSTRACT

The mechanfism by which neopr'fene coatings rai l i ofri n t e r e s t because a i r traffic

wi l l be carr ied on for many

years to come in the at i tuLderange in which ra in is

still encountered and in theveloc l ty range for which

neoprene coatings are a solutionto the raiaverosion problem.

This report is an account ofst~tdies that nave been made

to determine the mechanism 'bymeans of o neoprene

coatings eventually fai l underhigh-speed rain impingement.

Resu l ts of t e s t s Involvingan t iozonan t appl icat ions

to

the neoprene coat-ls areencouraging enough to warrant

furt.L eIxperiments with suchap p l i ca t i o n %.

PUBLICATION R3VIEW

This repor t has been reviewedand is approved,

FOR THF, COMAN)ER:

Ohtc. (Organic Materials P-,TM a t e r ia l n Laborfa tory

VPY)r', Tit X.-111,FtI~1.

8/13/2019 Ad 0225488

9/102

TABLE OF COI'TNTS

. i t1roduct ion . . . .2, Models of a Waterdrop . . a . .

2.1 Steel Spheres and Deformaing Lead P e l l e t s . 22,2 Oil-Fil led Gelatin Capsu.1es * 3

Response of Several Different Neoprene Based Coatings

3.1 MMME(:-539 Neoprene . 1 . 7

3 .1i Damage Marks Produced on MMMEC-539Neoprene by Impingement of Oil-FilledGelat in Capsules 4 . . 7

3.1.2 Damage Produced. on MMM '10-539 Neoprene byImpingement of Waterdrops , 9

3.2 Modified MMXEC-539 Neoprene Coatings * 13

3.e.1 Damage Marks Produced on Modified MMME0-539 Neoprene by Impingement of Oil-Filled Gelatin Capsules . . 15

3.2o2 Damage Produoed on Modified MMMEO-539Neoprene by Impingement o f Waterdropm 18

:3.2.3 Resis tance of Modified IMM E0-539 Neopreneto Very High Speed Waterdrop Impingement 20

. l ates White Neoprene and the Standard NeopreneCoatings . . 2

.3, i Damage Mai .% roduced on Gates WhiteN,pz.ene and the Standard Neoprene0Coatingm by Impingement of Oil ; I )1edGelat in Capsules and o f Deforrl,.Lg ueadPel l e t s . . 4 27

.- 3.2 Damago Produced on Gates White Neopreneand tho Standard Neoprene Coatings byZmpingment of Waterdrops . . 32

4AJ.: . r ,-1 [ i't XIII iv

8/13/2019 Ad 0225488

10/102

TABLE OP CONTENTS (continued)

Section Page

3.3.2.1 *Gates KV-9433 White Neoprene 32

3.3.2.2 Goodyear 23-56 Neoprene overBostik 1007 P1imer . . 34

3.3.2.3 Gaco N-79 Npn~rene overGaco N-i5 Primer . 41

4. Poss ib le MoCes of Fai lure o f the Neoprene To p c a t 45

4.1 Rubber Abrasion . . - 45

4.1.1 Effect of Hardness in the Water Used

fo r the A r t i f i c i a l -Rain and of theUse o f a Wett ing Agent in the Water ' 48

4 .1 .2 Effec t o f Detergent\\Applled to th eCoating - . 53

4 .1 .3 Eff ec t of draph i te and Oil Applied to

the Surface of the Specimen . . 56

4.2 Chemical Deter iora t ion . 61

4.3 M'echanical Fat igue . . . . . . . 68

-. Menhats. of Rain Erosion of INeoprene Coatnrgs * * 7:

5.1 Waterdrop Impingement St r e s se s and the Response..... - t u r a Mater ia l s - 72

5.1.1 St r e s se s 72*

5'.1.2" Response oqf .l ,, 76

-i ?allure of the Sutstrate-PrImer--Coatir.g- -e q. tY78

;BESTAL]-AEt XIIIC

BESTAVAI[ABtECOPPY

8/13/2019 Ad 0225488

11/102

TABLE OF CO.NTENTS (coatlMnd)

Section Pace

5.2.1 Loss of Adhesion Due to th eCcepressive Stress . 79

$..--2rv--Loss o f Adhesion Due to ShearSt r e s s . .. . 80

5.3 Fa i lu re o f the Neoprene Coating so

5.3.1 Failure Ir.luced by the UoConditions . . . .

5 .3 .2 Fa i lu r e o f the Neoprene Coat lngas a Result of Vate r i rop

Impingement . . , . -25.3.2.1 Abrasion . . 82

5.3.2.2 Ozone-type Cracking , 83

5.3.2.3 Effect of Rate ofRecovery on FatIC1eLile . . . . 85

REFERNCES . . .*

W DCM :53-192 Pt XnII vi.

BESTAVAILABLECOPY

8/13/2019 Ad 0225488

12/102

LIST OF ILLUSTRATIONS

Figure Page

1. Damage Marks Made by Impact of u i l -F i l l ed GelatinCapsules on 1/8-in. Lucite Sheet . . 5

2. Damage Marks Made by Impact of 0il-Filled\ GelatinA Capsules on 1/4- in . Lucite Sheet . . . 6

3. Abrasion of MW EC-539 Neoprene 8

1. Waterdrop Impingement Abrasion of 4MU EC-539 Neoprene 10

5. Structure of Waterdrop Impingement Abrasion of EC-539-Neoprene * Ii

6. Structure of a Single Daneve Mark P-roduced on PM EC-539Neoprene by Impingement o f Waterdrops a t a Velocity of880 f~i e . . . . . .i

7. Damage Marks Produced by Oil-Filled Gelatin Capsulesn Collision against Three Neoprene Coatings ofDifferent Properties a t a Velocity of 900 ft/sec . 16

8. Mechanical Abrasion of MMMI.-:39 Neoprene . 17

9. Views of Waterdrop Impingement Damage on Coating-C 19

10. Waterdrop-Impact Fa i lu re of Coating-A . 22

11. Waterdrop-Impact Fai lure of Coating-B . 25

12. Watezdrop-Impac. Failure of Coating-C . . .

S1-esponse of Thri Neoprene Coatings to Impingementwith an O i l - F i l l e d Gelatin Capsule at-' Vnlocityo f 900 ft/set . \. 28

14. Response of Three Neoprene Coatir.us to Impingementwith Deforming Lead Pellets at a Velocity of 490ft/sec . 29

WAZC ,3-192 P t X II I v i l

A41IBE COpy

8/13/2019 Ad 0225488

13/102

8/13/2019 Ad 0225488

14/102

LIST OF ILLUSTRATIONS (continued)

Figure Page

27. Stesses that Result when a Waterdro, Runs overa Surface Prot rus ion -

28. Stresses that Resul t frorp the Coll is ion of aWaterdrop with a Rubber-Coated Surf_: . . 75

BESTAVAILABLECOPY

X111K

8/13/2019 Ad 0225488

15/102

1. In troduct ion

Of the large number of hard-.set t ing plas t ic coating3 anc.

of r e s i l i en t rubber and synthet ic rubber coat ings t ha t havebeen t es t ed for ra in-erosion res is tance, neoprene has beenfound to be one of the niost eros ion- r i sBi t an t up to impingementvelooitiet of 500 mi/hr. Air t ra ff ic will b carried onfor many years to come in the alt i tude range in which rainis still encountered and in the velocity range for whichneoprene is a solUtion to the rain-erosien pzoblem. Themechanism by which neoprene eventually does fail underwaterdrop impingement is, therefore, of considerable mmvdiateinterest .

Neoprene i t se l f does not adhere strongly to glass reinforcedplautic laminates. It must be bunded to the laminate by aprimer coating. The success of the ne6prene topcoat Irresist ing high-speed rain impact depends strongly on thesucces, of the topcoat-primer-laminate system. The vain-erosion resistance not only of tne neoprene topcoat but alsoof the topcoat-primer-laminate system rn'ast, therefore, beassessed in evaluating neoprene as a rain-erosion resistantmaterial.

Maj;,;,rrpt veleai3ud by the author June 1959 for pub l ica t i I 8. 1'a VIAD T r Phnil X1port,

8/13/2019 Ad 0225488

16/102

2. Models of a Waterdrop

Damage tha t is sus ta ined by a s t ruc tu ra l mater ia l onco l l i s ion uith waterdrops a t high speed is a di rec ;consequence of the impact pr ope r t i e s of . a waterdrop. Underimpact condi t ions a waterdrop behaves as thoigh it were asphere of hard-mtertal; in high-speed co l l i s ions withthe planar surfaces o f so l id s it acts l i ke an inden ter towhic~h a compressive load has been applied. Unlike asphere of hard mater ia l , however, a c o l l i d i n i watt..-!ropr e ta in s its l i q u id proper ty of flow. The rad ia l flow ofar impinging waterdrop exer t s a turning moment aga ins tpr o t r us i ons from the surface o f the so l i d t ha t are in itspath and a shear s t r es s on the surfacp l ayers of the s o l i daround the cent ra l poin t of the co l l i s ion . See Sect ion5.1.1. The Use o f models to reproduce one or more of thesedamaging a t t r ibu te s o f a waterdrcp in high-speed co l l i s ionswith so l id s is very informative. A model of a waterdropmight s imula te its hard - sphe re property or Its proper tyo f r ad i a l f low, o r both.

2.1 Stee l Spbeze_-nd Deforming Lead Pe l l e t s

Stee l apheres and deforming l ead pe l l e t ' have been used asndels o f : . . e rd rCp s in s t ud i e s o f the r a in - osion damagetha t occurs on methyl methacrylate p la s t i c an 1100 aluminum.On co l l i s ion with these m at e r i a l s a t reletaol- low ve loc i tya s t ee l sphere does not f low, a lead p e l l e t flows to abouttwice its or ig ina l diameter, and waterdrops f low to many-11ims t h e i r or ig ina l diameter. Comparison of the damage

-ark!- male by s tee l spheres, deforming lead pe l l e t s , an diteq ,1-i~ in . o l l i s i o n with methyl methac ry la te p l a s t i c an dwith 1100 aluminum has proved to be of value in unders tand ingthe mechaanism o f f a i lu re of these m at e r i a l s under high-speedwatorirnp Impingem-nnt 3-I,2 'a.

An Wt','-a....; rr:d, to u.e deforming lead pelle ts an:a f-, watetrrorr -:olll4ng agains t neoprr.aw-coated,-. It. was fournd, however, that the load pelle ts provided

1 it t ha t wi- t,.',v, r F'" the r.noprcnc oc,' Ino tha t wan

! : z - t r z r. . rvkr ts ref&r to the llteralure references at ,

I ... ... : r D . t

/ITIf1 BES AVAILABLC\BESTAVAILABLE COPY

8/13/2019 Ad 0225488

17/102

used ( the .ating Was completely removed froM the metal'41u1htrate) and it was concluded t ha t a so f t e r waterdropmodel would have to be employed. It appeared t h a t the sof tge la t in c losures t ha t are used fo r the co or ing oil ofoleomargarine o r even the somewhat rubbe ryge la t in c losurest ha t are used as capedles f o r medicinal oils might proveto be sa t i s fac to ry ' , a terdrop models to fre aga i ns t neoprenecoat ings.

2.2 Oi l -F i l l ed Gelat in Capsules

The r es t r a in in g ge la t in c losure co n s t i t u t es an importantpoin t o f difference between these possib le two-phase modelsand a homogeneous l iqu id drop t ha t has no re s t ra in ing skinexcept sur face tension to resist its flow, It was an t i c ipa tedtha t t h i s objectionable d i fference would be l e s s importantin the very so f t ge la t in closures used to .zntain th ecolor ing oil fo r oleomargarine than In tVe more durab. erubbery ge la t in capsu le s used fo r the medicinal o i l s .After ex tens ive correspondence with the Gelatin krmductsDivis ion of the Scherer Corporat ion , however, it was foundt h a t so f t ge la t in closures t h a t would be small enough to

enter the b ar re l o f the Benjamin Frankl in air rifle tha twas to be used were not aval lab le , and tha t , due to produc t iondifficulties it was doubtful whether such a closure oftha requ ired s i ze could be manufactured.

It was decided to use g e l a t i r -apcules containing ha l i bu to i l t h a t were found to be ava i l ab l e on the loca l market an dt ha t were s u f f i c i e n t l y small to enter the Iun barre l . Theundes i rab le cha rac te r i s t i c o f the more durable g e l a t i ncapsule o f i nh ib i t ing the normal r ad i a l f low o f the o ilmay be p a r t l y overcome by soaking the capsule in waterbefo re f i r i n g It.

The v a l i d i t y o f us i rg oil-filled ge la t in c a p s e e s oft h i s kind as a model fo r waterdrops was t e s t ed by f . r i n g themagains t methyl methecryla te p la te s o f d i f f e r en t th ickness .The type of 4 .amage mark tha t forms on methyl methacry la teas a r e su l t o f Impingement with s tee l spheres deforminglead pe l l e t s , and waterdrops is knowh [1, 2) . A high-spet 'dmovlng picture was a l so taken o f an oi i -fei-d ge ta t incapsule colidiii with a methyl methacry~ate p la te .

40AD, TR 53-192 P t XII I - 3 -

BESTAVAILABLECOPY

8/13/2019 Ad 0225488

18/102

Damage marks t t la t were made on 1/8-in , and on 1/4-in.Lu,'1te sheet are shown in Figures 1 and 2. Two points ofsimilarity between these damage marks and those hat wererrnduced on methyl methaerylate by impingement o is tee lspheris, def'rming lead pellets, and waterdrops a e that:

the damage mark consists of a circle of crazing and(b) the center spot of the collision, which is undercompression during the collision, is undamaged. Ihese3imilarit tes in the appearance of the damage marks indicatethat the mechanism by which the marks were produoed is thesamc, A

There is nn evidence of damage to the methyl methacrylateplastic as a result of the radial flow of the oil. In theease of collisions of deforming lead pellets -and of waterdrops?iit~n methyl methacrylate plastic the radial flow of theprojectile c a u s e a e n i n g of the craze cracks and arreaking out of material along the craze cracks in thedirection of the flew of the projectile n.1 2j.

From the high-speed moving picture of the collision ofan oil-f i l led gelatin capsule with methyl methacrylate plasticat a velocity of about 720 ft/sec it appears that the gelatincapsule acts as an efficient rebraining case for the"nontained oil. The gelatin capsule appears to burst duringthe col!_,.Lnn a t one or more of i ts weakest points and the oil,which must be under pressure, seems to be atomized or vaporizedthrough the resulting holes. This behavior is altogetherdifferent from that of an Impinging liquid drop,

Although the oil contained in the gelatin capsule doesnot undergo the r A.Lal-flow that a liquid drop would undergoan a zesult of such a collision, the gelatin capsule doesarpear to s.imulate the hard-sphere property of n impingingliquid drop without cutting a soft rbbery coatit entirelyoff the metal surface to wnich it was applied. It can behoped that the stretch of the gelatin capsule as it flattensagainst the surface of the solid may exert a shear stressthat will simulate the shear stress exerted by the radialflow of an impinging drop of water.

3. Response nf Several Different Neoprene Meed. Uoatings

In order to determine (a) what properties of a ncoprerpcoating operate to e.tablish i ts resistance to high-speedwaterdrop ImpIngcoment and (b) what properties of a neopren(I

WADe, TR 53-l9: Pr, XIII - -BESTAVAILABLECOPY

I I

8/13/2019 Ad 0225488

19/102

7 U )

4-4-

~' tii!

w

8/13/2019 Ad 0225488

20/102

8/13/2019 Ad 0225488

21/102

coating lead to its eventual fa i lure wuaer high-speedwaterdrop implnement, the response of several differentneoprene coatings wes studied. The coatlngs were:31 C 539, MHM EC-539 modified as to curing temperaturt and"curing time (Coatings-A 1 -B, and -C), Gates Whlte neopreneKV-9433 and the two neoprene coating systems that have metMIL-C-7439 requirements, Goodyear 23-56 and Gaeo N-79.

3.1 MMEC-539 Neoprene

Three 1/8-in,.thick flat panels and three 1/16-i.thick ai r fo i l shaped Cornell Aer.onalauical Laboratory 'ain-erosion t e s t specimens were sent to the Minnesota Miningand Manufacturing Company to be coated with MM PO-539neoprene coating. The specimens were given one dip coat-of EC-1022 general purpose adhesive asnprimer and two d ipcoats of EC-539 neoprene based coating. The f i lm thickness-,as from 8 -t O mils. The three flat panels were used fo rtes ts by Impingement of oi l - f i l led gela t in capsulea; thethree ai r fo i l shaped specimens were sent to thts CornellAeronautical Laboratory for t e s t by ar t i f i c ia l rainImpingement.

3.1.1 Damage Marks Produced on MW 1C-539 Neoprene byImpingement of Oil-Fi l led Gelat in Capsules

Oil - f i l l ed gela t in capsules were f i red a t a f l a t panelcoated with MM EC-539 neoprene. Each of the oil-filledgelat in capsules *Aat was fire' was soaked in wat..r fcr 2min before it was inserted in the gun barrel. The pointof the gun was maintained a t lapproximately 12 in. fromthe ta rge t panel for each shot. Views of the damage marksthat . were m i d E j - t h e impinging o i l - f i l l ed gela t in capsulesare shown In pictures 1, 2, and 3 of Figure 3.

At low magnification the damage mark that was producoda t an impingement velocity of 320 f t / sec appeared to consistof a very dim, more o r lose circular trace. It is showna t approximately X o magnification in picture 1 of Figure3. When the magnificatiorn was increased the trace was seei.to consist of raised e_-.aa of the coating along shorti rregular wrinkles or cuts.

BESTAVAILABLECOPY

WADC,TR 53-192 Pt XIII - 7 -

8/13/2019 Ad 0225488

22/102

rf-

0,4

IN)

.................

00

N,.

J" .4 4) .-

[- 1-4 I-.

1 41-.,

.. .. t- '

-..- . . . -. -. _ _ __ _ __-_ __ _ _ -- . _ __ _ -

8/13/2019 Ad 0225488

23/102

The damage mark made by an oi l - f i l led gelat in capsulea t an Impingement velocity of about 720 f t / sea was similarin general appearance to that made a t an mapingementveloci ty o f 320 f t / sec. However, the damage was moresevere; a t low sagni f ica t ion the circular t race appearednot cnly to consis t of a removal of gloss but of distinctva2'.e1s. When the magnification was Increased the damaeappdaxed to be a voarse I r regular wrinkling as though thesurface sk in -e i costing had been given a two-dimensionalstretch to the point or permanent net and then released.In some areas the wrinkling was more retgu and took onthe appearance of more or less para l le l rfdges. The edgesof same of these wrinkles were so sharp tha tI t seemed pecsiblet ha t they might be edges of coating tha t had curled or turedup along an array of cuts In the coating. Picture 2 of7igure 3 shows t h i s damage mark a t epproximately Xnomagnification.

Thq damage mark that resulted from the Impact of an oi l -f i l led gelatIA capsule a t an Imlingement veloci ty of about 900f t / sec had the same general appearance as the damage Marksmade a t the lower Impingement veloci t ies . However, th eImpression tha t the wrinkles are the ro l led back edges ofcuts In the coatlng is even stronger. PIcture 3 of Figure 3show t h i s damage mark aT app IMte l 10 magnification.ihan zhis damase mark was viewed with a stereomicroscope itcould be seen that It Is not flat as It appears to be In th epicture. The unrsmaged center Is not depressed but tae circleof wrinkles or outs In depressed.

3.1.2 Damage Produced on P M SC-539 ieoprene by MpMiemento f sterd-rops

An a i r fo i l shaped specimen of 1/16-in. aluinum al loy thatwas coated with MM E0-539 neoprene was tes ted on the CornellAeronautical Laboratory rota t ing arm tes te r rn -I./hrartificlal rain for 1.5 min at a relative veoolty of 880ft/see (600 m1/hr). Inspection of th is spec 1 e n showed thatit was marked with what appeared to be c i rc le s of damsae,Scae of these are sho* Inspicture 4 and 5 0 f Figure 3.r i c= s a t h i g e r ma=nifcatlon, Ir vYh-h -a--1r. or th eatr U r c of the damage can be seen, a. slmof in Pigure&4, 5, eW 6. The st ruc ture of the dmaage o f vhch th ecirc les are-rsde up strongly r4sembles the wr1mrng icutting of which the circles produced by the impact ort heoil-filled gela t in capsules were made up. T e st ruoture

VADC TR 53-192 Pt XII -BESTAVAILABLE COPY

\

8/13/2019 Ad 0225488

24/102

* js:

VSLOCrIf 85Y Fl/SFICjcP

6 Iz .11.

AZ r

BEST ~ ~ *AVIALECP

8/13/2019 Ad 0225488

25/102

~~T-

BESTPL IL BLECOPY

8/13/2019 Ad 0225488

26/102

*Ik

8/13/2019 Ad 0225488

27/102

Is similar to the abrasion pat tern for rub~er tha t has beenreported by Sohallamach no* 5# 6.7, See etion 4.1.

It in possible tha t these circ les of abrasion may becaused by the radia l flow of Individual waterdrops af te r th eneoprene surface has become weakened Dy other watere&opblows tha t were not themselves able to cause vis ib le damage.The .iroles are not caused by every waterdrop tha t impingesagainst the specimen because. If th i s were the case, therewould be so many circ les of abraslon af ter 1,5 min of tes ta t a ',elocity of 600 mi/hr In 1- ine /h r r z.L that they wouldoverlap. If these clrclese are caused by the radial flow ofwater from single drops, the wrinkles or trenches of whichthey are composed should be perpendicular tu radi i of th eflow. This may be t rue of the circ le of abrasion In pic ture1.of Figure 5, but In the ci rcle of abrasion shown in picture2.of Figure 5 the trenches seem a l l to be oriented in th esame direction.

These circ les of abrasion may have an entire ly differentorigin. They may be simply ralsed circ les in the coatingthat were left when bubbles in the coating opened duringits cure. Any protrusion above the planar surface of th ecoating would be more subject to abrasion by the flow ofwater over the a i r fo i l shaped specimen tha4 the planars " , a c e i t se l f . On th i s picture of thwe aC n of these marksit would be expected that. the treenohas for a given circleshould have about the same orienl.: and chat t1 : s sameorientation should be seen on other l es s symmetrical non-circular protruding areas. Picture 1 of Figuro 6 providessome evidence for th i s explanat Io n of the )Aigin of thesecirc les of abravion. The fac t that areas exis t that are notcircular but tha t are still marked with the more or eess paralleltrenches Is evidence in favor of th i s exp lana t ion . \ \

3.2 Modified 9M BU 539 Neoprene Coat.ngs

Flat pla tes and Cornell Aeronautical Laboratory ra in-erosion t es t so ens of aluminnm alloy were aent to th eMinnesota Miinig and Manufacturing Company to be coated withthree neoprenesof different properties-. T o nebprene f ilmswere applied by dip coating to a dry de.- of 10 milsbefore cure which produced a coat l ig thiolkness of 8 to 10mils after cure. The coatings tha t were applicd are lesignated

as Coating-ACoating-B, and Coating-C. The different. inthe propert ies of these coatings was produced by varyirg tne

WAD, TR 53-192 Pt XIII B A13EC

BEST AVAILABLECOPY

8/13/2019 Ad 0225488

28/102

cnirlnjkconditlons of MM EC-539 neoprene plus EC-566 accelerator.Coatin -A was cured at 140OF for 72 hrs; Coatn-B was cureda t 210 F for 8 hrs; Coating-C was cured a t 275 for 1 hr.

The ten31le e t2*nth , per cent elongation, and shear strengthof the coatings that remilted from use of these curing conditionswere determined by the Minnesota Mining and I1autacturingCompany. A pandulum-type tens i le tes te r was used to determinethe tens i le strength and the percent elongation. One-inch la pshear bonds were pulled In . pendvlum machine to determine th eshear strength. The lap shear bonds were prepare-d aandwichingthe neoprene between alumlnum sheets using EC-1022 as the metal primerand curing under the specified conditions. The true shear s t rengthnf the EC-539 neoprene f ilm was not measured because fa i lure alwayoccurred ei ther a t the primer-to-coating or a t the primer-to-metalbond. The t ens i l e strength, percent elongation, and shearstrength (of the adhesion bond) data for a lO-mil fi lm thicknessare given in Table 1.

Tat .e 1.

Curing Schedule and Physical Properties of Coating-A, Coating B,and Coating-C

Curing Schedule Tensile Shear Elongation

time temperature Strength Srength+

h r A ps i psi pe r cent

Coating-A 72 140 2,700 \,250 170

: , - , tng-B 8 210 500 137

1 275 45800 500 96

'T.hear st rennhof e adhesive bond

Ji.DC TR 50--92 Pt XIII 14 BEST V IL BLECOPY

8/13/2019 Ad 0225488

29/102

.3.2,1 Damage Marks Produced on Modified HW4 30-539 Neoprene- by Impingement of Oil-Filled Gelatin Capsules

Oil-filled gelatin capsules were fired from a BenjaminF:-zi~lini &1r rifle at flat plates coated with modified PMM EC-539naaprene (Coating-A, Coatirig-B, ana Coating-C3). The platescarrying the neoprene coatings were diamped against a backingplate, The gun was held 12 ins from the neoprene coated plate

andl the gelatin capsules were fi.redat

the1l o at 900 Inciden2080

The gelatin capsules were poaked in water for P Iit before the ' ihotfwere made, Shots were made at velo6itles or aproximately 320,pT20,, and 900 ft/sec. Microscopic inspection of the spots thatware i~ruiok by the gelatin capsu~es provided t~ie followinginformation,

On Coating-A no marked damage was'observed as a result ofthe shot made at a velodity of about 320 ft/sec. The *hot made&t a velocity of approximately 720 'V/uec produced a ha.rdlydiscernible semioizgole of what appeared to be auts or wrinklesin the coating. The shot made at a velocity e~ about 900 ft/secproduced a complete circle of what appeared to be cuts orwrinkles in the coaibing., A view of this damage mark is 3hownIn picture L--of-1-i~tre 7.

Q. Coaving-B no damage was produced by, the shot3 that werematdr, at the approximate velocities of 320 abd 720 ft/sec.The shot made at A velocity of jabout 9n ft/sec produced arcso~f what appeared to Ue cuts or wrinkles In the coating. Avie~w of this damage mark Is shown In pictidre 2 of Figure 7.

On Coating-C no marked damagte was produced by the shot made.ta velocity of about 320 ft/sec or by the shot made at a

i'#loni t~y or appoxmtely 720 ft/sec. The shot made &t a

Ift1ocity of ab~out 9"00 t s o p o d u c e aros ofwhat appearea to

',.e cuts or wrnlsin th otn.This damagte' was siaroundedhy a large blister of ,cbating. A view of thin damage maric is

"~'wziat two magnifications In plcteres 3 and 4& f ftguro 7.

T. -t ' s ea t ion of Coating-A contann the damage mark shown1::. picture I of Figure 7 was out loose fk'w h aa~mrimn plate

:1 .1-aO oirole of damage was Inspected .. nit oaffni4cationsTO.r"Je1 of damage was seen to oonklit or z'#ised ridgdsrAges of m'bber, The some typ~e of damage (raised flaps of rubbe ')

i11.iults w'hers a sha~rp poin~ted marking' pencil or a razor 1*. .de isdreggeet ex.-resh the ntsuprene surftace. Such kinds of damtWQ are#J'J-i,.rxin Figure 8. This type of damage to rubber has bt~en divcussedby Schellawach EJ 5 7 a d h s en4kfo te r iued as rubbert,.J-slon. Throuhoout tile remainder of tnis repo~rt Wei +Ypoeof derage us, he surface of a neoprene oa~ting will be referredt-) as rubber abrasion, The way in which this abrasio & is producedin disrateaed in Section 4*.l

U~DCTR 5-19t liiBEST V IL BLECopy

8/13/2019 Ad 0225488

30/102

8/13/2019 Ad 0225488

31/102

hjJ5O.BY T ~ i i r i - J I I E M A I A E I N G PL-NCtL MOVING DjO'N,42D .

Btb3ciHAZO~R F LGV IN FNjo1 WL TO' HtG'A't

8/13/2019 Ad 0225488

32/102

The pactu res o f Figure 7 show t ha t oa l y Coating-C faileddue to l o s s o f adhesion. Frsom Table I It can be seen thatCoatirg-C had the lowest perceno e longa t ion o f the th reecoat ings. If a rubber coa t ing has a high degree o f rigiditya shear s t r e s s exerted on its surface may be t ransmit ted through-It to the adhesion bond and If the s t r es s . is m i f i c l en t l y grea tthe adhesion bond w,,. fa l l . The area o f uoat ing t h a t was given,,--a- ad ia l s t re t ch to the po i n t o f permanent se t during th ecollision winl then be raised from the surface to which ithad been bonded in the form o f a coa t ing bubble o r coa t ingblister.

It wi l l be seen in the fo l lowing sect ions t h a t damage markecomparable to those shown in Figure 7 are produced on thesecoat ings b1 v e y high-speed waterdrop impirgement and thatthe use o f oil-filled ge la t in capsules as a zode l fo r waterdropsin very h igh speed co l l i s ions is f u l l y J u s t i f i e d .

3.2.2 Damage Produced on Modified XMM EC-539 Neopren.Impingement o f Water,4s

Hine airfoil shaped r a i n - e r os i on specimens were poatedwl.t,h MMM EC-539 neoprene by the Minnesota Mining and WanufacturingCompany. The cur ing schedule used fo r these specimens\was sucht ha t th ree of the specimens were of Coating-A, three o f th especamern dere of Coating-B, and th ree o f the-specimens were ofCoating-C. These specimens were t e s t ed on the Corne l l AeronauticalLaboratory rota t ing arm t es ter at a vele,-.ty of 600 mL/hr L,. l -on . /h r artificial-ain-. The three specimens o f each coat ing weret e s t e d f o r 25 sec, 1 min, and 2 min, respect ively. The visualnppearance of these specimens a f t e r test was repor ted . Theopecimens were returned to the National Bureau of Standards fo rStudy.

Examination of the specimens o f Coating-A, Coating-B,and Coating-C t h a t were t e s t ed f o r 25 sec, 1 mi, and 2 min pro-vid-ed the in fo rma t ion t h a t the coa t ings on all o f these specimenswere characterized by I r regu la r patches o f the small more or0less rarallel shal low t renches o f rubber abrasion. In manycases these patches were c i rcu la r o r consisted of arcs o f c i rc l e .Plt ture 1 -if Flmure 9 is a view a t high magnl ; .Lc ' 'on t ha t showsI . h e pa ra l l e l - t rench s t ructure; t h i s p le tu re ouss taken on apeclmen of Coating-C. It is difficult to take p ic tu res at

high magrftAi' a l . in -on the r a l n - e r os l on test specimens becau:,o! 'their curvature.

WAW TR 53-192 Pt XIII . - i0 -

\SECO

8/13/2019 Ad 0225488

33/102

,vot Ol. < > N 3

I414

8/13/2019 Ad 0225488

34/102

The quan t i ty of these patches o f abrasion Increased inamount as the tes t time for the coatings varied from 25 sec to 2 mine"There seemed to- -- as much or more of this rubber abrasion withinany given period of test in the order of Coating-A least toCoating-C most. In add i t ion to being bese t wth the rubberabrasion, the specimen o f Coating-B t h a t was t e s t ed fo r 2 minhad lost adhesion on the high-speed end with formation ofbubbling; it had a l so to rn open there.

The waterflow from in te rcep ted drops fo l lows curvedt r a j ec to r i e s on the airfoil shaped specimens; tUe f1ow runsoff on both s ides o f the leading edge. The curved t r a j ec to r i e swere clear ly marked by the more o r l e s3 p ara l l e l t rench s t r uc t u r eo f rubber abras ion on the specimens of Coating-A, Coating-B,and Coating-C that were tes ted f o r 2 min. Pic ture 2 o f Figure 9is a view of these curved t r a j ec to r i e s on Coating-C; th eleading edge o f the specimen is on the diagonal from upper leftto lewer right-11TFle pictu re . The c i r cu l e r nature o f manyof the patches o f rubber abrasion can be seen in th i s pitt.-ure.

The evidence that has been presented appeiws to i nd i ca t et ha t the damage t h a t inc reases with t ime on this pa r t i cu l a rneoprene coating system f o r the test veloci ty and r a i n densi tyt ha t were used is the rubber abrasion which appears to beprogress i r- t ? poin t a t which it wi l l cover the en t i r e l ead ingedge of the specimen. *The percent e longa t ion o f these coatingsis in the order of Coating-A most to Coating-C l e a s t ; th edevelopmx-nt of rubber abras ion on the su r iace of the neoprennappearec to be in the order o f Coating-A l e a s t to Coating-Cmost. From t h i s evidence It would appear t h a t a neoprenecoat ing hav5ng a high elongation prope r ty should be more r a i n -erosion r e s i s t a n t than a neoprene coating having a low e longa t ionpropert-. Coating-A had the lowest t en s i l e s t rength of the th reecoatings. This, however, does not mean t h a t t en s i l e s t reng this not an important property in a rain-erosion resistant coating.A winimum tensile strength 10 necessary and it Is possible thatIf Coating-A had' had a higher tensile strength it would havebee-n more erosion resistant .

3.P.3 Resistance of Modified 4M1MEC-539 Neoprene to VeryHlh Speed Waterdrop ImpinLtr.neL:u

Specimr.nd for use In waterdrop impingement testa at veryhigh velocity were coated with MMMEC-539 neoprene by he M!i',eso 4;aMIning and Manufacturx.-g Company. The curing Bchedules that ",',rotised produced Coating-A, Coating D, and Coatin-C on the burfaces

WADC TR 53-192 Pt XIII - 20 "B3T AVAILABLECOPY

8/13/2019 Ad 0225488

35/102

to which the MM EC-539 nboprene was applied. These specimenswere sent to Convair where very high velocity impingement t e s t swere carried out.

The damage marks made by approximately 2-=, waterdrops onCoating-A a t a.n impingement velocity of l ,540 f t / sec are shown

In picture 1 (o) 'of Figure 10. Dimensions of the damage markclosest to the edge of the specimen are shown in sketches l (a)and l (b) a t the left of the picture. The damage mark consistsof a oircOlar depression bounded by a circle of cutting.See cross section In sketch l (b). The oirouiar depressionmarks the region of maxImun pressare exerted by the collidingwaterdrop. The bottom of the circular trough Is roughlyhalf a radius from the stagnation point (center of the col l i s ion) .The diameter or the ci rc le of cutting Is approximatel- the sameas the diameter of the waterdrop tha t produced the dabiage mark.In general, the damage mark Is similar to tha t produced byimpigement of an o i l - f i l l ed gelat in capsule a t a velooit,) ofabout 900 f t / sec. See picture 1 of Figure 7,

There are two poesible ways In which the c\ role of cut t ingmay be produced. (a) Because the coating siaffers severe com-pression, a shallow disk-shaped cavity may form in It. If thishappens, strong t ens i l e stresses wil l exis t in 'he sharp kneeof cc-' .. at the periphery of the cavity. These t ens i l es tresses may be responsible for the circle of cutting. However,the high percent elongation of Coating.- makes ' th is exp anationseem unlikely. It should be poksible wo bend th is coating i n toa very sharp knee without producing cuts. (b) A second'explanation Is that the circle of cutting is produced by th eshear s t ress exerted by the radia l flow of water from th edrop during the coll ision. On this explanation the circle ofcutting is severe zrubber abrasion. Inspection o f the s tructureof the circle of cutting with a stereoscopic microscope showedthat the structure of it Is completely comparable with tha t ofrubber abrasion.

The damage mark produced a t an impingement velocity of2,480 f t / sec is shown in picture 2(c) of Figure 10. Thisdamage mark consists of a central mound in the dott ing surroundedby a circular fold of coating. Dlmensbna of "%e damage markare shown in sketches 2(a) and 2(b) a t the left of the picture.

WAD" TR 53-192 Ft XIII - 21 -

BSTAVAILABLECOPY

8/13/2019 Ad 0225488

36/102

.(a) W-

I b)

PAUI~~~~~~R #,WISSdS~ VY PROWP/

Ali~~A

wC~h1-29utXle

8/13/2019 Ad 0225488

37/102

It can be seen t h a t the diameter of the eircular f o l d o f coa t ingIsamioh l a rge r thn the diameter of the c i r c l e o * c u t t i n g(rubber abrasion) seen In p ic tu re l(c). Inspebt ion o f tJilsdamage niark with a stereoscopio microscope revealed t h a t th ec i r o l e o f cut t ing ex i s t s a t the base o f the fo ld o f coa taand Is concealed by the overhang o f the fo ld itself. Althoughthe r ad ius o f curvature o f the fo ld Is very small there are nocuts in the coating along the top o f the fo ld .

It would-appea- tha t the first s tage In the formationof this damage mark was the same as that IeTn picture (c)'.The shear s t r es s exe r ted by the r ad i a l f low o r water from the dropwas apparen t ly great enough to break the primer bond and movethe coating out in to a r ad i a l s t r e t ch th a t was severeenough toinduoe permanent set . The f ac t t h a t there are no cuts in th eacute bending of t h i s fo ld o f coating is evidence t h a t th ec i r o l e o f cu t t ing shown In pictu re 1(o) Is rubber abras ionand is no t the re su l t o f a t ens i l e f a i l u r e at the p e r i ' h e r yo f a disk-shaped depress ion ,

The damage mark t h a t was produced a t an impingementveloci ty o f 2,588 ft/teo is shown In pictu re 3(c) o f Figure 10 .It oonsis ts o f a a i r cu l a r spo t o f primer surrounded by a circulara rea o f bare metal, Beyond t h i s is a reg ion over which th ecoat.i has beeo-rbrpped from the primer. Dimensions o f th edamAge mark are shown in sketches 3(a) and 3(b) to the lef t ofthe picture. From a comparison of the dime"lona shown insketches 2(a), 2(b), and 3(b) it c.: are thatbthe diameter of thec i r cu l a r spot o f primer in p i c t u r e 3(0) corre4\;ond to thediameter of the cen tra l ra ised mound in pictu re ' 2(c) and tha tthe outside diameter of the circle of bare metal In picture 3(0.)corresponds to the diameter of the fold of coating in picture2(o), It would appeAr that In the formation of this damagemark the s tages shown In pictu res 1(c) and 2(c) a re reenac ted .At t h i s h ighe r Impingement veloci ty it appbars t h a t th eturning moment exerted bjy the r ad i a l flow of water .om th ecol l id Ing drop Is strong enough to break through the c i r c l eo f cu t t in g (rubber abras ion) at the base o f the fo ld ofc3ating and to rip the coating off the primer. The mechanismby whic.h this may be accomplished is shown schematicaLly inWipure 3(a).

Ds, ehavior, that is observed for Coting-A in this very highvelo.%Ity watcrdrop, impingement Is In agreement with Its proper tyo f high elongat ion. It would appear t h a t whereas the k.ighe longa t ion pr i pe r t y qf Coating-A was desi rab le at an ! .^Ingement

WADC TR 53-192 Pt XIII 3

STAVAILABLECOPY

8/13/2019 Ad 0225488

38/102

valqc1.ty of 8&A-?!'r (600 mi/hr) In that it seemed to re tarda rraiirual rbber arailun, the proper ty is undesi rab le whent. 'e impingement veloci ty Is Increased by a fa*tor o f th ree .

Damage marks produced on Coating-B a t fouzdi ffe rentve loc i t i e s are shown In Figure 11 and damage mar'ks produced onCoating-C at two di ffe ren t v e l o c i t i e s are shown In Figure 12.The damage marks shown in p ic tu re 1(c) o f Figure 10, p ic tu re I of

Pigure 11 and p ic tu re 1 o f Figure 12 were all made at anImpingement v e lo c i ty o f approximate ly l ,5 0 f t / nec . Each ofthese damage marks co n s i s t s o f a c i r c l e o f severc rubber abrasion.Comparison o f them Ind icates t h a t there are minor di fferences Inthe response o f the three coatings a t this veloc i ty, however.In the case o f the damage mark on Coating-A there Is a c i r cu la rt rench o r depress ion within the c i rc le o f cut t ing o r rubberab t i s ion ; in the ease of Coating-B and o f Coating-C t h i s c i r cu la r

0,!-rarsion is essen t i a l ly absent. Tis observa t ion would seent- :0ndlcate t h a t Coating-A Is more sub ject to being deformntcynn1i its ability to recover than is Coating-B o r Coating-C.

nThlei r c l e o f rubber abras ion in these damage marks Is wid(;rf. r ( r a t I f t g - B and f o r Coating-C than fo r Coating-A. This mayi n d i ca t e a grea te r su sc e p t i b i l i t y toward abras ion In Coating-Band CuaLitg-C "taxi In Coating-A

The damage marks shown In plcture 2(e) of figure 10 and- I c tu re . . o f r igu re 12 were made at almost the same impingementvelocity. These damage marks are quite diffe nt In appearanceand are in agreement with the high perc-... elo ation propertynf Coating-A and the low percent elongation prper ty of Coating-C.

The damage mark shown In picture 3(c) of Figure 10 was madeat almost the same Impingement velocity as thosp shown in picture4; if 1gure 11. Although the damage mark mad%,on Coating-A

c.prears to be the most severe the coating in th vicinity ofthe crank on Coatinr-B Is loosened from the primer' and, probably,had the impIngcannt ve loc i ty be-enal i t t le hig*er, It wou.d haveteca torn loons to produce a damag.. mark comparable to t h a t on

C.at.JIng-A. However, t h i s has not occurred at the ve loc i ty atwhich the, damage mark was made and It may be concluded thatCiatin-B3 is more resis tant to waterdropImpingement a t a

.It- *" ~abn'ut 2, s0t/n than Is Coat l - , .A-

-. ; : , ".it" :!.!-)preue and the Stanax' Neoprene Coat ings

,, ni'-rfrli : ; telm,,ns -nd two f l a t ponlzs were ana te r. . t*..,- ":,t,--L 81.~:tro.r4 i.-an, with tUnco N-79 'tn neoprene a:.4

.r',.!, ,.. r.-:: w.--i 1,t.- E. ,-r , r~,rp~ectively. The G(ao N-79

BESTAVAILABLECOPY

8/13/2019 Ad 0225488

39/102

2c r'i r/ Vfl-OACIY 2025 YTr/S$ Q

VV,1WCITY 2250 FT/SSC VELOCITY 2605 FT/S&C

FIGURtE 11 WiiUZri1A(POI'-1MPACT FAILURE OF COATING-B

VALUCITY 1510 FT/SiSC Vic

V13U06 1V WniceJtlIO0P-IH)PA,& FAILIMI. 10,04Z VA 4 e

8/13/2019 Ad 0225488

40/102

tan. eonprene was applied over Qaco N-15 isooy ate primer.The rates KV-94#33 white neoprene was applied @er KV-8600tie-cement and KV-8582 primer. Similar epec vnen were coatedat the Cornell Aeror~autical Laboratory with Goodyear 23-56neoprene over Bosti~k 1007 primer. Physical pr~perties of thesee'n-atlng systems are given in Table 2.

The airfoil specimens coated with each of the threen-~oprene coating systems were tested for rain er L~ion *'euistanceat the Cornell Aeronautical Laboratory. The f la t panels wereused for tests of the resistance of these coating systems toimpingement by oil-fi l led gelatin capsule3 and-by deformiingleaxd pellets.

Table 2

Pilysical Properties of Gates White Neoprene ~nd the StandarO9

Neoprene Coatings

coatin-g Syte Physical Propertles-~ -e 6'ff-W~ie ee -F

Prrtinner Cc.-icnt Strength Adhesion at Break

'a'n 11-79 qcI-52,OpM 25-4io 750

'at~s-h~iteGatesYV- 'IY V-582 Kv-e60C) 3,0o26 2-4886

BESTAVAILABLECOPY

w~r~:. 29 t. XIII- 4-

8/13/2019 Ad 0225488

41/102

3.3.1 Damsae Marks Produced on Gates White Neoprene and th eStandard Neoprene Coatings by Impingement of011-Fil led Gelatin Capsules and of DeformingLead Pel le ts

The ol l - f I l lod-Se la t in capsules tha t were used for the shotswere "*Saked in water for 2 min. They were fired a t the coatelpanels

fran a Benjamin Franklin a i r rMfle. Th distance fr=the muzzle of the gun to the target coating w" 2 ft. Wieea 8sules were f i red a t veloci t ies of 320, 720, i7O, and 900ft/sec.

Inspection a t low power with a stereoscopic microscopicshowed that very little damage was done to any of the ooatedpanels by Impingement of the o i l - f i l l ed gelat in capsule,t veloci t ies tunder 900 f t / sec. Evidence of loss of adhesion

. d d te seen In the whiite neoprene a t an Impingement veloci tyof 870 f t / sec and evidence of a circle c f abrasion could beseen in the Gaco H-79 neoprene a t t h i s Impingement te1c'lty.Magnified views of the marks pr-Auc-4 on the three coatingsa t an impingement veloci ty of 900 f t / sec are shown In Figure 13.

The Gaco N-79 coating appeared to be the l eas t affected.A circular dent In the coating was observed. The Goodyear 23-56reepre- n g was abraded around the central point o f Impactposs~bly due to the scraping of the capsule as It spreadradia l ly on the surface during the collision. The Gates XV-9433white neoprene coating f a i led in adhesion and bubbled or'' lifted around the central point of the col l is ion; It appearsalso to have suffered some abrasion.

Deforming lead pel lets , which flow when they coll ide witha salid surface a t igh velocity, were also

f i r~d a t uhe neoprene--a ted panels. The pel l e t s were red from the =e n FranklIna i r r in le; the pel le t .- ' - c i t i e s 9e0, 5 0d640 f t / sec .

Magnifled views of the damage marks produced on the three:voprene coatings a t a pel l e t veloci ty of 490 f t / sec are shown InFigure 14. In each case a circular collar or bubble of coatingwis raised about the central point of the col l ls ion . The Gaco?1-79 Neoprene coating appeared to- suffer It %-. t avag int-t the en t In the coat i was a sen-c~oe r a t h e r thaz \a ecomplete

A T -E192t 11 - 27CP

\ BEST V IL BLECopy

8/13/2019 Ad 0225488

42/102

GOODYUi 23 5

S'TANU4RD NhOO1IREM

QACO N-79

STANDARD NEOPflZNE

OATO XV-94+33

lpIn3,OF -rImr'?', Itoirntili C~OATTros5 to 4 1 M W W~Tit

ANTL-FILtM)3 (,PLfr'IN CAll"ULS AT A VgE '.T1Y OF, 900 5,/eIV

8/13/2019 Ad 0225488

43/102

Q~ 00 Li

r44

6A4

6.4

8/13/2019 Ad 0225488

44/102

The cire'-uar patch ofr;,-atIrig at, h n e fthe damage mark was cut completely free of the \remaInder ofthj- coating in the case of the Goodyear 23-56 n oprene and inthe case of the Gates KV-9433 white neoprene. The whiteneoprene appeared to suffer the most extensive damage; a sectionof the coatlng wes torn open. Through the open hole In thewhite neoprene coatine It appeared that the Mdhsoion railure-was between the neoprene and the underlying coio ( t l e-"e me n t and primer).

Virews at about 2X magnification of the damage . s thatwere p rdu*ced on the three coatings at pel let velocities of 530anl 6 ) t /sec are shown in Figure 15. The Gaco N-79 neopreneappeared ti suffer the least damage and the Gates KV-9433white neoprene appeared to suffer the most extensive damage.

At a vel~nity of 530 ft/sec the raised collar or bubbler f !cating In the Gaeo N-79 neoprene was of about the samed' .meter as that produced In It a t a veloe i t to f 490 "t/sec butthe cut In the coating was a complete circle. The r&ised zection(,f -natlng in the Goodyear 23-56 neoprene was nearly twice aslarge as that pr3duced in Gaco N-79 neoprene a t this velocity.ThIs may Indicate that the adhesion bonds between the metal and prlmeand/or bctween the primer and topcoat of the Goodyear-23-56-Bostik-1r-7 syste= are weaker than those of the Gaco-N-79-Gaco-N-15 system.The adhes: .. fa : lure of the Goodyear coating system appeared totE between the primer and the metal. The raised section of coatingin the Gaf 1es KV-9433 white neoprene was me"e than four timesl iager tohan that which formed in Gaoo iN-(q neoprene a t thisve~i,,ty. There was no tearing of the white neoprene topcoat

*.he..peripher 7?T-he coating bubble but this did occur In the^ase of Goodyear 23-56 neoprene. This may indicate either thatthe adhes:,n' bonads of the white neoprene system are weaker thantlf se 'If ther Gooyear-23-56-Bostik-l007 syste and permit ther ' rwth e.. a large bubble witho~ut undue st ress t' the topcoat

t,-'.hat the white neoprene" topc-tat has higher strength propertlen.nan theo 1--,dyeair 23-56 neoprene t i p c a t permitting the growth. a iarg, bubble In sp i te of strong adhesion bonds.

At a viV'city of 640 f t /sec there is no tearing mf t#he Gaco:2 t". rnr: n 8.t the perIphery of the aoating )ubble but sitvh

.I ng did onniar both in the Goodyear 23-7 npi. -. e aud in the*1tes KV-9'-:3 white neoprene. There is so.a abrasion of th erei r e surface around the circular nut in the coating in the

rf. t},e white neoprene and ra ther extensive abrasion in .he

WrD,;TU' 53-152 P1. XIII - 30 - BESTAVAILABLECOPY

8/13/2019 Ad 0225488

45/102

IMPIWO"Wrt VZ'EoDITY 3 /880sl EMIN0i'E4ANT VZLOCITY 61#'0 ?TWAKC

N m 1 f t~ ad ram i OACO 14445 PRIMP~

GOODIAM 23-56 NBCPI3RiK, DOSTK 1007 PHI Mai

I mlWl mdr(.oarr V.LvCtra 530 FT/81CC TMPTMIC1M9.1"VlviWCTTY (,40O

(I'L-O -4., jH IV 9 0T MM

_ _ _ _ _ _ _ _Pjj_ _

8/13/2019 Ad 0225488

46/102

case o f the Goodyear 23-56 neoprene, but no abras ion in th ecase of the Gaco N-79 neoprene.

Oil-2illed gelatin capsules and deforming lead pellets whenthey collide with a solid surface simulate the damaging propertiesof a waterdrop when it collides with and flows on the surfaceof a solid. The two damage tools that are active in a collidingwaterdrop are the localized impact pressure and the radial flow-of the substance of the d&op with the concomitant stresses

thatare introduced by each. The damage produced on the threeneoprene coatings by Impingement of oll-f i l led Rel.atin capsulesand of deforming lead pellets suggests some trends that may belooked for in the rain erosion response of these coatings. Withregard to adhesion it can be expected that the white neopr-enemay fail extensively, the Goodyear 23-56 neoprene may failmoderately, and the Gaco N-79 neoprene may show l i t t le , If any,failure of this kind. With regard to abrasion it can be expectedtha t the white neoprene may abrade s l igh t ly, the Goodyear -?-56neoprene may abrade to a cons iderab le extent , and the 0aco 21-79neoprene may suf'er l i t t le or no abrar-ion. (With regar tot e a r i ng f a i l u r e it can be expected t U t the\',white neoprene

may be somewhat suscept ib le , the Goodyear 2 -56 may be th emost suscep t i b l e o f the three , and thi GacoN-79 may be th el ea s t suscep t i b l e o f the three . The extend to whic1 t he sepredic t ions are fulfilled can be seen in the observa t ionsrecorded - the fo l lowing sec t i on .

3 .3 .2 Damage Produced on Gates White NenDrene and the StandardNeoprene Coatings by Impingement o f Waterdropa

Rain-eros ion test o f the airfoil specimens t h a t were coa tedwith the three neoprene systems were made,,on the ro ta t ing armat the Cornel l Aeronau t i ca l Laboratory. Tes t o f the specimenswas to have been made a t a v e lo c i ty o f 600 mi/hr in 1 - i n . / h rra in fo r periods o f 10, 20, 4Q, and 60 min. Pai lu re o f th ewhite neoprene coat I -g wa.s so rapid, however, t ha t two of th e_p-cimens t h a t were roa ted with white neoprene were t e s t ed at

a ve l oc i t y o f 500 mi/hr.

3 . 3 . 2 . 1 Gates KV-9 4 33 White Neoprene

The specimens t ha t were coated with white n o..,rene werenumbered 2119 A and B and 2120 A and B. A picture of the foureroded specitens is shown in Figure 16. Only one specimenwas t es ted a.A. e ioc i ty, in a ra in dens i t y, and fo r a t a ' ein terval comparable v i th t h a t used fo r the other neoprene cta.inr

;WADC TR 53-192 Pt XIII 3 -

BESTAVAILABLECOPY

8/13/2019 Ad 0225488

47/102

40-4 Z

20

Al, X

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -2 8_____. -__ _ _.

8/13/2019 Ad 0225488

48/102

systems, This was specimen 2119 A which was tes ted a t ave loc i t y o f 600 ml/hr In 1 - i n / w r a i n fo r 10 min. The coa t ingon t h i s specimen fa i led d ras t i ca l l y in adhesidn betweenthe primer and the metal specimen-base. The coatInM loosenedto f o m a l a rge bubble t h a t covered about - th ree-four ths ofthe leading edge* I pec t i on of the eroded coating on th i sspecimen with a-i-ereoscopio microscope at a magnif ica t ion ofX20 revealed an In d i s t i n c t chevron pa t t e r n in the neoprenea t the low-speed end o f the specimen and some\abrasion o f th ecoating surface on the leading edge, There wire a lso randomlyspaced gouges in the coat ing along the leading ed'?.e from aboutthe cente r o f the specimen to the high-speed end. The damagedone to this specimen was obvious to the unaided eye whereasthe sped1mmcoa ted with the Gaco N-79-Gaco-N-15 and with th eGoodyear-23-56-Bostik-1007 systems t ha t were tes ted under th esame cond i t ions appeared to be undamaged to t h i s degree ofInspec t ion .

3 .3 .2 .2 Goodyear 23-56 Neoprene over BecsLik 1007 Primer

The specimens t ha t were coated wxth the Goodyear. 23-56neoprene over BU.TW!1007 primer were numbered 2123 A and B an d2124 A and B. A pictu re o f the eroded specimens Is shown inFigure 17. Specimen 2123 A was t e s t ed fo r 10 Iin at a v e lo c i tyo f 600 ml/h r in 1 - i n . / h r rai. Inspec t ion o f this specimen witha s tereosp-- .o microscope revea led a soa t t e r i ng r o f small roundholes In the coating down the lead ing edge. Evidence o f th eformation o f a chevron pa t t e r n in the neoprene was observeda t the low-speed end o f the specimen, A crack in the coatIzj&could be seen at about the cente r o f thb lead ing edge.

Specimen 2123 B was t e s t ed f o r 20 min uxnder the same Oondittonso f veloci ty and r a i n ra te . , Microscopic inspec t ion o f this specimena lso r evea led a scat ter ing o f small round ho les In the coatingdown the leading edge. The chevron pa t t e r n in the neoprene atthe low-speed end o f the specimen was more d i s t i n c t . There wasa -oat ing bubble a t abo i t the cente r o f the leading edge anda humpy s t r uc t u r e a t the high-speed end ind icat ing l o s s ofadhesion. There was some abras ion o r tear ing away of the surfaaelayer espe ' . l a l ly at the high-speed end. There were a l so some su tureao r cracks in the coating mainly a t the hlgh-spoed end o f the specime

Specimen 2124 A was t e s t ed fo r 40 min under the same cond i t ions :o f veloci ty asd r a i n ra te . Microscopic inspec t ion o f this specimen..again revealed a scat ter ing o f small round holes down the i . id l rgedge. The chevron pat tern in the neoprene a t the low-speed end

WADCTR 53-192 Pt XIII " KBESTAVAILABLECOPY

8/13/2019 Ad 0225488

49/102

I']

'1

8/13/2019 Ad 0225488

50/102

Ui

waa di s t ino t . There was strong bubbling o r lifting o f th ecoat ing along the leading edge Ind icat ing l o s s of adhesion,and there was abras ion by souff ing up o f the sur face layerespecially a t the high-speed end. There were a l so somesutures o r cracks in the ooating.

Specimen 2124 B was t e s t ed fo r 60 ian under the samecond i t l o r s o f ve l oc i t y and r a in r a te . Inspection

o f t h i sspecimen with the microscope revealed a strong chevron patternIn the neoprene at the low-speed end of the s p e o i e n . Thecoat ing on t h i s specimen was torn open down v a s t M f th eleading edge. There was a strong abrasicn of the surfacel ayers and a gra iny struwture a t the h igh-speed end o f th especimen.

Some additional data are available for the progress oferos ion on Goodyear 23-56 neoprene S3x epoxy l amina teairfoil shaped specimens were coated with Bost ik 1007primer and with Ooodyear 23-56 neoprene at the Cornel lAeronau t i ca l Laboratory to be t e s t ed along with somenylon specimens fo r the purpose o f ocvparlson The specimenswere numbered 2152B through 2157B, inolusive.\ The coatingsystem was 12 mils tick, Rain erosion tests vwere conduotda t a veloci ty o f 500 mi/hr in 1 - i n . / h r artificial r a in fort ime In te rva l s Of 8, 10, 30, 35 4 and 100 min.

Microscopic inspec t ion o f specimen 2152B hat wast e s t ed fo r 8 min showed the presence o f su tu res o r shor tcracks o r cu ts at the high-speed end o f the specimen.There was a l so a background o f patchy o r spot ty removal\o f a t h I n surface l aye r o f coat lng on t h i s p a r t o f th especimen. In the cen ter of the specimen the sur face glosswar-removed from pr o t r us i ons .

Inspect ion o f specimen 21533 t h a t was t e s t ed fo r 10 min,showed a strngr general abras ion o f the sur face at th e

.. gh-speed end; a t h in surface l ay e r of coat ing was removed.There were a few broken out spo ts in the remain'ng dullsur face . In the cen t r a l area o f the l ead ing edge there wereonly c l ose l y spaced areas In which a t h in l ayer o f coa t inghad been removed. This prov ides a clue as to how the t o t a lremoval o f the surface l ay e r at the high-spe- . e-.." ofthe specimen was accomplished. It appeared as thoughmater ia l had been scraped , to rn , o r pee led from the sur faceIn many c l ose l y spaced spots . Close to the low-speed end

WADC TR 53-192 Pt XIII .6 - B._.. -ssBEST AVAILABLECopy

8/13/2019 Ad 0225488

51/102

of the specimen there were only i s l ands of etchI.= in anLnetched surface. At the low-speed end of the peolmon th e3.jrfa-e was es sen t i a l ly unetohed and only i so la ted patcheso f etching exis ted . Ti l t i n g o f the specimen to iewthe area s l i g h t l y o ff the leading edge showed tha tthere was poor methanical coverage of the g l a s s fab r i cwith the neoprene; cu t t i ng f a i lu re was evident in th e

areas between the woven g l a s s f ibe rs .Inspection o f specimen 2154B t ha t was t e s t ed fo r 30 min

showed t ha t a-eompl-ete surface l aye r had been uniformlyrenoved- from the high-speel end. Many im..11 c ra c . 3 orsu tu res exis ted t ha t could be a general de te r io ra t iono f the coa t ing by u t t ack of ozone of of hydr6xyl ions.At the cen ter and a t the low-speed end of t h i s specimen th esur face l ay e r was simply removed.

Inspect ion of specimen 2155B t ha t was t e s t ed fo r 35 minshowed tha t the sur face l aye r had been completely removedfrom the high-speed end of the s p e c l e n . Severa l very

deep holes and many of the su tu res could be seen in t h i sp ar t of the specimen. The etching away o f the surfacel aye r extended completely to the low-speed end. In th ecenter of the lead ing edge the coa t ing was ro l l ing inappearance Ind icat ing poss ib le l o s s of adhesion and there wereocc~sional brokeerrt sports. At the low-speed end of th espe: -L....e coating also had a ro l l ing appearance. Therewere several pits t h a t appeared to have been 'ormed bybubbles t h a t opened in the coa t ing : these pi t did not seemto be nuc l e i fo r more severe eros ion .

Inspect ion of specimen 2156B t ha t was t e s t ed fo r 40 minshowed t ha t the high-speed end was s t rong ly abraded and wasa l l iga to red with su tu res to a degree t ha t might be re ferredto as a ro t t ing of the surface l ayers by cracking. SeeFigure 18. It appeared tha t the sutures even tual ly circumscr ibedhunks of coa t ing which then broke away. In the cen t ra l par to f the leading edge the su tu res were broad and sha l lowand there were broken out spots , At the low-speed end ofthe specimen there were shallow broken out spots wi thevidence of small cracks or sutures, There were alsospl ier ical p i t s t ha t may have formed when bubbles openedin the coat inp ' but again these did not a.e.. to have growni n t o damage centers.

WDC IM 53-192 P t XII I - 37 . BESTAVAILABLE Copy

* K

8/13/2019 Ad 0225488

52/102

SVisual Inep.nlWfi of specimen 2157B that was tested fo r100" min showed 1wo areas in which the coatIng was torn awaydown to the laminates Microsoopic inspection j i owed severeloss of material between sutures a t the hignis 'eed end ofthe spezimen; deep cracks or sutures existed in \ the coating,See Figure 18, In the center of the.nading edge'there wereisolated areas containing sutures and Isolated brokenout spots existed; there were several deep brokenout spots. At the low-speed end of the specimen there wascomplete removal of the surface layer, general shalloweroslon, and evidence or formation of shallow autz es,

The erosion tha t was produced on Goodyear 23-56-Bostik1007 neoprene coated specimens a t a velooit7 of 500 =I/hris remarkably different from tha t produced a t a veloci tyof 600 mL/hr, At a velocity of 500 m/hr the erosionappears to consist of a.,general mechanical abrasion accompaniedby a crack formation t ha t could be due to a t tack of theneoprene by ozone or by hydroy l Ion. See Section 4.2. Ata velocity of 500 mi/hr the crack formation appears to bethe factor that contributes most towerd eventual drast icfa i lure of the coatings Blemishes such as bubble& thatmay have opened In the coating during cure do not appearto be nuclei for erosion attack, This is evidence thatholes in the surface of a neoprene coating do notl e .d to . serious fa i lure of the coating unless or unt i lthe coating as a whole over the area of ra in Impingementhas become rot ted or degenerate, An inta.%;'a network.of cracks appears to serve t h i s purpose and t o 0 ead todrastic failures.

At a velocity of 600 mi/hr the mode of fai ure Is th e

production of a chevron pat tern along the t raje tor ies ofwater flow from the Impinging drops. There is notablecementitious deposit especially on the ends of a spealmenthat was tes ted a t a veloci ty of 600 ml/hr whereas thisdeposit Is n ligible when the t e s t is 'carried out a t avelocity of 500 ii/hr Precipitat ion of calcium sa l t sappears to occur in the water of the intercepted drops whensolid specimens are rotated through the ar t i f ic ia l rainof tap water a t a velocity of 600 u!i/hr, The hadriessin parts per llion-of calcium c a r boa t s for- e vap waterused is 22. This Is a re la t ive ly low hardness.

OtTADC TR 53-192 Pt XlII,, 38 .- BESTAVAILABLECopy

8/13/2019 Ad 0225488

53/102

L'A: TM I

I'I11iNK 18 VlIiaa OF GLCUY3Ait AYTi~fz~O3~ANitrNO T'iTPHVft]Ai OY TluC AT A vx'1PCTy or 5oo mr/Hm U 4 v - m

8/13/2019 Ad 0225488

54/102

-the bollerswcale equation is i

Ca(H1C03 )2 V CaC03 + H2 \+ C02 I'

and the equi41brium reaction is forced to the 'Ight by

loss of the gaseous product, It is possible that theloss of carbon dioxide may be caused In, the open systemof a colliding waterdrop by the. Impact pressure o r bya temperature r i se in the water of the drop as a resultof the high-speed collision,

In an e*tQr~t to throw some light on the mechanismtha t produced precipitation, experlments of a preliminarynatare were performed in which concentrated solutions ofcalcium bicaroatj were i r radia ted with sound.It was found t ha t precipi ta t ion of calcium carbonate onlyoccurred If the solution was allowed to beoeme warm durinrthe i r radlat ion,

The fact that mineral sa l t s preolpi ta te in the water ofthe & rt if i c i a l ra in drops as they col l ide with a rapidlymovin solid Is Important in the problem of tes t ing th eresistance of materials to highpspeed rain-erosion damage,It is es ' -Ua l ly important in studies of the mechanism bywhich tniA type of damaae occurs because the erosion damageis part ly caused by the flow of the impingi. waterdrops.The erosive action of the flow of water t ha t Is carrYina precipi ta te 3s nore dest ruc t ive than tha t o f the f lowof water alone; It s a serious problem in dredging pumpsand-y u l lo ibghes . Eroslon due to flowing water thatis carryiM a precipi ta te resu l t s from the e"l l i s ion ofpart ic les suNpeid in the water a g i n s t the 4solid surfaceover which the flow occurs and by the ra of these par t ic lesover the solid gsurfage under the act ion of h74c fodycesa

Ilgaz / I / has studied the wearing of a plane surface bya sand-laden Jet of water, He found tha t the eqnation

u- i s '7 o.7 Vo

where U Is the rate of wear, Ic is a proportionality con-t-i.,C, is I.e -o*cmntra.lon of the transported materia&4Vo ie tX,

WADC 7.7% PZ.XZ11 - BEST V IL BLECOPy

IK

8/13/2019 Ad 0225488

55/102

velocity of the mixture, and S is the cross section of the Jetp,could be used to describe the resul t s . A picture of the wearproduced on rubber by impingement of such a Jet is similarto the wear pattern produced on a neoprene coating duringt es t on the ro ta t ing arm apparatus in an artificial tapwater ra in aS-xayelocity of 600 ml/^r. In the caseof- the Impinging sand-laden Jet the wear grooves areradia l l ines from the stagnat ion point, tha t is, the weargrooves are in the direction of the water fNow. In th et a se of the airfoi l-shaped rain-erosion spocimenscoatpa with neoprene, the wear grooves are h)evron-shaped,that jis he wear grooves are also In the direc t ion of th ewater flow down and away from the leadig edge of th eairfoil shape.

The effect of the flow of gr i t - laden water may bemore serious on one material than on another; it is, furthermore,not a genuine aspect of real ra in lmpinementp Consequently,it may cause divergence of t es t rej.ults from what is i',undunder seivice conditions, Every effort should be m.de toreove . th i s undesirable feature from the rotating-am ra in-erosi t i t e s t ,

A further discussion no' the effect of hardness in thewater.used for the artificial rain is given in Section .oll,

3.3.2,3 Gaco N-79 Neoprene over Qaoo U-15 Primer

The specimens coated with OGao N-79 neopren. over GaooM-15 primer were numbered 2121 A and B and 2122 A and B.A picture of the eroded specimens is hown in Figure 19.r0%cimen 2121 A was tested for 10 min at a velocity of

M mi/hr in 1-in/hr rain, ti-oescopio inspection of thisspecimen showed that there wat a scatterlnsgof small roundholes down the leading edge. There were alo quitea few much smaller holes that may possibly have been producedby the tearing loose of grains o f coating because thiscoating was very grainy In structure, Thaer was noevidence of a chevron pattern in th i s 0aco 0oating. Athigh magnifcat3 1 some evidence of what might be the

=ni ta t lon of sutures could be seen,

3pecimen 2121 B was tested for 20 mrin under the 'samcconditionr o f veloci ty and rain rate. Microscopic inspecti.onof this specimen again revealed a scattering of smallr holes down the leading edge. The-coating was,f~thermore,,Deppered-with the mai le r holes tha t were notedon specinm-m 21 A. There was an abrasion or tearing awayof surface la-ers especiolly around tae =03 holes ofboth sizes. !here was no evidence of a chevron pattetrn

WADC TR 53-192 P t XIII - -

BESTAVAILABLECOPY

8/13/2019 Ad 0225488

56/102

WAPbC U- i . - , 9 2 Tt X I 1 4

8/13/2019 Ad 0225488

57/102

in the neoprene* At high magnification there was evidenceof the i n i t i a t i on of sutures or cracks in the coating.

Specimen 2122 A was tes ted for 40 min underi--the saconditions of velocity and ra in rate . There was a general 'abrasion o t the surface of t h i s specimen and many of th every mu l l holes tha t t1-e notice" on speo iens 2121 A and B.There were quite a few du tue s or craftcs the coating.There was a tearing out of pieces of ooat fram th esurface a t the higho-spedd end. There was also a peelingor tearing back of the coating from thoe extreme edge ofthe high-speed endp A val id comparison be~een theGoodyear 23-56 and Gaco N-79 coatings cahnnt be made ont h i s point, however, because the Goodyear 23-56 coating liadbeen applied not only on the specimen but also on eshoulders where the cl ips fasten the specimen to th;propeller so tha; the edge of the coating was protectedfrom the coll iding waterdrops; in the case -of the ,,ecimen

coated with Gaoo N-79 the shoulders had been left bareof coating so tha t the water from the Impinging drops hadaccess to the - d g e o fhe coating. There was an undistioctchevorn-liki pat tern in the neoprene of th i s specimen;it exis ts a t the high-speed end of the specimen whereasthe chevron pat tern in the Goodyear 23- neoprene apperaiedin the low-spaed end' of the specimen.

Specimen 2122 B was tes ted for 60 mi under the sameconditions of velocity and raIll kata. Microscopic inspectionof t h i s specimen revealed increased abrasion wear Andtear ing 'out of pieces of coating mater.Al from thesurface. The coating was also peeled or torn back extensively

from the extreme edge of the high-speed end where the impingingdrops had access to the edge of the coating. There wasan indis t inct chevron-like pat tern in the neoprene oft h i s specimen mainly a t the high-speed end. There weremany sutur~.s. acka in the coating.

In general it can be said tha t the white neoprenefa i led preiaturely due to loss of adhesion; the in t r ins icresistance of the coating itself was easent ia l ly nottented. Although the adhesion failur. ; oa the Goodyear, 23-56wrs less dras t ic than tha t of the white neoprene, It showe'zdef ini te adhesion failume. At the end of 40 min of t e s t th e

WADC TR 53-192 P tXVI - -43

BEST V IL BLECOPY

8/13/2019 Ad 0225488

58/102

3aoo-N-79-Gaco-N-15 system showed more abrasion than theGoodyear.23-56-Bostik-1007 system but the Goodyear ooatinghad lost adhesion so that It failed drastically a t theend of the 60-main test period whereas the Gaco coatinghad not as yet failed drastically. Adhesion appeared to .be the weakness-leading-to-fallure of the Goodyearcoating; graininess appeared to be the weakness-leading-to-failure of the Gates coating. Both coatings developedsutures or cracks.

The email round holes tha; were observed In boththe Good . e re556 and the Qaco N-79 coatlngs are""ver1ylikely due to air bubbles. The bUbples may haverisen to the surface and burst during thp cure of thecoatings or the bubbles may have risen oose to thesurface during the cure and may have burst only when theywere struck by intercepted waterdrops.

The very small holes that were >nly characteristio

of the Oaco N-79 coating may be the result of the tcaringout of grains of this very grairnr ooating. The numberof these holes appeared to increase with the length of theteat time. The holes of both sizes appeared to providea foothold for abrasion failure of the Oaco-N-79-Gaoo-N-15system. Material was torn out from around the holes.

in general the failure of the three neoprene coatingsin rain erosion test was in agreement with the predictionsmade for them on the basis of their resistance toimpingement with oil-f i l led gelatin capsules and withdeforming lead pellet%.

WADC 'T, 53-1r. 12 Pt XIIX 411 BESTAVAILSLEup y

8/13/2019 Ad 0225488

59/102

4. Possible Modes of Failure of the Neoprene Topcoat

The most probable mechanisms that may be responsiblefor the t a l lu re of the neoprene topcoat ore the rubber abraelocthat has been referred to extensively throughout t h i s report,chemical deterioration and mechanica fatigue* Thesemechanisms are discussed In the following sections.

k . l Rubber Abrasion

The process of abrasion is not clearly understood.There appear to be as many abrasion mochanisms as thereare wajs of produoing abrasion. on surfaces, E3perlmentalwork that has been done on the moving of one metalsurface over another without lubr ica t ion indicates thatt h i s process is not continuous but Is oharacterized bya "s t ick-s l ip" behavlor. Bowden and Leben tejstatethat the ir experiments suggest that f r i c t ion Is due toa welding together of the metals a t local points of conraot.This is the st ick"-step. They regarded theme Iunvt.nftsas being large compared with the dimensions 'f a moleculeand concluded that when they are broken the metal isdis torted to a considerable depth. The breaking of he Junctionsis the "s l ip"-s tep and is aoocepanied by a temperature f lash .Morgai. Muskat, ard Reed E3 la te r concluded that meltingis now- necessary to establise the "s t i ck -s l ip" behavior andascribed the temperature f lashes on the "s l i"-s tep todissipa t ion of enorgya

The oaftlnfrubber Is also a "s t i ck -s l ip" process.Schallamach [is, n has found that the abrasion trace leftby a needle on a pure gum vulcanl2ate rubber is not continuous.It consis ts of i sola ted p i t s or tears; they surfaoe materialbetween the p i t s or tears is undamaged. In regard tothe abrasion caused by a needle, Sohallamach postu la tedtha t (a) the needle pricks a mal l hole In the rubber, and,as the needle moves a bulge of rubber is bu i l t Up in frontof it; eventually, (b) the needle moves over t h i s bulgesimultaneously pulling out a thong of rubber tha t breaksei ther a t I t s root or a t the needle t i e . Step (a) Is th e"stlckw-step and step (b) Is the "s l ip -step. See Figure20. The mechanism of needle abrasion p o t U '*ad bySchalljosaft depends on two factors, namel7, the bulgeof rubber mu t a d h e re by f r i c t ion to the t ip of the needleagainst the-IX- 'of the elas t ic forces, and the ten t i l -

properties of the material must admit of ae deforma;t'.-nwlhou t fa i lure .

WADX TR 53-192 Pt XIII -5 - BESLA.h . LOpY

8/13/2019 Ad 0225488

60/102

C)-4CI)01

' -.--

Ci)01

C)

Ci)C,001

0 U)i-s IU) C)

-- 5. H*'s5* H

U)

C)

(\I C)4-4

I)

02

r01C)HF-'

ricwl(LI

C,.,)(-iJ

I)I-.)

I-I

[f , -i9 Y) II

8/13/2019 Ad 0225488

61/102

Schallamach has found that an array of nearly parallelridges is often produced on rubber surfaces-by abrasion.He refers to th i s array as the "abrasion pattern". Theridges of the abrasion pattern run at r ight angles tothe direc t ion o brasion and are asymnetric with respect,to th i s d-c ton. They are characterized by overhangingcres ts tha t lean against the directSP~n o' abrasion.According to Schallamach the ridges are gll bent backwardaduring the abrasion and material is "removed by abrasion"from what in the relaxed s tate was the underside of th eridges. Schallamach .remarks tha t the orie.r. of th eabrasion pattern id not yet fu l ly underetood and s tatestha t It "is most" probably a consequence of the combinationof high e las t ic i ty and high coefficient of f r i c t ion whichIs character is t ic of rubber".

In view of the observation tha t the ci rcles of damageprdueied on a lO-mil thickness of MM EC-539 neoprenei bythe impingement of waterdrops may be the resu l t of anabrasion proce5ss -some elementary observations of the

-abrasion orbhis rubbery ooatIzZ were made. An abrasionsimilar to tha t observed by Schallamach w's found to -resu l t when a sharp-pointed tungsten carbide markingneedle was drawn across the surface of the\coating.See nictures 1 and 2 of Figure 8. The damage isdiacontinuous and consista of a succession of p i t s ortears. The flap of rubber torn out of the surface in th emaking of each of these t ea rs re".:"crs in such a way thatit points in the direction from which the marking pencilwas moving. In pictures 1 and 2 of Figure 8 the markingpencil was moving downward and it can be seen that the flapsof rubber ripped up In making the individual tears pointupwards.

The abrasion of th i s rubbery coating tha t resul ts whena razor edge is drawn across it was also observed*I r regular i t i es along the razor edge produce a resu l t thatis equivalent to tha t which would be produced by a seriesof needle points moving along a l ine. See picture 3 ofFigure 8. From time to time the cut in the rubber Is

oeontinuous over the space of several oe the i r regularprotrusions on the razor edge. The &iVs w.' rubberthat tre raised from the surface again point in thedirection from which the razor edge moved, that is, towardthe le f t of the picture.

W,.w, TR 53-192 PL XIII - 47- BBET\l

8/13/2019 Ad 0225488

62/102

8/13/2019 Ad 0225488

63/102

INV*,

;t v

___________ ____ ______ _ _ _ _ _ _ _ _ _ _ _ _ _

8/13/2019 Ad 0225488

64/102

NiNMcroscopic inspection showed the presence of isolated,surprlsing round, small holes especially o the lowspeed end of specimen 1893 B. Deep circul r holes thatwere surrounded by a torn-out area wero ob rvedoThis may be evidence that the coating was torn awayaround holes of th i s kind by the radia! flow of drops

af ter the holes themselves had formed. In the case ofa t leas t one such round hole on specimen 1893 B, a th inlayer of coating material t ha t was about the size of th emouth of the hole appeared to be lying on the bottomsurface of the hole. This observation ma.y Indicate thatthese holes are formed when the th in skin .of coatingover bubbles, which exis t in the coating I t se l f andwhich are near the surface, is broken In.

The process of erosion appeared to be the tear ingaway of th in segnents of coating material a t the surfaceof the coating. After a roughened surface Is formed,further erosion eauld be expected to progress rapidlybecause the impact pressure of addit ional dropz UxatImpinge is multipl ied in depressu.ons of the surfaceroughness and the velocity of the radia l flow, whichgoverns I t s ab i l i ty to tear more of the coatin away,Is increased. Cuts or tears which could be due to at tackof the neoprene by ozone or by hydroxyl ion were alsoevIde', in the roughened surface of the coating onspe-amen 1893 B. Quali tat ively similar dae featureswere observed on specimen 1893 A. Tlnerodec areas onthe sides and away from the leading edge of\ both speciLnen1893 A and 1893 B were grainy In appearance.

Specimens189

A and B were eroded at velocity of600 ml/hr In 1-in./ ,r artificial rain for 1 0 min. Thear t i f i c i a l ra in was again of ordiuary hydrant water having areported hardness of 22 ppm. The only difference i0 th etest conditions of specimens 1894 A and B and specftmjns 1893A and B was the velocity at which the specimen struck thewaterdzops of the artificial rain and the test time.Visual inspection of these specimens showed that althoughthey were tested at a higher velocity-and for a longerperiod of time U m -specimens 1893 A and P hey wereeroed t a esser degree. Microscop i cx~m.m._%tnindleated that the erosion was also of a dIfferent charanter.

The low-speed end of the leading edg was charawii..-tze.by a chevron pat~ern of erosion in the neoprene coati!n 3e-BSIdtture 1 of Figure 22. The point of the chevron faced th e

2IBEST AVAILAB[F~ WADW TR 51-192 Pt XIIT 50-

8/13/2019 Ad 0225488

65/102

0.3

-I4

8/13/2019 Ad 0225488

66/102

low, speed and of the specimen. The eroded surface wasrippled but was suw isingly inooth and almo t free ot theerosion characteris~ios that'vere found an s9teimrs893A and D a~ ae few xa holes and to ~uara"could b e ee At the high-speed end of the leading edgeor the specimten the chevron pattern was almost obliterated

bya generally beats texture of the eroded sufaceseo icture 2 at Fi8gure 22. Cracks that c = 4 be due toattack of the neoprene by ozone or by Wqro,.~ i)n could

be seen, A camentitious mineral deposit existed at b~othends ot the specimene The appearance of sppi34 l89S was quaZitatively similar.

Speo3mens, 1895 A and B were tested under the sameconditions and for the same length ot-time as were specinens1894 A and D but the water used for *the artificial rainwas passed-throughy-a -softener and was reported by theCornell Aeronautical Laboratory to havea, h.ardness of ler-mthan 2 pM.9 These specimens eroded in a "n r slimilar *tothat of specimens 1894 A and B, Comentitiobus depositswere observed at both ends of these specimens showingthat the softened water used for the artificial raincontain"e enaugh hardness to allow precipitation to occur,The chevron pattern at the low-speed end of the leading edgewas less 4istinct than on specimens 1894A and R but thebeady Uextiare of the surface at the high speed end wasabout the same. The Cornell Aeronautical Laboratoryreported that In the case of specimen : 394 A and D the

chvron pattern was first detected after 25 aki of test butthat In the case of specimens 1895A and B it appeared onlyarter 40 mmi of test, This observation appears toipdicate that-the-chevro pattern Is related to the amount

of-Me~dness In the water and hence to the amount atcementitious precipitate that can comie out \ot it. Theobservation that the general beady texture ~t the high-speedend ot the specimen after the-surn test timx (140 min)was about as distinct as on specimens 1894A and B seemsto Indicate that it may be a result of the water flowand/or of the texture of the neoprene coating now thanof the amount at precipitate present in the water*There were quite a few sutures or cracks In % b coatingan both apecimens.

BEST V IL BLECopyWADCT53-19t2Pt XIII -52 -

8/13/2019 Ad 0225488

67/102

Specimens 1896 A and B were tested under the sameconditions as specimens 1895 A and B except that the tota l testtime was 130 mnn and tha t 1.0 g ot My=ine 1622 germicldal1wetting agent was added to each 0.5 gal of water that Was_--used to make the a r t i f i c i a l rain. The addit ion of th ewe6tin sfiToreduced the surface tension of the softenedwater frce 75 d/ci to 35 d/cm. The Corne~j AeronauticalLaboratory reported tha t the Ooodyear 2 3 - b neoprene coatingappeared to be sof ter on these two specimens and that achevron pat tern similar to that which appeared on specimens1895A and B was detected a f te r 40 min o f tts..

microscopic Inspect ion of these specimens showed th epresence o f a eementitious deposit a t both ends. Theerosion was similar In appearance to that which formed onspecimens 1895 A and B, namely, a din chevron pat tern atthe low-speed end and a beedy structure of the surfacea t the high-speed end. There were a few round holes ',d

torn-out spots. On specimen 1 8 6 S a section of coatingwas turned back so tha t the reverse side of It (the sidethat had been bearing against the alumumn alloy pecimen-base) could be seen. The revezrse ide of the coating hada beady s tructure without having suffered erosion a t a l l .See picture 2 at Figure 21. This observat ion appears toindlecate t ha t the beady s tructure of the surfade tha t formeda t Wle hugh-speed end o f the specimens durin t o r t Isre la ted to the coating I t s e l f as wel as to the waterflow o r to the presence of a p r e . 4 . : e t e In the water*The presence of globules o r beads In the coating o t 1 dr e su l t from use of neoprere that was not e ff ic ien t lydispersed during compm ng or It could be due to th e

use of old material in w h c h par t ia l coagulation hasoccurred. Addition of di luent to such maser 'al wouldnot disperse the aggregates. nspe t ion 'a tornaedge o f the coating on spt -men 1696 B showed some evidenceof separation of layers of the coating. Both speailmn 1896A and spe'elmen 1896 B contained maniy nuturqs or cracksthat could be due to a t ta .k by ozone or byFbydrol ion.

4.1.2 Effect of Detergent Applied to the Coating

Four a i r fo i l specamens were tab at f -2 \alf1oa t thn Cornell Aeronautical Laboratory. They vore GO&% eW thBostik 1001T primer a&n4 with Goodyear 2356 neoprene (brushcoat). After a i r drying for one week,, the specimens i-re

WADC TIM53-192 Pt XIII 53 - BEST VA1A ,A.

8/13/2019 Ad 0225488

68/102

dried in a dessicator overnight and were then weighed ona chemical balance. Two of the specimens (CornellAeronautical Laboratory numbers 1400 A and 1401 A) werebrished with a concentrated solution of alkyl arylsu l forate and were air dr ied fo r a t l e a s t one hour. *Each of these specinens was t e s t ed with an, unt rea tedcontrol specimen (Cornel l Aeronaut ica l Laboratory

numbers 1400 B and 1401 B) a t a ve l oc i t y of 600 mi/hr in1 - i n , / h r artificial r a in * After each 10-min I t e rva l of th efirst two hours o f the test, the four specimen were d r i edin a d ess i ca to r overnight and were then wei.cad\ beforeano the r f i lm o f detergent so l u t i on was appl ied to th et reated specimens. This procedure was repea ted a t 20-mini n t e rva l s during the remainder of the test. weightl o s se s o f these specimens a t the end

of the l0-"win

i n t e rva l s o f test during the first two hours o f t e s tare p lo t t ed In Figure 23.

Inspect ion of the eroded speeimens at X 20 m a g : i a t i o nr evea led t h a t specimens 1400 A ( t r ea t ed with dete rgent and1400 B (control), which were tested for ek total of 260miin, were in about the same state of erosion, There wasa fairly dist n ~heVrOn pattern at the low-speed endof these specimens and small .oating bubbles and torn-outspots at the high-speed end. Specimen 1401 _4( t rea t4 witn detergen t) appeared to be s l i g h t l y moreeroded than specimen 1401 B (cont ro l ) . These 'specimenswere tested for a total of 320 min. T wes a strongchevron p a t t e rn a t the low-speed end, a genera l beadinessin the cen t ra l pa r t of the l ead ing edge, coat ing bubblesover most o f the leading edge, and t o r n - ou t spots at th eh igh-speed end o f these specimens. All four specimens

conta ined qui te a few cracks or su tu res t h a t may be due toa t t ack o f the neoprene by ozone o r by hydroxyl ion.

The microscopic inspect ion showed t