1205

REPORT ON THE MEASUREIViENT OF TAUTNESS OF DOPED FABRIC.

B y F . W . A s T o ~ , D .Se . , of t h e R . A . E .

P r e s e n t e d b y CONTROLLER, T e c h n i c a l D e o a r t m e n t , A i r c r a f t P r o d u c t i o n .

Report~ and Memoranda, No. 569. December, 1918.

SIIM~IARY.--(a) Introductory.---Early in the year 1918 the Dope Panel of the Engineering Standards Committee expressed the opinion that a s tandard method of measuring and expressing tautness of doped surfaces was urgently needed. The representatives of the R.A.E. were asked to undertake experiments on the matter . The following report deals with the results of these investigations and with general considerations of the question.

(b) Range of investigation.--Direct methods not being feasible, the numerous indirect ones received careful consideration. The R.A.E. depression method, the G.A.C. tensionmeter and the proposed s tandard spherometer methods are discussed and their results compared. The great importance of hygroscopic conditions in tautness testing is indicated by a diagram showing change of tautness with varying humidity.

(c) Conclusion. The conclusion arrived a t is t ha t the best method of measuring tautness is to establish a fixed difference of pressure between the air on the two sides of the doped surface by using an air t ight frame and then to measure the radius of curvature of the very nearly spherical surface so produced. A spherometer specially designed for the purpose is described in detail.

(d) Application and further developments.~This apparatus, which works very satisfactorily, is now being installed at the R.A.E. in place of the former " depression " method. I t is hoped tha t its use will give valuable information.

Introduction.--Tautness of d o p e d f ab r i c is i m p o r t a n t in a g r e a t v a r i e t y of w a y s . A e r o d y n a m i c a l l y i t is e s s e n t i a l t o r e d u c e t h e s ag to c e r t a i n l imi t s , b u t i t is p r o b a b l e w i t h t h e m a t e r i a l s n o w in use t h a t a n y s u r f a c e s d e t e r i o r a t i n g to a s l a c k n e s s suffi- c i en t t o a f fec t t h e a e r o d y n a m i c a l p r o p e r t i e s of t h e m a c h i n e s e r i o u s l y w o u l d h a v e b e e n c o n d e m n e d long be fo re on t h e i r a p p e a r a n c e , t h o u g h w i t h su f f i c i en t ly e l a s t i c a n d e x t e n s i b l e m a t e r i a l t h i s m i g h t n o t b e t h e case.

A p p e a r a n c e is a c t u a l l y a v e r y i m p o r t a n t c o n s i d e r a t i o n i n d e e d . One c a n n o t e x p e c t a p i l o t t o v i e w w i t h f a v o u r a

1206

machine with its fabric hanging in reefs, however good it is in other respects ; also slack fabric when in flight, and particularly when exposed to the slipstream of the propeller, may get into a state of flutter, the violence of which may tend to increase and lead to its complete collapse.

Another point which has been somewhat neglected is how very greatly the strength of the wing structure is influenced by the tension of the doped fabric upon it. In general, slackness of the dope will weaken the strength of the wing structure, but on the other hand, too great tautness will lead to deformation or even fracture. There seems little doubt that were we able to guarantee a definite tension constant in all weathers and con- ditions, sweeping improvements could be made in wing design.

Method of measurement.--Direct measurement of the tension of the doped fabric can only be car~ed out with elaborate apparatus and is quite out of the question as a practical method. One must, therefore, rely upon indirect methods. Of these a great variety have been suggested, nearly all depending upon the application of a force normal to the surface under considera- tion and the measurement of the relation between this force and the resulting displacement.

Thus, since 1916 comparative measurements of tautness have been obtained in this laboratory by placing a small sphere at the centre of a standard frame, loading this with a known weight (100, 200 or 400 gms.) and measuring the consequent depression with a cathetometer. This " depression " method is quick and delicate and works well for comparison of dopes, but., as the mathematical relation between the stresses in the material and the depression measured is extremely complex, it is unsuitable as a standard method. I t is also obvious that it can only be used on test frames of exact dimensions.

The G.A.C. acre tensionmeter.--In this instrument the last objection is overcome by the instrument in a sense supplying its own frame in the form of its heavy circular brass rim, which rests on the doped or undoped surface. The depression of a specially designed disc at the centre of this circle is applied by a standard spring and measured by an Ames dial gauge. This instrument is simple and portable, it also possesses the great advantage that it can be used upon aeroplanes direct.

In order to obtain some idea of t h e accuracy with which readings could be repeated by different observers, a G.A.C. tensionmeter was tried on various machines in service at the R.A.E. This was done first on May 7th, an exceedingly wet day, and again on May 9th, which was bright and very dry.

The measurements were done by two independent observers, the time interval not being long enough to introduce change of tautness owing to variation in humidity. In the following table of results the figalres obtained by tbe second observer are always

1207

given below those of the first, the three readings for each wing being those of the tip, the middle, and the body end reading from left to right : - -

G.A.C. TENSIOlqMETER. TABLE OF TESTS DONE ON 7.5.18 AND 9.5.18.

Sopwith Dolphin C.4191. Wings new. Right Wing. Right Wing. Left Wing. Left Wing.

Wet. Dry. Wet. Dry. 82 67 72 63 64 69 69 71 68 64 61 fi8 75 76 81 66 62 62 70 75 71 60 63 fi5

R.E.8 A.3406. Wings new 24.8.17. Flown very little. Right Wing. Right Wing. Left Wing. Left Wing.

Wet. Dry. Wet. Dry. - - 65 83 52 44 55 - - 76 81 54 60 56 - - 7 4 81 4 9 4 1 4 9 - - 7 3 7 9 5 3 5 4 5 9

D.H.6 B.2963. Wings new 13.4.18. 5.7 hours, flying. Right Wing. Right Wing. Left Wing. Left Wing.

Wet. Dry. Wet. Dry. 47 44 45 65 70 77 47 46 45 64 71 71

D.H.9 C.6060. Wings new 24.1.18. 18.5 hours flying. Right Wing. Right Wing. Left Wing. Left Wing.

Wet. Dry. Wet. Dry. 82 94 89 80 80 - - 98 96 97 85 82 90 83 88 83 76 76 77 96 93 99 86 83 85

N.E.I No. 3. Wings new 28.11.17. 24 hours .flying. Right Wing. Right Wing. Left Wing, Left Wing.

Wet. Dry. Wet. Dry. 77 86 80 55 55 45 93 83 92 54 55 62 79 83 77 49 52 45 83 83 89 53 53 60

R.E.9 3911. Old Machine. Right Wing. Right Wing. Left Wing. Left Wing.

Wet. Dry. Wet. Dry. - - 96 100 56 60 65 96 94 92 54 54 56 - - 9 7 1 0 0 6 1 6 4 6 4 9 7 9 6 9 5 5 4 5 3 5 6

The mean numerical difference in readings on 62 observations is 3.0, the mean percentage error, calculated from the higher figure 4.1 per cent. : the consistence in the case of corrugated surfaces was better than one might have expected.

The table shows well the progressive increase of the difference between the wet and dry tautness with age. The case of D.H.6 B2963 is of interest as ~:ndieating the effect of a short exposure to sunshine in the case of the right wing while the left remained in the shade. R.E.9 and D.H.9 in the wet gave alrhost identical readings, hut their appearance was absolutely different, the former being exceedingly corrugated, and the latter apparently quite taut , a good illustration of the importance of elasticity as regards appearance.

1208

There seems little doubt that the irregular readings (v, Fig. 2) obtained with the instrument are due to two causes--the friction in the Ames dial gauge, which can be eliminated sufficiently for practical purposes by the use of some such device as that described m the spherometer below, and the friction between the brass rim and fabric surface upon which the isolation of the circle depends.

In working out a formula for the instrument, Dr. Silberstein has assumed the latter to be infinite (to give consistent results it is enough for it to be constant), while in practice approximately reliable results can only be obtained by judicious tapping which makes it virtually zero.

In order to get a rough idea of the real value of the coefficient of sticking friction, the instrument was placed on various surfaces and pulled in a horizontal direction by a spring balance. From the readings of the balance when slipping just stal~ed and the weight of the instrument (1,200 gins.) the following values were obtained : -

Surface. Coefficient of Fr ic t ion . U n d o p e d i a b r i c . . . . . . . . . . . . 0 " 3 7 5 R a f t i t e u n w e a t h e r e d . . . . . . . . . . . . 0 " 4 0 0 R a f t i t e w e a t h e r e d 0 " 3 7 5 • . . . . . * , * . ° .

Dope weathered and badly cracked . . . . . . 0"568 Pigmented dope unweathered . . . . . . 0"465 P.C. 10 unweathered . . . . . . . . . . . . 0"393 Emaillite pigmented oil varnish weathered... 0"400 Emaillite pigmented oil varnish unweathered 0"67-0"85*

As the rim has a circumference of 47 era., the difference of tension between the inside and outside necessary to cause slipping is approximately 10 gins. per cm. The extension caused by the depression of the plunger at the centre for ordinary doped sur- faces requires u force of about the same order, so that it is doubtful if an instrument working on this principle is suited for the accuracy rcquired in the laboratory, but as regards its use on planes direct, its portability and simplicity are strong points, which, together with the fact that it can be used on undoped fabric, should render it an extremely useful instrument for technical inspection work.

Standard method of measuring tautness.--There seems little doubt tha t the best method of applying a measurable force to doped fabric is in the form of a fluid pressure. If air is taken as the fluid and the pressure chosen properly, the surface is approximately under actual flying conditions. The measure- ment may be made independent of the size or even the shape of the frame, and when obtained can be expressed in such a way as to give direct mechanical information to the designer.

* D e p e n d i n g on how long the i n s t r u m e n t was a l lowed to r e s t o n the s u r f a a s before the pull was applied.

1209

In consequence of these suggestions submitted to the meeting held July 26th, 1918, the Dope Panel of the Engineering Standards Committee adopted provisionally the following definition : --

"Tha~ tautness ' T ' of a doped surface shall be defined as the evaluation of the expression

RP T ~ . . . .

2

where ' R ' is the radius of curvature when a difference of pressure ' P ' is maintained between the sides of the doped surface, tha t the difference of pressure should be as nearly as possible 1 inch of water (2490 dynes-cme), and tha t the value of the tautness should be expressed in lbs. per inch or kilogrammes per metre."

The tautness expressed in this way though not strictly the same as the tension of the surface is sufficiently so for all practical purposes. Hence a designer being given the tautness under various weather conditions can calculate the stresses he may expect to be caused in the structure of the wing by the doped fabric covering it.

I t will be seen that the definition above makes no restriction as to the method of applying the difference of pressure or measur- ing the curvature.

The method recommended in this report is that airtight frames should be used, the pressure inside reduced to 1 inch water below atmospheric by moans of a filter pump and ~[ariotte bottle, and the radius of curvature determined by a spheromoter.

Standard frames.--The form of frames adopted at the R.A.E. for all weathering experiments requiring determination of taut- ness is square, 12 inches × 12 inches external dimensions, built up as indicated in Fig. 1, the wooden sides 1 inch thick and 1.5 inches deep being reinforced as shown by a strip of iron 1 inch wide by ~ inch thick bent into a 10-inch square and welded up.

The frame has a permanent back of doped fabric, and the fabric for the experiment is stretched over the front of the frame at constant tension (2 lbs. per inch warp, 1 lb. weft) and doped down to the wooden face of the frame. When dry, the edges are pulled over and stuck down to the outside edges of the frame with V.114. This method, which does away with nails alto- gether (except one at each corner to secure the loose flaps of fabric), has proved the best yet devised to prevent slipping.

For use with the sphorometer tautness meter the breathing holes normally pierced through the back are replaced by a single one ~ inch wide through one of the sides. The outer part of the hole is enlarged to a width of ½- inch and lined with a short piece of brass tube driven flush. By means of the small tube and rubber stopper shown, airtight connections with the interior carl be made easily and rapidly. At the same time the exterior

1210

of the frame is left free of any excrescences or alterations in dimensions. There is no difficulty in obtaining airtightness sufficient for tautness determinations, small leaks being negligible.

Method of obtaining constant difference of pressure:--Several devices were tried in the hopes of finding something more satis- factory than the M ariotte bottle, but without success ; the form of the latter giving the most constant results is shown in Fig. 1.

The air overflow consists of a ring of sx inch hole in a 1-inch metal tube. It seems practically impossible to arrange for the device to be entirely independent of the rate at which the air overflows, so that the rate of exhaustion should be kept as uniform as possible.

The pressure is measured on a separate manometer and is exactly adjusted by sliding the air overflow tube through the cork. Small momentary variations of pressure causing unsteadi- ness of the manometer may be damped out completely by the introduction of a large capacity between the Mariotte bottle and the manometer- -an empty dope frame does very well.

The spherometer.--This instrument, which was designed by the R.A.E. Instrument Department, is illustrated full scale in Fig. 2. By means of a specially arranged movement modified from that of a Clift air speed indicator, the depression of the contact-pin 10 below the level of the edge of the circular base 16 is measured on a horizontal dial graduated in mm. and divided in tenths, which division, being large, may be easily subdivided by eye if necessary.

I t is as well to state here that a horizontal dial is very much more convenient to read than a vertical one as fitted at present on the G.A.C. tensionmeter, which might be altered with advan- tage. This movement is practically frictionless, and the spring necessary to take out back-lash is so weak that its effect in depressing the contact pin beyond the point at which it just rests on the doped surface is negligible in cases of ordinary tautness ; neither is the weight of the instrument, 205 grins., sufficient to cause any serious disturbance, so that it acts simply as a spherometer. The zero is obtained by placing it on a truly plane surface, and can be adjusted by rotating the scale.

Owing to the desirability of keeping the scale uniform, the depression h is read in mm., its reciprocal being a measure of the tautness for

T I~P PD~ - - 2 -- k]_6_~ (for small curvatures).

D being the diameter of the spherometer rim 152 ram. Hence the tautness T is obtained b y dividing a constant (determined for the instrument) by the reading on the dial expressed in mm.

With this particular instrument, the constant is 34..3 for T in kg.-metre and 1.92 for T in lbs.-ins.

1211

As soon as the interior of the test frame is at the correct pressure it is only necessary to place the spherometer on the frame to obtain the reading. Naturally, a central position is desirable, but no particular care need be taken in this respect,, as so far all doped fabrics tested were found to approximate remarkably closely to a spherical surface when submitted to the standard pressure. I t is for this reason tha t the instrument gives quite reliable results even with a slightly warped frame, although it does not necessarily read zero when placed on such

frame before the pressure is reduced.

Comparative results.--Fig. 3 shows results obtained on frames of various tautness by the depression method and the O.A.C. tensionmeter plotted against the reciprocal of the tension as determined by the spherometer method. I t will be seen that the curve given by the 100-gram weight is not so nearly a straight line as that given by the 400-gr., but the latter passes further from tile origin. The C.A.C. tensionmeter gives an irregular series of points, the locus of which is definitely straighter than the other two curves, from which it would appear that this instrument if calibrated to read the reciprocal of tautness direct would have a uniform scale--a great advantage.

Corrugations or reefs.--These appear whenever the tension in any direction becomes less than 0. They wel,e a source of serious error in the "depression " method as the centre of the frame might be a crest or a hollow of a corrugation, which made a considerable difference even if a 400-gr. weight was used.

The spherometer method apparently does away with this difficulty entirely, and as many of the most interesting com- parisons of dopes are made under conditions of humidity which cause corrugations to appear, this is one of the most important advantages of the air pressure method.

Limitations of the method. The Sphingometer.--It will be seen that, although independent of the size of the frame, this is essentially a laboratory method, as the latter must be reasonably airtight. I t is, therefore, not directly applicable to the measure- ment of tautness of undoped fabric or to actual parts of aero- planes. For the first only the roughest measurements are necessary, as experiments have shown that variations of the tautness of fabric before doping over a very wide range have practically no effect on the final tautness of the doped material, so that an instrument of the G.A.C. type will probably serve all requirements.

In order to apply the air pressure method to any doped surface without regard to the structure upon which it is stretched, Dr. Burr, of the N.P.L., has devised an instrument which he calls the sphingometer. The accuracy of the results of this, unfortunately, depend on the frietionaI isolation of a circle of

1212

doped fabric by the rim of the instrument. Dr. Barr claims that this isolation is good, but a calculation from the values of fric- tional resistance given on page 1208 show that it is still of the same order as the forces tending to cause slipping; it is, therefore, quite impossible to guarantee either complete sticking or complete slipping.

I t may be possible in the future in instruments of this type to render isolation certain or, by replacing the solid rim by a ring of rollers, to eliminate it altogether, but until this is done they cannot be recommended for accurate work.

Rigidity effect.--It was suggested by a member of the A.I.D. that a considerable part of the resistance to depression of a doped surface might be due to rigidity as distinct from tension, i.e., the surface might act as a piece of cardboard loosely resting on a frame would do. Although it seemed very unlikely that this force would be serious, a direct measurement was made by testing strips of doped fabric, 10 X 3 cms., clamping one end in a vice and measuring the bending--mean of results in opposite directions--caused by a small weight placed at the end.

In this way the pressure found necessary t6 cause the curva- ture obtained in ordinary measurements of tautness by the spherometer was, roughly, 0.02 cm. water, i.e., less than 1 per cent. of the total, so that the rigidity effect is quite negligible.

Accuracy of tautness measurements.--Accuracy is o~fly desirable or, indeed, possible in comparing dopes directly in the laboratory, and here the utmost hoped for is about 1 per cent. The reason for this is easily seen in the extreme sensitiveness of dopes in use at present to changes of moisture. Fig. 4 shows the change in tautness in fabric coated with six coats of standard raftite under varying humidities. ~easurements of tautness ander ordinary laboratory conditions are not of much value : the relative merits of dopes are only to be gauged by taking their tautness at various humidities, and particularly near saturation po in t - - it is for this reason that the dial in the spherometer was given the very long range of 4 mm.

Anyone experimenting in this field will soon realiso that to maintain a high and invariable relative humidity and to be certain that the doped surface is in eqlfilibrium with it, is a much more difficult problem than the actual measurement of tautness.

Now that deterioration due to exposure has virtually been abolished, the elimination or at least minimisation of change of tension due to moisture is by far the most important problem hi the technology of dope, and it is to be hoped that the standardisation of tautness and apparatus for the measurement will be a considerable step towards its final solution.

[~.~'PO~T N °. ,9¥~;'9

SEGTION OF FRANCES

F'I6. 0

P'IAR R~ OTTE BOTTLE

To ~c~nomePer ¢~d Fram~

#.:Air M

a~

u

0 - - ©

___ii _~ ±+:~'~-~:-~-jpo~ ~o~ +-~ ....... : . . . . . . . . -0 . . . . . -- 0 ....

- o --II--- - 0 . . . . . .

-::-_2~- - +

l'

REPORT N ° ,D-~9__,

A

/B

...... FIG; ~_ R.A.E.

TAUTNESS METER

B A

O N ( .~

I DIAL 2 ~PRING PLATE 3 PINION 4 QUADRANT 5 POINTER 6 HAl R SPRING 7 MAIN ,SPINDLE 8 I~ATiO ARM 'A"

. . . . "B"

S ECT I [ ~ N B.B.

I0 CONTACT PILN II SPLIT RING I~ B E Z E L k3 DIAL FIXING PiN 14 B E Z E L LOCKING PiN I~ DIBTAI~CE RiNG 16 SPUN ALUMINIUM BABE I'I iSELLULOID WINDOW.

REPOR_._T N......_~-J',~_.. ~, F~G N ° ,,~

READING~ OF COMPARATIVI-~

"T~UTN E55 METERS,

,D

o

o

e o o m

oo~

De

o •

f ~J

j ~

J

m m

o o

e oo Q

o o

o o e

o

o

J /

/

/ /

/ /

/

• 4- " 6

..~ PHE RO M E'T'E R

-8 ~.0 0.°.- 1-4

D E P R E $ $ ~ 0 N I N M . M .

VARIATION OF TAUTHE55 WITH HUHIblTY RAFTITE. Six COATS

"~IG.4.

"~o H~Jrnic :ity_ .40 ~0 ~0 ' 7~ 80 ~ I00

r-- -T- '~ ......... ~ ," /,o ]

- d ~ SO

i tO 1 --__ -4 ! o

4 0 5 0 GO ~ 0 8 0 9 0 t