7-JZ_~~/~~ ~~ ~"--- ~~-=-__e~ ~/2-6/G s /¥ .,flSC- ~/-#'S#/275

Navy Department Bureau of Ordnance Contract NOrd 9612

PITCHING MOMENT BALANCE FOR THE

HIGH SPEED WATER TUNNEL

By

R. W. Kermeen

Hydrodynamics Laboratory California Institute of Technology

Pasadena, California

Memorandum Report No. EM-12. 4 April 15, 1952

Copy No. IZ

J. T. McGraw Project Supervisor

ABSTRACT

The measurement of model drag in the High Speed Water Tunnel

has been difficult because of the interaction of pitching moment and the

measured drag forces. Due to the mechanical arrangement of the force

table used. pitching moment will produce reaction at the drag gage caus­

ing an erroneous reading. An internal balance has been fabricated which

measures pitching moment. allowing a correction to be applied to the

drag gage reading.

Introduction

Discrepancies in drag measurement have been observed in the High

Speed Water Tunnel which have been traced to the interaction of pitching

moment with the drag measuring apparatus. Tunnel models are rigidly

mounted to a vertical spindle which is pivoted in the center and restrained

by a load cell attached to the lower end of the spindle, as shown in Fig. l•.

Because of the mechanical arrangement, it is possible for pitching moment

to produce a force at the drag cell. Pitching moment can be produced by

shield interference, mechanical deflection, etc. A marked change in the

character of the drag curve has been produced by changing the lengthwise

position of the support point of the model.

Some pitching moment seemed to be unavoidable so an internal

strain gage balance was fabricated to measure the pitching moment pres­

ent. A correction could then be applied to the drag reading.

De scrietion of Balance

The pitching moment balance which is used in conju:tlction with the

tunnel three-component balance consists of a l-in. diameter cylindrical

section l-3/8 in, long. It is attached to the top end of the spindle and forms

a part of the model cylindrical body section. As shown in the sketch, Fig. 1,

the pitching moTnent balance, hence the model. is attached to the spindle

thr~ugh ball bearing pivots, A triangular, unif-o~~' strength, cantilever

beam extends outward from the spindle. Any pitching moment on the model

is transmitted through a knife edge to the end of this beam. Drag forces

are transmitted through the ball bearing pivots to the spindle. A pair of

DRAG CYLINDER

TO DRAG GAGE

KNIFE EDGE

UNIFORM STRENGTH BEAM

PRELOAD SPRING

STRAIN GAGES

FLEXIBLE DIAPHRAGM

-2-

PITCHING MOMENT BALANCE

DRAG PRELOAD SPRING

STRAIN INDICATOR PRESSURE COMPENSATING BELLOWS

BALL BEARING PIVOT

Fig. l - Schematic of tunnel balance and pitching moment balance

-3-

Baldwin SR-4 resistance strain gages is cemented to the beam and the

wire leads are taken out through the spindle. The entire assembly is

made water tight and filled with oil. A thin neoprene diaphragm seal is

used around the spindle and a pressure compensating bellows is installed

in one end of the section. The body section and beam are shown in Fig. 2.

l .

11'1'1

Type SR 4 strain gages on uniform strength beam

Ball bearing pivot

Fig. 2 - Pitching moment balance section before assembly

The Baldwin SR-4 Type L strain indicator is used to measure the

bending of the beam. This instrument was calibrated against the tunnel

balance drag gage and the strain indicator readings are applied as a cor­

rection to the drag gage data. The limits of operation of the pitching

moment balance are ± 17 in-lbs of moment which correspond to 1. 75

lbs of drag on the three-component balance. To insure proper alignment

of the model in the working section and to have maximum range with the

pitching moment balance, the models were statically balanced about the

support point before installing them in the tunnel. The range of operation

of the pitching moment balance can be increased by preloading the model.

- - - - ·~

-4- ~ - -

Results

· The balance was first used with a long model which had produced

erratic results. Figure 3 is a plot of drag coefficient versus Reynolds

number. The uncorrected drag data is represented by the lower curve,

while the upper curve represents data corrected for pitching moment

interaction. In general, the pitching moment correction was in the same

direction as shown in this figure, that is, a negative downward pitching

moment which causes a smaller drag gage reading. The negative pitching

moment is apparently caused by the balance spindle deflecting downstream

causing a small upward pitch of the model. This results in a negative pitch­

ing moment on the balance as the stable model used in these runs attempts

to align itself with the stream. It has been found that the opposite is true

and a positive pitching moment results when an unstable shape such as a

pointed large calibre ogive nose is used on the pitching moment balance.

As a test of the effectiveness of the pitching moment balance a

series of runs was conducted in which the 5-in. A. S. Projectile was sup­

ported at various points along the cylindrical body section. Figure 4

shows the results obtained on the three -component balance without the use

of the pitching moment balance. In these runs the model was supported at

the 0. 30 L, 0. 35 L, and 0. 48 L points. The curves (Fig. 4) were obtained

by making one drag run with a single spindle shield and then repeating the

run with the addition of a second image shield supported above the model.

This method is based on the assumption that the interference caused by

the image shield is the same as that caused by the spindle shield, and the

diffellence between the curves for these two runs is applied to the no­

image run as a shield correction.

Figure 5 shows the results obtained on the pitching moment balance

with the 5-in. A. S. Projectile supported at the 0. 30 L, 0. 35 L, and 0. 41 L

points. Image shield runs were made using the pitching moment balance

but in every case the results were identical with the no-image runs.

The procedure in the High Speed Water Tunnel has been to support

all models as near the 0. 50 L point as possible. For the runs without

the pitching moment balance, the image shield correction for the o. 48 L

support point was negligible and it is interesting to note that this curve is

almost identical with those obtained on the pitching moment balance.

.3 5

0 (.) .30

...... z w - .25 (.)

lL lL w 0 (.)

.20

"' <(

"' 0

. 15

. 50

0 (.) .40

...... z w

~ lL lL

.30 w 0 (.)

"' <( a:: 0

.20

. 50

0 (.) .40 !-"" z w 0 lL lL w . 30 0 (.)

"' <(

"' Cl

.20

0 C; 0

0 L\

C;

2

-5-

) ~RRECTED FOR PIT CH ING MOMENT

~ u L C; C;

UNCORRECTED

l 4

REYNOLDS NUMBER X I 0- 6

l

v 0 0 C; C; C;

6 8

Fig. 3 - Pitching moment correction for model of 6-in. projector, 0. 22L support point .

I I I

10

I SUPP ORT PO I NTS

=

I 2 4

REYNOLDS NUMBER X 10-6

/ / ,

I I ~.48L Vr . 35L

'/ 30L

-

6 8 10

Fig. 4 - Effect of mode 1 support point -on drag coefficient of 5-in. A. S. Projectile without pitching moment balance. Curves corrected for shield interference .

61

SUPPORT POINTS

0 . 41 L

C; . 35 L

0 .30 L

i 2

.@ &SID eP

4

REYNOLDS NUMBER X 10- 6

~ ~ E

6

~ ~ ~ 4iD ~

8 10

Fig. 5 - Effect of model support point on drag coefficient of 5-in. A. S. Projectile. Model mounted on pitching moment balance. Image shield runs are included.

12

12

12