roswell

AuGusT 1948 AT HE L STAN F. S PI L H AU S, C. S. SCHNEIDER, AND C. B. M 0 0 R E 135
s
so
FIG. 7. Height-time curve of balloon Flight 17. Released at
Alamogordo, New Mexico, on 9 September 1947 at 1647 MST
(105th meridian). Recovered near Pratt, Kansas, 530 miles
distant.
constant leakage held the balloon at 16,800 ± 700ft
for 7 hours. The duration of the flight was 9i hours.
When the original 2700-gram ballast was expended,
the balloon descended rapidly. Even had the auto-
matic ballast valve been functioning, the constancy of
altitude would have been the same. This seems to indi-
cate that only a minimum of automatic control is
needed, provided that diffusion losses are slightly
overcompensated by a constant ballast leak.
Other flights also indicate the importance of a check
valve in the balloon appendix to prevent dilution of
the lifting gas with air. If this is not done, the altitude
reached is far under the theoretical altitude deter-
mined by the displacement and gross load.
7. Control systems
Two systems of control are possible with the equip-
ment as described. The balloon is controlled between
an upper level (ceiling), where the full balloon buoy-
ancy just equals the load, and a lower level (floor),
below which the automatic ballast valve operates.
Schematic curves for these two systems of control are
shown in fig. 6.
In the first system of control the rate of static ballast
leakage is greater than the diffusion loss of lift, and
the balloon will stay at the ceiling. If it is displaced
above the ceiling the buoyancy is insufficient to bal-
ance the load and it will descend again. Provided the
rate of ballast discharge is greater than the rate of lift
by loss of gas this ceiling will slowly rise by valving of
gas, and as gas is lost by diffusion. The less the amount
of gas the lower the pressure (higher ceiling) must be
for the gas to fully distend the envelope. Unnecessary
valving is undesirable and may, in part, be minimized
by use of a restraining safety valve set in the appendix,
which will allow some slight pressure to be carried in
the balloon, preventing gas loss at the peaks of minor
oscillations but still valving gas before the balloon
ruptures due to too great an internal pressure.
In this system of control, the automatic valve is not
sealed off until the balloon starts a descent due to
cooling or other changes in lift, as when night falls.
Upon descent the valve is activated and starts drop-
ping ballast 'immediately; this continues until the
balloon is no longer losing lift at a rate greater than
the diffusion losses. The balloon will then rise above
its former ceiling to a height determined by the weight
of ballast dropped, and remain there as long as there
is ballast to compensate for lift losses. Flight 17, repro-
duced in fig. 7, used a low-leakage balloon and is an
actual case of ceiling control. It may be compared with
the idealized time-altitude curves in fig. 6.
In the second system of control the static rate of
leakage is less than the diffusion loss of lift. hi this
case the balloon will descend to the floor, where the
automatic control operates and the balloon floats at
an equilibrium altitude where the rate of ballast re-
lease exactly balances the rate of loss of lift. Floor
control conserves ballast, since only that needed for
altitude control is released. However, the altitude of
the floor varies diurnally as the temperature of the
entrapped air in the automatic ballast valve is affected
by solar radiation. Two methods are being investigated
to circumvent this undesirable feature. One is to
w
s
FIG. 8. Wind vectors at 16,000 feet for El Paso (EO), Albu-
querque (AB), and Roswell (THJ), at 03h, 09h and 15h (MST) on
7 July 1947, in connection with balloon Flight 11, mean motion
of which is shown by the balloon vector. Cross-hatched sector
contains all wind vectors at these three stations for the three
observation hours and for the three levels, 14,000, 16,000, and
18,000 feet.