Skip to main content

Full text of "NASA Technical Reports Server (NTRS) 19850007393: The UH-1H helicopter icing flight test program: An overview"

See other formats

yrtosJ-X/fP- *7^ 

NASA Technical Memorandum 86925 


The UH-1H Helicopter Icing Flight 
Test Program: An Overview 

Robert J. Shaw and G. Paul Richter 
Lewis Research Center 
Cleveland, Ohio 


Prepared for the 

Twenty-third Aerospace Sciences Meeting 

sponsored by the American Institute of Aeronautics and Astronautics 
Reno, Nevada, January 14-17, 1985 

E- 2421 


Robert J. Shaw and G. Paul Richter 
National Aeronautics and Space Administration 
Lewis Research Center 
Cleveland, Ohio 44135 



An overview is given of the elements of an 
ongoing joint NASA/Army program to study the 
effects of ice accretion on unprotected helicopter 
rotor aerodynamic performance. This program inte- 
grates flight testing, wind tunnel testing, and 
analytical modeling. Results are discussed for 
helicopter flight testing in the Canadian NRC hover 
spray rig facility to measure rotor aero perform- 
ance degradation and document rotor ice accretion 
characteristics. The results of dry wind tunnel 
testing of airfoil sections with artificial ice 
accretions and predictions of rotor performance 
degradation using available rotor performance codes 
and the wind tunnel data are presented. An alter- 
native approach to conducting future helicopter 
icing flight programs is discussed. 




P s 

P m 







(iCdo) g V g 

a HP 





main rotor area (168.11 m ) 

airfoil lower, upper surface pressure 
coefficients where C p = (P s - P^J/Q, 

rotor airfoil chord (0.533 m) 

icing cloud liquid water content 

local static pressure 

free stream static pressure 

free stream dynamic pressure 

rotor radius (7.315 m) 

radial coordinate 

rotor shaft horsepower 

airfoil axial coordinate 


average increase in rotor section drag 

change in rotor power coefficient 
where Cp = (Power)/p„A(fiR) 3 

average increase in horsepower required 

change in rotor horsepower required 

density ratio 

ambient density 

rotor rotational speed 

Some of the most difficult aircraft icing 
problems are those associated with the helicopter. 
In recent years, an increased emphasis has been 
placed on the need for both military and civilian 
helicopters to be able to fly into forecast icing 
conditions. To date only one civilian rotorcraft, 
the French Super Puma, has received icing certifi- 
cation from the FAA. The rotor of the helicopter 
presents some of the most complex icing problems. 

As Ref. 1 points out. 

"Rotor icing and ice protection techniques 
present a unique set of problems (aerodynamic, 
dynamic, thermodynamic) chat are not easily 
solvable through fixed-wing ice protection 
technology. Ice formations on the rotor cause 
degradation in helicopter performance, the 
degree of performance change depending upon 
specific rotor characteristics (airfoil pro- 
file, blade loading, Mach number distribution, 
catch efficiency, blade flexibility, rotor/air- 
frame reaction, control loads, etc.)" 

A key step in developing the required tech- 
nology base needed for solving helicopter rotor 
icing problems is the acquisition of flight data 
for both simulated and natural icing conditions for 
unprotected rotors operating in hover and forward 
flight. This data base should include detailed 
measurements of the rotor ice accretion shape 
characteristics and the resultant rotor performance 
degradation levels (e.g., increased levels of 
required horsepower to sustain flight). This data 
base can then be used to validate analytical 
techniques being developed to predict rotor ice 
accretion and resultant aerodynamic performance 
degradation, ^ and to evaluate the usefulness of 
studying rotor icing characteristics using experi- 
mental test rigs such as full scale oscillating 
rigs and model rotors in icing wind tunnels. Also, 
the data base can be used to determine the effects 
of forward flight on ice accretion characteristics. 

The validation of simulation approaches such 
as those described above should reduce the amount 
of natural icing flight testing required for future 
helicopter icing certification programs. It is 
well recognized that natural icing flight testing 
is a very expensive, uncontrolled approach to 
evaluating aircraft performance in icing. This is 
particularly true for the helicopter which has a 
limited range of operation. Often many seasons of 
flight testing are required to acquire an adequate 
data base. ^ 

With these thoughts in mind, NASA and the U.S. 
Army have undertaken a joint program called the 
Helicopter Icing Flight Test Program (or HIFT Pro- 
gram) to begin to acquire the needed rotor icing 
data. Complementary efforts have also been started 
by NASA to evaluate the analytical and ground-based 
experimental simulation approaches. This paper 
will present an overview of the current program 


status and indicate possible future efforts. The 
HIFT Program required coordinated efforts from many 
organizations. Table 1 shows the responsibilities 
of the various organizations involved in the 

The combined efforts of many individuals from 
the U.S. Army, Air Force Arnold Engineering 
Development Center, the Ohio State University, 

Hovey and Associates, and Bell Helicopter Textron 
made the HIFT program possible. 

HIFT Program Objectives and Organization 

The phase one efforts of the HIFT program 
involved helicopter icing in the hover mode. 
Specifically, the major elements of the program are 
shown in Fig. 1 along with an indication of how the 
elements relate to one another. The following 
paragraphs will discuss some important details 
associated with each of the elements. 

Helicopter Icing Flight Testing 

The hover icing tests were conducted in the 
Canadian National Research Council's Ottawa Spray 
Rig located adjacent to the Uplands airport. This 
outdoor facility, the only hover icing facility 
currently available for flight tests, is fully 
discussed in Ref. 4. The spray rig is a framework 
(Fig. 2) which measures 75 ft wide by 15 ft high 
and is attached to a 59 ft mast. A total of 156 
steam atomizing nozzles are located on this frame- 
work, and they are spaced 3 ft apart on 30 vertical 
bars. The icing cloud is created by using pres- 
surized steam to atomize water into a spray with 
the desired water droplet sizes. Icing cloud 
liquid water content (LWC) is primarily governed 
by water flow rate and wind speed while the droplet 
size is primarily a function of the steam pressure 
level. The framework can be rotated in order to 
align it perpendicul ar to the prevailing wind. The 
spray rig requires a wind velocity of at least 
6 knots to carry the cloud over the helicopter. 

The helicopter used for the HIFT Program was a 
Bell Helicopter Textron UH-1H "Huey" provided and 
operated by the U.S. Army Aviation Engineering 
Flight Activity (USAAEFA). The UH-1H is a 
thirteen-place, single engine configuration which 
has a single, two-bladed teetering main rotor. The 
maximum gross weight of the UH-1H is 9500 lb. The 
helicopter is powered by a Lycoming T53 engine, and 
the main rotor transmission is limited to 1100 shp 
for continuous operation. Figure 2 shows the 
UH-1H helicopter in the Ottawa spray rig facility 
while Fig. 3 shows a schematic of the UH-1H with 
important details noted. The required instrumen- 
tation was provided by the Army and installed by 
Army personnel (USAAEFA) with the exception of the 
main rotor blade load instrumentation which was 
installed by Bell Helicopter Textron personnel. A 
complete listing of the helicopter instrumentation 
is contained in Ref. 5. 

Ice Shape Documentation 

Three different approaches were employed to 
document rotor ice accretion shapes and an evalu- 
ation was made of the relative merits of each. The 
three approaches employed were, (1) stereo photo- 
graphy, (2) silicone molding, and (3) cross- 
sectional tracings. Ice shape documentation 
occurred after the helicopter was flown in the 

spray rig and then landed, with the iced rotor 
brought to rest. 

The ice accretion documentation phase was con- 
ducted in a special climate controlled work station 
shown in Fig. 4. The work station, a modified 
airline galley truck with inside dimensions of 20 
by 7 by 4 ft, was backed up to the helicopter so 
that one of the main rotor blades could be posi- 
tioned within the truck. The truck interior was 
then sealed from the outside by using the truck 
door and a tarp and thejntenor temperature 
brought up to about -5 “C using space heaters. The 
documentation station was kept at subfreezing tem- 
peratures to ensure that no ice melted during the 
time the liquid molding material required to 

The stereophotographic technique was developed 
by Arvin/Calspan Field Services, Inc. for the 
Arnold Engineering Development Center (AEDC) of the 
Air Force Systems Command. Wide angle stereo pairs 
of photographs of the rotor leading edge ice 
accretions were taken using two 70 mm Hasselblad 
cameras equipped with 50 mm lenses. The stereo 
pair photographs were taken over the span of the 
iced rotor blade and were centered on 1 ft-wide 
segments every 2 ft starting 5 ft from the hub. 
Figure 5 shows the overall stereophotography setup 
employed. Three sets of photographs were taken at 
each location with the two cameras located 
approximately, (1) 1 ft above the rotor chord 
plane, (2) on the chord plane, and (3) approxi- 
mately 1 ft below the rotor chord plane. Post 
flight analysis of the stereo pair photographs in 
order to determine local two-dimensional ice 
accretion shapes was performed at AEDC using an 
available stereo-compiler. Reference 6 gives a 
more detailed discussion of the stereo photography 
set up and data reduction procedures employed. 

The silicone molding technique involved placing 
1 ft wide plywood mold boxes around the iced UH-1H 
rotor every 2 ft along the span. The mold boxes 
were designed and fabricated to ensure a tight fit 
around the airfoil contour and to allow for molding 
of an ice accretion as much as 2 in ahead of the 
airfoil and 6 in aft of the leading edge. Figure 6 
shows the details of the mold box construction. 

The material used to form the molds was a mixture 
of Dow Corning silicone rubber (RTV 3110) mixed 
with thinner and catalyst compounds. The choice 
of this mixture was based upon the results of 
studies conducted at NASA Lewis and information 
from previous research efforts available in the 
literature. Prior to use, the molding mixture was 
degassed to remove unwanted air bubbles using a 
vacuum pump arrangement. The degassing process 
took about 15 min to complete. 

The molding mixture was made thin (i.e., the 
consistency of cake batter) so that when it was 
poured into the mold box cavities, it would flow 
freely around the ice in order to document the 
small scale characteristics of the ice accretion. 

It generally took about 2 to 3 hr (at temperatures 
below freezing) for the silicone mixture to harden 
to a sufficient degree that the mold boxes could 
be removed from the rotor blade. Figure 7 shows 
the UH-1H rotor blade with mold boxes attached. A 
more complete discussion of the molding technique 
is given in Ref. 7. 


The tracing technique involved making chordwise 
cuts of the ice accretion every 2 ft of the rotor 
span using a hot wire electric knife. A cardboard 
template was then inserted into the cut, and a 
pencil tracing was made of the leading edge ice 

Dry Transonic Wind Tunnel Tests 

Selected silicone molds were used to fabricate 
artificial leading edge ice accretion shapes which 
were attached to full scale rotor sections. The 
sections were tested in a dry transonic wind tunnel 
to measure local airfoil section performance deg- 
radations due to the ice accretions. Figure 8 
shows a typical casting fabricated from epoxy and 
affixed to a UH-1H rotor section. Surface static 
pressure taps were installed over the forward 35 
percent chord of the models to determine the 
effects of the ice accretions on airfoil boundary 
layer development. 

The wind tunnel tests were conducted in the 
Fluidyne Engineering Corporation (Minneapolis, 
Minnesota) 66 by 66 in Transonic Wind Tunnel 
Facility. This Facility is an atmospheric total 
pressure wind tunnel with a slotted wall test sec- 
tion and is driven by a set of ejectors aft of the 
test section. Model drag levels were measured with 
a wake rake. A more complete discussion of the 
Facility is given in Ref. 8. 

Analytical Predictions of Rotor Performance 

As already indicated, the Ottawa Spray Rig 
requires a wind speed of at least 6 knots to carry 
the icing cloud out over the helicopter. Thus, the 
helicopter flight environment is not truly hover 
and this small but finite forward velocity must be 
accounted for in the analysis of the rotor per- 
formance. To perform this analysis, two different 
approaches were tried. The first approach was a 
conventional forward flight analysis methodology 
developed by Bell Helicopter Textron and called 
ARAM 45.9 -[-his analysis is based on blade ele- 
ment momentum theory and includes a capability for 
treating unsteady aerodynamic, compressibility, and 
finite blade effects. 

The second approach was an extrapolation tech- 
nique also developed at Bell which corrects the 
predicted hover performance power levels by making 
momentum theory corrections to predictions of both 
profile and induced power levels. This approximate 
approach is discussed in detail in Ref. 10. 

The next section will discuss some of the 
results for each of the major elements of the phase 
one program. 

Results and Discussion 
Hover Flight Testing 

As already indicated, the two major objectives 
of the flight testing were, (1) to measure rotor 
performance degradation due to icing, and (2) to 
document the rotor ice accretions. Pretest dis- 
cussions between NASA, Army, and Bell Helicopter 
Textron personnel resulted in a decision to attempt 
to fly a series of tethered hover flights before 
and after accreting ice in the spray rig. However, 
initial test flights revealed that a large portion 
of the ice was shed due to rotor blade flexing 

when the second set of tethered hover points was 
acquired. The test technique finally adopted was 
as follows: 

(1) Determine rotor baseline profile power 
levels required by conducting a flat pitch 
speed runup with the aircraft on the ground 

(2) Fly a baseline out-of-ground effect (OGE) 
free hover point 

(3) Enter the spray rig cloud, accrete ice, 
and exit the cloud 

(4) Fly a free hover point to determine hover 
performance degradation 

(5) Land the aircraft and conduct a second 
flat pitch speed runup 

(6) Bring the rotor to rest, move the ice 
shape documentation station into place, and 
document the rotor ice accretion 

This test technique was judged to be successful in 
that ice shedding was minimized. 

A total of ten research flights were conducted 
during the flight test program which lasted from 
mid January to mid March (1983). Unfortunately, 
unseasonably warm weather occurred in the Ottawa 
area during this period, preventing any more 
flights from being attempted. For five of the ten 
flights, the ice accretion was retained on the 
rotor blades. A summary of these five flights 
(designated as flights A through E) is given in 
Table 2. The liquid water content figures shown 
were determined from the NRC calibration of the 
spray rig^ which requires an estimate to be made 
of the wind gust conditions ("low, medium, or 
high"). The estimates for the various flights are 
also given in the table. The volume median droplet 
diameter for all test conditions was 30 u as 
determined from the available NRC calibration. It 
should be noted that the main rotor torque instru- 
mentation was inoperative for flights C and D. 

Thus, "complete" data sets (i.e., ice accretion and 
rotor performance degradation levels) were acquired 
for flights A, B, and E. It is these three flights 
which the remainder of the paper will focus on. 
However, tracings for flight D are presented in 
Ref. 5 and molds and stereo photographs were 
acquired for flight C. 

The flat pitch speed runup procedure was 
employed to acquire data which would, (1) verify 
that the main rotor performance instrumentation was 
functioning properly, and (2) allow an estimation 
to be made of the mean profile drag coefficient for 
the iced rotor while operating at an approximate 
zero lift condition. 

Profile power is an indication of the power 
required to pull the rotor through the air and 
varies linearly with the cube of the angular 
velocity. Figure 9 shows the profile power curves 
for the baseline (clean rotor) measurements and the 
corresponding curves for the iced rotor for flights 
A, B, and E. The data shown in Fig. 9 were cor- 
rected for compressibility using a standard 
Prandtl-Glauert correction. Average rotor profile 
drag coefficients ( Cd 0 ) avg were calculated. 

The results summarized in Table 3 indicate flight 
E had the highest ( Cd 0 ) avg value (0.0110) for 
the three flights. Table 3 also indicates the 
average increases in profile rotor horsepower 
ranged from 60 for flight E to 25 for flight A. 


Table 4 summarizes the hover performance 
measurements made for flights A, B, and E. Also 
noted in the table are the maximum ice thicknesses 
measured at the rotor mid span station and the 
percent of span of the rotor over which ice had 
accreted. (That is for flight E, ice was accreted 
over the first 92 percent of the span.) As the 
table indicates, the rotor hover performance deg- 
radation was the greatest for flight A (+101 shp 
expressed in eguivalent standard day conditions). 
This is just opposite to the results of the profile 
power measurements (Table 3) where flight A showed 
the smallest average profile power coefficient for 
the three flights. 

It is hard to definitely explain this apparent 
anomaly. However, the flight log for flight A 
indicates that some ice over the rotor span was 
shed during landing. Thus, the rotor span coverage 
noted on Table 4 corresponds to the extent of ice 
remaining after the second flat pitch runup was 
completed, and thus a direct comparison of per- 
formance losses for flat pitch runup and for hover 
for the three flights is not justified. Previous 
analytical studies of rotor aerodynamic performance 
degradation due to icing 11 have shown that the 
rotor is most sensitive to ice accretions over the 
outermost portion of the span, and the loss of 
outer span ice for flight A may explain why the 
horsepower increase for flat pitch runup was so 
low. The instrumentation appeared to be working 
properly for this flight as indicated by the 
agreement in clean rotor profile power levels 
measured for the three flights (Fig. 9). 

The available rotor ice accretion tracings for 
the three flights are shown in Fig. 10 for various 
radial locations. As the figure indicates, only 
three tracings were taken for flight A, those being 
at r/R = 0.31, 0.40, and 0.49. However, it has 
already been indicated that ice coverage was noted 
out to r/R = 0.75 but unfortunately no ice shape 
details were preserved for this critical outer span 
region. The three tracings for flight A show rel- 
atively rounded ice accretions which look something 
like rime ice shapes which characteristically 
result in smaller airfoil performance losses than 
do glaze ice accretions. 

The tracings available from flight E show less 
streamlined characteristics over the inner span 
(r/R = 0.56) with a transition to smaller, more 
rounded shapes over the outer span region. In 
particular, the ice tracing at r/R = 0.81 is very 
small and rounded. 

On comparison of these three sets of ice shape 
tracings, one might conclude that while the inner 
span ice accretion shapes for flight E appeared to 
be worse than those for flight A, the outer span 
ice accretion characteristics for flight A (which 
were not documented) might have been more disrup- 
tive to the local flow over the rotor than were 
those for flight E. Clearly, more outer span 
documentation for the two flights should have been 

This possible anomaly with the hover ice 
accretion/aerodynamic performance degradation data 
for flights A and E points out some inherent prob- 
lems with the basic test procedure which had to be 
employed. As already indicated, the desire was to 
measure a thrust-power curve for the UH-1H with the 
rotor ice accretions. However, the flexing of the 

rotor blades during the data acquisition resulted 
in a large amount of ice shedding and thus the data 
was rendered useless. Thus, only one point on the 
thrust-power curve could be obtained prior to 
landing the aircraft and completing the flat pitch 
runup. This is obviously an undesirable situation 
in that helicopter flight test data and in partic- 
ular that for the UH-1H characteristically exhibit 
some scatter, and it is desirable to acquire large 
quantities of data in order to determine a statis- 
tically significant curve. It is possible that 
some of the discrepancy in increased power 
requirements for flights A and E could be explained 
by data scatter. 

Also, it should be noted that the Ottawa spray 
rig requires a minimum wind speed of about 6 knots 
to carry the icing cloud out to the rotor. Thus, 
the helicopter is not operating in a true hover 
mode either when the ice is being accreted or when 
the performance measurements are being made. The 
original test plan for the flight testing had 
specified that a precision low airspeed sensing 
system be installed in the UH-1H and be used as an 
airspeed indicator. However, such a system was not 
available and as a result, a cup anemometer located 
on top of the spray rig had to be used. This 
anemometer system would be expected to be somewhat 
less accurate than the low airspeed sensing system. 
The low forward flight velocity regime is char- 
acteristically a very sensitive one for a heli- 
copter as the aircraft is very sensitive to gusts 
which can cause significant changes in power 
required. At such low velocities, the induced 
power is sensitive to wake position which can 
significantly vary with small changes in forward 
velocity. In order to assess this sensitivity, 
analysis was carried out at Bell Helicopter 
Textron using a forward flight rotor performance 
prediction code to determine change in power 
required for the UH-1H as a function of forward 
velocity (information received in private 
communication from L.F. Berkowitz of Bell 
Helicopter Textron). The results are shown in 
Fig. 11 where the horsepower required increases 
for flights A, B, and E are compared to the 
decrease in horsepower required when the forward 
velocity was increased from 5 to 15 knots. As the 
figure clearly indicates, the magnitudes of the 
changes in horsepower required are the same for 
the icing flights as that due to a change in 
forward velocity. This figure indicates the 
forward velocity must be known very accurately 
when measuring performance losses due to icing at 
very low forward flight speeds. 

More recent high speed wind tunnel tests of 
scale rotor airfoil sections 12 have indicated 
that appreciable erosion/subl imation effects can 
occur when the iced airfoil is exposed to a high 
speed airflow for several minutes. In particular, 
that study showed that the section drag coef- 
ficients could be reduced by as much as 40 percent 
below the level measured immediately after the ice 
accreted if the iced airfoil was exposed to a high 
Mach number air flow for several minutes. The 
performance measurement technique employed during 
the HIFT phase one flight testing required several 
minutes to complete the rotor hover performance 
measurement, land the aircraft, complete the flat 
pitch power runup, and bring the rotor to rest to 
document the ice accretion characteristics. Since 
the outer regions of the UH-1H rotor are subjected 
to relatively high Mach number environments 


( i p ~ °-7 t0 0.8), ice accretion erosion/subl l- 
mation could have occurred to significantly affect 
the rotor performance measurements made. However, 
the magnitude of this effect cannot be quantified 
at the present time. 

With the above thoughts in mind, some comments 
regarding possible approaches to follow in future 
tests seem to be in order. It would seem to be 
very important to have some in flight photographic 
techniques to document the rotor ice accretion 
characteristics. If a hub mounted camera system 
such as that developed by the British^ or the 
rotor head video camera being developed by the Army 
were available, a determination of span coverage 
of the ice and any shedding which occurred could 
be determined. However, these systems could not 
provide any quantitative details of the ice accre- 
tion. A proposed system which might yield such 
information would be a stereo photography system 
such as that proposed by NASA and AEDC person- 
nel.^ This approach is attractive in that 
conceptually it would be possible to document ice 
accretion characteristics at the same time the 
performance degradation measurements were being 
made. However, it must be acknowledged that the 
stereo system discussed in Ref. 13 still requires 
significant development. 

Also, an alternate and possibly more attractive 
experimental approach can be suggested. The 
helicopter rotor would be allowed to accrete ice 
for the desired length of time and then the air- 
craft would be immediately landed, taking care not 
to shed any ice by any abrupt changes in collec- 
tive pitch, etc. The ice accretion along the rotor 
would immediately be documented using one or more 
of the techniques discussed above. Bypassing the 
acquisition of any iced rotor performance data 
would minimize the ice accretion exposure to the 
high speed erosion environment and thus preserve 
the "best" ice accretion definition. After the 
icing flight testing is completed, the ice shape 
documentation can be used to fabricate artificial 
ice accretions which would be affixed to the lead- 
ing edge. Of course these artificial ice accre- 
tions should have the proper mass characteristics 
to accurately simulate the effects of real ice 
accretions on rotors. Conventional dry air flight 
test techniques could then be employed with the 
artificially iced rotor to accurately determine the 
performance degradation due to icing. Of course, 
this approach is predicated on the assumption that 
it is possible to accurately model the key features 
of natural ice accretions so that the aerodynamic 
flowfield will be adequately duplicated. Research 
to verify this assumption is currently underway.^ 

This method does suffer from the drawback that 
additional flight testing is required. However, 
it is felt that the additional expense associated 
with this testing can be justified on the grounds 
that the quality (and hence bel levabi 1 ity) of the 
performance degradation measurements will be 
greatly improved. 

Ice Shape Documentation 

On site inspection of the rotor ice accretions 
indicated that a considerable amount of local three 
dimensional detail existed 1 ^ as revealed in 
Fig. 12. The tracing technique and to a certain 
extent the stereo photography technique capture the 
two dimensional nature of the ice accretion at the 

spanwise station of interest. The tracing tech- 
nique requires the least time and effort to employ, 
but it cannot be used to document any of the 
details of the surface roughness characteristics 
of the ice accretion. However, as Ref. 6 points 
out, the stereo photography approach does show 
promise for being able to document the roughness 
characteristics, but additional research and 
development is required. 

The silicone molding technique appears to make 
almost exact reproductions of the ice accretion 
characteristics, including the most minute three 
dimensional details. Figure 13 shows two views of 
a mold obtained during the program. While the 
photographs do give some appreciation for the 
detail preserved, a personal inspection is required 
to truly appreciate the details. However, it 
should be pointed out that the silicone mold does 
require a significant amount of time to harden 
(approximately 2 to 3 hr), and thus the amount of 
data which can be acquired is limited. 

Figure 14 compares the ice shape details 
available from the three methods. The profiles are 
shown for r/R na 0.46 for flight E. To reduce the 
silicone mold to a two dimensional cross section, 
a photograph was taken of the edge of the appro- 
priate mold. A comparison of the three tracings 
indicates that some differences in ice shape char- 
acteristics did exist between the three methods; 
however, these variations are judged to be no 
greater than those that would be obtained by using 
any one method and taking several closely spaced 
"slices." That is, the ice accretion exhibited 
some locally three dimensional details as already 
discussed (Fig. 12). 

Dry Wind Tunnel Test Program 

Flight E was chosen for follow on dry wind 
tunnel tests of rotor sections with artificial 
leading edge ice accretions and then computer 
analyses of rotor performance degradation. This 
choice was dictated primarily by the relatively 
large degradation in hover performance measured as 
well as the existence of rotor ice accretion out 
to about 92 percent span. 

Based upon rotor performance analyses conducted 
by Bell Helicopter Textron personnel (information 
received in private communication from l.F. 
Berkowitz of Bell Helicopter Textron) and available 
rotor ice accretion documentation, four stations 
were chosen for wind tunnel testing. The four 
stations chosen were at nondimensional radial 
locations (r/R) of 0.44, 0.60, 0.77, and 0.94. 

Since ice accretion occurred to r/R of about 
0.92 for flight E, only three "iced" airfoil models 
were tested in addition to a clean airfoil model 
which served as a baseline reference. The con- 
ditions chosen from wind tunnel testing were 
determined from the rotor analysis performed by 
Bell Helicopter Textron. Since the rotor ice 
accretion and performance data were acquired in a 
near hover mode (i.e., a finite forward wind speed 
existed), a forward flight analysis code had to be 
used and thus the local Mach number and angle-of- 
attack conditions for each radial station did vary 
with azimuthal position of the rotor blade. 

Figure 15 shows the two boundary curves for rotor 
section operation conditions, the lower curve 
corresponding to flat pitch runup conditions and 
the upper curve corresponding to the near hover 


performance conditions. The figure also shows the 
critical flow boundary for the NACA 0012 airfoil, 
and it can be seen that supercritical conditions 
would be expected for the r/R = 0.77 and 0.94 
stations. Wind tunnel test points for each model 
were chosen to be within the two boundaries and 
along the appropriate operating line shown in the 
figure so as to adequately document any Mach 
number or angle-of-attack effects on section drag 

The wind tunnel test models were fabricated 
using the appropriate flight E silicone molds and 
full scale sections of UH-1H rotor blades supplied 
by the U.S. Army. Replicas of the leading edge ice 
accretions were cast to the rotor section models 
using epoxy as the casting material. To cover the 
span of the wind tunnel test model, the mold was 
used to make repetitive castings starting at the 
midspan. The models were instrumented with surface 
static pressures over the first 35 percent chord 
to determine boundary layer growth characteristics. 
Figure 16 shows one of the models installed in the 
Fluidyne Transonic Wind Tunnel. 

The section drag coefficient levels measured 
for the three test models corresponding to r/R = 
0.44, 0.60, and 0.77 are shown in Fig. 17 as a 
function of angle-of-attack which has been cor- 
rected for tunnel downwash interference. The clean 
airfoil reference drag levels are also shown. It 
should be noted that the Mach number was varied as 
a function of angle-of-attack for those models 
corresponding to the r/R = 0.60 and 0.77 iced 
airfoils so as to choose test points which were 
close to the respective operating lines shown in 
Fig. 15. The experimental Mach number variation 
was 0.42 to 0.54 for the r/R = 0.60 accretion 
model and 0.53 to 0.69 for the r/R = 0.77 accre- 
tion model. For the most inboard model (r/R = 
0.44), the Mach number was taken to be constant 
(0.36). Also for reference, the figures show ref- 
erence ice shape tracings which are the same as 
some of those shown in Fig. 10. 

As the figure indicates, the largest drag 
levels were measured for the inboard ice accretion 
(r/R = 0 44) with a maximum drag coefficient of 
0.080 being recorded for a corrected angle-of- 
attack of 8.12°. The rapidly increasing drag 
coefficient level with increasing angle-of-attack 
suggests a large scale, upper surface flow sepa- 
ration had occurred. Figure 18 shows leading edge 
surface pressure coefficient variations both for 
the iced airfoil and the reference clean airfoil 
for an angle-of-attack of 5.9°. These profiles 
shown in differential form indicate a difference 
in the surface pressure characteristics for the 
iced airfoil occurred from x/c = 0.05 to 0.35. 

This reduction in level of pressure coefficient 
differential is indicative of a reduction in lift 
for the airfoil, and the nature of the iced airfoil 
curve (a plateau existing from x/c = 0.2 to 0.3) 
suggests the presence of a separation-reattachment 
zone. The presence of a localized separation- 
reattachment zone has been observed for other iced 
airfoils in other experimental programs. While the 
pressure differential plot of Fig. 18 suggests only 
a small change in lift occurred at a 5.9 angle-of- 
attack, the rapid increase in drag at the higher 
angles shown in Fig. 17 for the r/R = 0.44 ice 
accretion suggest the separati on-reattachment zone 
grew to a much greater size. For these higher 
angle-of-attack conditions, the effect of the 

leading edge ice accretion on the lift (and also 
pitching moment) characteristics would be expected 
to be more significant 

The drag coefficient curves for the other two 
ice accretion shapes (r/R = 0.60 and 0.77) show 
smaller increases relative to the clean airfoil 
baseline curves. This is to be expected as the 
corresponding ice accretion shapes are somewhat 
more "aerodynamic" in cross section as Fig. 17 

Even though the test matrix was very limited 
for each of the iced airfoil models (primarily due 
to program funding constraints), some understanding 
of the effect of the ice accretions on the UH-1H 
airfoil performance was determined. However to 
gain a more complete understanding of the airfoil 
performance many more wind tunnel test conditions 
would have to be run. These conditions should 
include Mach number variations at constant angle- 
of-attack and angle-of-attack variations at con- 
stant Mach number (for both low and high Mach 
numbers) . 

Rotor Performance Degradation Analysis 

As already indicated, two analysis methods were 
used to predict the UH-1H rotor performance degra- 
dation due to icing - a conventional forward flight 
analysis code and an extrapolation technique based 
on a hover analysis. Both analysis codes required 
that an airfoil data table be prepared which 
related iced airfoil lift and drag coefficients as 
a function of Mach number and angle-of-attack. 

Since airfoil section surface pressure profiles 
were not measured, the iced airfoil lift coef- 
ficient had to be assumed to be equal to the cor- 
responding clean airfoil lift coefficient. As 
already indicated in the previous section, this 
does not appear to be a bad assumption at least for 
moderate angles-of-attack. Thus, only a drag 
coefficient specification had to be provided. 

Both the hover and forward flight analysis 
codes required local section data for 21 radial 
locations for each 15° azimuthal position. (Of 
course for the hover analysis, there is no azi- 
muthal dependence.) Primarily due to program 
funding limitations, it was possible to conduct 
wind tunnel tests for only three different radial 
locations (r/R = 0.44, 0.60, and 0.77). Thus it 
was necessary to provide entries in the data table 
for the other radial locations. In the absence of 
any additional information, it was decided to use 
the small amount of data available (Fig. 17) to 
extrapolate/interpolate to the other conditions 
required. Obviously this is a gross approximation 
as it implicity ignores the variation in ice shape 
along the span of the rotor and the effect of 
changing ice shape on aerodynamic performance 
levels. However, it was felt that this was the 
best approximation which could be made. 

The results of the two analysis predictions 
performed by Bell Helicopter Textron personnel 
compared to the flight test measurement for flight 
E are shown in Fig. 19. The results are judged to 
be surprisingly good. The flight test results 
indicated an increase in required horsepower of 96 
while the forward flight analysis predicted 100 and 
the hover extrapolation method predicted 88. Of 
course, these results may be somewhat fortuitous 
in view of the several concerns already mentioned 


with regard to the normal scatter of the flight 
test data, limited amount of available flight test 
and wind tunnel data, and sensitivity of helicopter 
performance to small changes in forward velocity. 
Nevertheless, these results do appear to somewhat 
justify the correctness of the method and suggest 
the need for continued research is warranted. 

Concluding Remarks 

While the results of the phase one HIFT efforts 
were limited in terms of amount of flight test and 
supporting wind tunnel data which could be 
acquired, the program was judged to be successful. 
In particular, the agreement between predicted and 
measured rotor performance degradation was sur- 
prisingly good. Also, experimental techniques were 
developed which allowed rotor ice accretions to be 
documented in great detail without any apparent 
loss of any ice due to blade flexing during 

These techniques were employed to conduct a 
phase two of the HIFT program which involved flight 
of the UH-1H helicopter behind the Army's Heli- 
copter Icing Spray System (HISS tanker) to accrete 
ice in forward flight conditions, measure rotor 
performance degradation, and then land the heli- 
copter to document rotor ice shapes using the 
silicone molding technique. The results of these 
forward flight icing tests which were conducted 
near Duluth, Minnesota during the winter of 1983-84 
are given in Ref. 16. Analysis of this data is 
currently underway. Follow on wind tunnel tests 
and analytical predictions are also planned. 

In spite of the limited data available from the 
HIFT program some tentative conclusions can be 
drawn which should be considered for future heli- 
copter icing flight tests being planned. 

1. Both the phase one and two programs have 
shown it is possible to accrete ice on the rotor 
blades of an unprotected helicopter either in hover 
or in forward flight and then land the aircraft 
without shedding any significant amounts of ice. 

2. However, it does not appear to be possible 
to make a sufficient number of inflight performance 
measurements with an iced rotor to get a meaningful 
performance curve. Shedding a significant amount 
of ice from the rotor usually occurs during these 
performance measurements. The number of points 
which may be acquired are so few that the inherent 
data scatter may mask icing related trends. 

3. The time required to make the few inflight 
performance measurements which can be acquired is 
long enough that concern about erosion of the ice 
shapes especially near the high Mach number tip 
region exists. This ice shape erosion has been 
seen in at least one previous high speed icing wind 
tunnel test of scale rotor airfoil sections. 

4. The silicone molding approach appears to be 
a desirable documentation approach to take in that 
the most minute three dimensional details of the 
rotor ice accretion can be preserved, and highly 
accurate three dimensional castings of the rotor 
ice accretion can be fabricated from these molds. 

5. In view of the concerns suggested in (1) 
through (3) and with the availability of the 
silicone molding technique, it is suggested that 

future helicopter icing flight tests be limited to 
documentation of rotor ice accretion character- 
istics only, whether for hover or forward flight 
conditions. That is, no rotor performance 
degradation data be taken, rather the ice accretion 
molds acquired would be used to fabricate arti- 
ficial leading edge ice accretions and a second 
flight test be conducted with these artificial ice 
shapes of proper mass characteristics affixed to 
the leading edge. While this admittedly increases 
the amount of helicopter flight test time required, 
it is felt that with the artificial ice accretions 
a significant number of performance points could 
be acquired which would allow a more accurate 
determination to be made of rotor degradation due 
to icing. It is necessary to have a high level of 
confidence in the flight test data before the rotor 
icing analysis methodologies under development can 
be rigorously evaluated. A "one point" evaluation 
such as described in this paper is certainly not 
sufficient to validate the analysis methodologies. 

6. Additional research needs to be conducted 
to validate the assumption that artificial ice 
accretions affixed to the leading edge of rotor 
airfoil sections will cause the same changes in 
airfoil aerodynamic performance as will real ice 
accretions. The amount of surface detail required 
when fabricating the artificial ice shapes must 
also be determined. 


1. Peterson, A. A., Dadone, L., and Bevan, D. 
Rotorcraft Aviation Icing Research 
Requirements Research Review and 
Recommendations. ( D210— 1162— 1 , Boeing Vertol 
Co., NASA Contract NAS3-22384) NASA CR-165344, 
May 1981. 

2. Shaw, Robert J. Progress Toward the 
Development of an Aircraft Icing Analysis 
Capability. NASA 

TM-83562, 1984. 

3. Dunford, Philip J. New Techniques for Opti- 
mization and Certification of Helicopters in 
Icing Conditions. Presented at the AHS 
National Specialist's Meeting on Helicopter 
Testing Technology. (Williamsburg, VA) Oct. 

29 - Nov. 1, 1984. 

4. Bailey, D.L. Description of the Spray Rig 
Used to Study Icing on Helicopters in Flight. 
Second Revised Edition. Aeronautical Report 
LR-186A. National Research Council of Canada, 
Sept. 1960. 

5. Abbott, William Y., et al Evaluation of 
UH-1H Hover Performance Degradation Caused by 
Rotor Icing. USAAEFA Report 82-12, Aug. 

1983. ( AD-A141252 ) 

6. Palko, R.L., and Cassady, P.L. Photogram- 

metric Analysis of Ice Buildup on a U.S. Army 
UH-1H Helicopter Main Rotor in Hover Flight. 
AEDC-TR-83-43, Oct. 1983. (AD-B077844L) 

7. Lee, John D., Harding, Rorry, and Palko, 

Richard L.- Documentation of Ice Shapes 
on the Main Rotor of a UH-1H Helicopter in 
Hover. NASA CR-168332, Jan. 1984. 


8. Berger, Jack H, and McDonald, Timothy J.- 
Wind Tunnel Tests of Airfoil Shapes Altered by 
Icing and Airfoil Shapes With Deicer Boots. 
FluiDyne Report 1402, FluiDyne Engineering 
Corp., Jan. 1984. 

9. Harris, Franklin D., et al : Helicopter 
Performance Methodology at Bell Helicopter 
Textron. American Helicopter Society, Annual 
National Forum, 35th, Proceedings, American 
Helicopter Society, May 1979, pp. 79-2-1 to 

10. Harris, F.D.: Model 222 Performance Data Base 
for Hover and Low Speed Flight. BHTI 10M81, 
Bell Helicopter, Apr. 10, 1978. 

11. Korkan, K.D., Shaw, R.J.; and Dadone, L.: 
Performance Degradation of Helicopter Rotor 
Systems in Forward Flight Due to Rime Ice 
Accretion. AIAA 83-0029, Jan. 1983. 

12. Flemming, R.J., and Ledmcer, D.A.: Experi- 

mental Investigation of Ice Accretion on Rotor- 
craft Airfoils at High Speeds. AIAA-84-0183, 
Jan. 1984. 

13. Palko, Richard L., et al Initial Feasibility 
Ground Test of a Proposed Photogrammetric 
System for Measuring the Shapes of Ice 
Accretions on Helicopter Rotor Blades During 
Forward Flight. AEDC TR-84-10, Aug. 1984. 

14. Lee, J.D.: Aerodynamic Evaluation of a Heli- 

copter Rotor Blade with Ice Accretion in Hover. 
AIAA 84-0608, Mar. 1984. 

16. Abbott, W.Y., et al: Evaluation of UH-1H 

Level Flight Performance Degradation Caused by 
Rotor Icing. USAAEFA Report 83-23, July 1984. 





1. NASA Lewis 

Overall technical guidance, program coordination, 

partial fundinq 

2. Army Aviation Engineering Flight 

Instrumented Helicopter, flight test crew, performance 

Activity (AAEFA) 


3. Army Applied Technology Laboratory (ATL) 

Partial funding 

4. Dept, of Aeronautical and Astronautical 

Ice shape documentation coordination, dry wind tunnel 

Engineering, Ohio State University 

testing coordination 

(Prof. J.D. Lee) 

5. Canadian National Research Council 

Spray rig operation 

6. Hovey and Associates (Ottawa, Canada) 

Silicone molding, tracing 

7. Calspan (AEDC) 

Stereoscopic photography 

8. Fluidyne Engineering Corporation 

Dry wind tunnel tests 

9. Bell Helicopter Textron 

Rotor icing performance evaluation (analysis and 

experimental ) 









Time in 




































fi HP a vg 

(aCdo) aV g 























in cloud, 

ice thickness 
at r/R = 0.5, 




(xl8 5 ) 

standard day 
power change, 

























■=> c 

Cl, Cd, Cm 


Figure 1. - Major elements of HIFT program (phase one). 

14.6 m (48 ft) 

Figure 3. - Schematic of UH-1H helicopter used in the HIFT program. 

Figure 4. - Helicopter with the rotor ice accretion documentation station. 

Figure 6. - Mold box details. 

Figure 7. - UH-1H rotor blade with mold boxes attached, 

0 4 2.5 

Figure 8, - Epoxy casting affixed to UH-1H rotor section 




Figure 9. - Incompressible profile power curves 
(Ret 5 ). 



2 .3 

1 1 





.6 .7 

1 1 



.9 1.0 

_J 1 

Figure 10. - Rotor ice accretion tracings for flights A, B, and E. 

Figure 11. - Comparison of increased horsepower required 
due to icing with reduced horsepower required due to 
forward velocity increase. 

Figure 12. - Photo of typical ice formation on rotor blade showing three dimensional details (the white 
stripe is 1 in. wide), (ref. 7) 

Figure 13. - Typical molds obtained of ice formations, (ref. 7) 


Figure 15. - Rotor section operating boundaries (Ref. 15). 

Figure 16. - UH-1H rotor section model with artificial ice accretion 
installed in Fluidyne wind tunnel. 

Figure 18. - Surface pressure coefficient differential 
curve. 0. 35, a = 5. 9°, r/R = 0. 44. 


Figure 19. - UH-1H rotor performance degradation due to 
icing for flight E - experimental and predicted. 

1 Report No NASA TM-86925 2. Government Accession No 


3 Recipient's Catalog No 

4 Title and Subtitle 

The UH-1H Helicopter Icing Flight Test Program: An 


5 Report Oate 

6 Performing Organization Code 


7 Authors) 

Robert J. Shaw and G. Paul Richter 

8 Performing Organization Report No 

E— 2421 

10. Work Unit No 

9 Performing Organization Name and Address 

National Aeronautics and Space Administration 
Lewis Research Center 
Cleveland, Ohio 44135 

11 Contract or Grant No 

13 Type of Report and Period Covered 

Technical Memorandum 

12 Sponsoring Agency Name and Address 

National Aeronautics and Space Administration 
Washington, D.C. 20546 

14 Sponsoring Agency Code 

15 Supplementary Notes 

Prepared for the Twenty-third Aerospace Sciences Meeting sponsored by the 
American Institute of Aeronautics and Astronautics, Reno, Nevada, January 14-17, 

16 Abstract 

An overview is given of the elements of an ongoing joint NASA/Army program to 
study the effects of ice accretion on unprotected helicopter rotor aerodynamic 
performance. This program integrates flight testing, wind tunnel testing, and 
analytical modeling. Results are discussed for helicopter flight testing in the 
Canadian NRC hover spray rig facility to measure rotor aero performance degrada- 
tion and document rotor ice accretion characteristics. The results of dry wind 
tunnel testing of airfoil sections with artificial ice accretions and predictions 
of rotor performance degradation using available rotor performance codes and the 
wind tunnel data are presented. An alternative approach to conducting future 
helicopter icing flight programs is discussed. 

17 Key Words (Suggested by Authors)) 18 Distribution Statement 

Helicopter icing; Rotor icing; Wind- Unclassified - unlimited 

tunnel testing; Analytical modeling STAR Category 03 

19 Security Classif (of this report) 20 Security Classif (of this page) 21 No of pages 22 Price* 

Unclassified Unclassified 

*For sale by the National Technical Information Service, Springfield, Virginia 22161 

End of Document