[ See also COANDA and LANIER -- these 3 ought to get
"It's the speed of the air, not the airspeed."
1. Custer Channelwing: Intro
2. Custer Channelwing: The Facts
-- Too Good to be True ?
3. Custer Channelwing: Dreams and
Schemes -- Developing a Great Idea.
4. The Custer Channel Wing - P-20
5. Raul Colon : Willard Custer
Ideas, Ready for the Aviation World?
6. Williard Custer : Theory of
Channel Wing Aircraft
7. Popular Mechanics ( May
1947 ) : The Wing that Fooled the Experts
8. Air Progress Magazine (
October/November, 1964 ) ; Custer's Production Model
9. Walt Boyne : Airpower
Magazine, Volume 7 No. 3 ( May, 1977), pp.8-19, 58
; The Custer Channel Wing Story
Custer Channelwing: Intro
"It's the speed of the air, not the
CCW-5 five seconds into takeoff run.
This aircraft can be seen at the Mid Atlantic Air Museum, in
The idea of the channelwing pre-dates most of those who are
reading this site. It all began in the 1920s, when Willard
Custer took shelter in a barn during a near hurricane velocity
storm. Much to his surprise and fascination, the roof of the
barn suddenly lifted off, and soared through the air. He
wondered why an airplane had to gather speed on a runway, while
a barn roof, a poor airfoil by any reckoning, could fly from a
standing start. He soon came to the realization that it was the
speed of the air over the surface, not the speed of the surface
through the air, that created lift. Bernoulli principle in both
cases, but an application that had eluded aviation up to that
time. He settled on the idea of pulling the air through
channels that were, in fact, the lower half of a venturi.
He was reversing the normal method of powered flight. Instead
of using the engines to move the airfoil through the air, he
used the engine to move the air through the airfoil. His
channel had the effect of going several hundred miles per
hour, due to the induced air flow,while standing still.
The airflow over the surface of the channel created conventional
lift, and a lot of it. It was at this point that Custer settled
on," It's the speed of the air, not the airspeed", which
became his mantra of, "aerophysics".
CCW-1 hanging in the Smithsonian Garber facility.
Photo by Sam Smith
Many experiments followed with all nature of devices. The first
real aircraft to which he applied his princi;ple, was the CCW-1,
or Custer Channelwing number one, which now hangs at the Garber
facility of the Smithsonian. It is still strangely modern, even
after all these years, with a smoothly rounded fuselage, and a
wrap around plexiglass canopy. But, close inspection reveals the
channels appended with two by four struts. Two 75 HP engines
were fitted into the two six foot diameter half-barrel like
channels, and the tests were started. First flight was November
12, 1942. The CCW-1 was used for test purposes only, to prove
the concept. More than 300 hours of flight tests did prove that
the Custer not only flew, but was capable of flight without
wings. After the first flights proved stability, the wings
were progressively cut off or had spoilers attached to the
point of having no lift from the wings at all. The test pilot
noticed no difference because the channels furnished all the
lift needed! Most of these tests were low, straight ahead
hops. A demonstration took place in Beltsville Maryland for
Brigadier General W. E. Gilmore. Gilmore was noted for his gruff
temperament, but after seeing the demonstration, was excited
enough to place a call to Orville Wright, asking that he come
out to witness the Custer phenomenon. Orville didn't make it,
but the plane was placed in a military test program. The results
of these tests proved to be typical of the many government tests
the Channel wing received over the years. The Army Air Force
technical report concluded that the lift generated by the
channels was similar to normal induced lift created by other
wing /propeller arrangements. Although this was a complete
falsification, the damage was done, and Custer was on the
defensive. What they forgot to mention, was that the channelwing
created more static lift than the weight of the test vehicle,
and was, in fact, capable of vertical takeoff! The report
stated that the channelwing was inferior to the helicopter in
creating static lift and did not show sufficient promise of
military value to warrant further testing. This was at a time
when every conceivable concept from flying wings to rocket ships
was being tried. The conclusion, both then and now, seems
incomprehensible to say the least. To Custer, it was obvious
that the tests had been too good, and consequently helicopter
interests were pushing him out of the picture. That seems to be
the most likely scenario, as later tests proved the
channelwing to outlift helicopters of the day, with 13.8
pounds of lift per horsepower recorded. Custer was a good
inventor, but a little naive about politics and government
contracts. He also felt that the engineering staff and theorists
just didn't understand the Custer phenomenon, as they didn't
understand "aerophysics". But, if faith in the government was
dimmed, faith in himself wasn't. Over the next forty years, he
obtained financial backing for a series of aircraft from CCW-2
to CCW-5. He had enough data and tests to convince enough
investors to bring him near full production on at least two
In 1951, he co-operated with the Baumann Aircraft Company, and
modified one of their twin pusher aircraft to a Custer
configuration. This was the CCW-5, and had two 225 HP engines,
and weighed en excess of 4300 pounds. Walker Davidson made the
first flight of the CCW-5 in July of 1953. As usual, the
aircraft was highly successful. Demonstrations repeatedly showed
hair raising maximum performance takeoffs, nose high climbs
at speeds so low it seemed obvious that the Custer would fall
out of the sky. Three second takeoffs, with nose high steep
turns of 45 to 60 degrees bank, at speeds below 30MPH gave
the CCW-5 the ability to take off and do a 180 before most
planes could lift off. Video of these flights still confound
experienced pilots. Although I have personally logged 20,000+
hours, in all nature of aircraft, I was absolutely stunned the
first time I saw the videos of the Custer doing a 150 foot
takeoff, roll into a steep bank at speeds that would have
insured a stall - spin - crash,in any other plane, and leave
town going the other way, while staying within what appeared to
be about a 250' square area. Slow flight was a specialty, and
the CCW-5 flew at a measured 22 MPH and on August 27,
1954 hovered against an 11 MPH wind, although it was not
modified to use maximum lift potential. Cruise speed
remained a normal 170 mph.
These tests attracted more investors, and it seemed that Custer
and Noordyun Aircraft Ltd. of Canada were going to do a
production run of at least 100 aircraft. On the strength of this
proposal, a production version of the CCW-5 was built and rolled
out on July 4 of 1964. Although it looked like the original
Baumann conversion, the second model was built from scratch,
rather than modified from an existing aircraft. Now came the
securities and exchange commission who claimed the stock was not
issued correctly, and the deal fell through, in a manner
reminiscent of the Tucker car.
Since then, the Custer channelwing has virtually disappeared,
and few have even heard of the aircraft, let alone its'
capabilities. It has taken me 15 years to amass all the
information on the Custer. I now have in my possession reports
from Langley, Wright Patterson, several universities, several
independent engineering consultants, flight videos, use of the
CCW-5, reports from test pilots, and contact with Harold
"Curley" Custer. Curley is the son of Willard Custer, and has
more time in channelwings than any other man on earth.
Custer Channelwing: The Facts
Too Good to be True ???
In this section, I will show you the facts, as I have found
them. I know that I was a bit more than skeptical as I
researched the channewing. To me, it was obviously, "too good to
be true". As I uncovered more and more information, one thing
was constant - everyone who has firsthand knowledge of the
channelwing says, "yes, it works just as advertised". So that
you might join the ranks of, "true believers", this section will
contain only research data or pure verifiable facts, with the
documentation to prove it. Obviously there is too much data on
the channelwing to reproduce completely, but I hope to whet your
appetite with the following figures and data. But, before you
read this section, I will ask you to think of this exercise.
Design a simple (not even flaps) five passenger plane with
450 HP, capable of slow flying at 20 mph, 160mph cruise, 200''
takeoff and landing, with extreme load carrying ability.
If you can do it, you know more than any aeronautical
department, or aircraft manufacturer, and I'd like to see the
plans. On the other hand, the Custer CCW-5 did exceed all those
parameters, and did it many years ago. Take a look!
1. The custer channelwing is capable of vertical
CCW-2 tied to a windsock standard and levitating! As shown, it
is approximately 1000 pounds and powered by two 90 HP
engines. The aircraft would levitate at 2400 rpm.
Because Custer had extreme difficulty in convincing aviation
academics and manufacturers that the channelwing created a great
deal of static lift, he concocted this demonstration. The CCW-2
was tied to a windsock standard on a calm day, and run up. As
you can see, the aircraft levitated with the tie down rope
parallel to the ground. It was lifted on the channel power
alone. Even this did not convince the many skeptics. It was
thought to be swinging on a pendulum, or some kind of trick. It
was a classic case of NIH. (not invented here). They saw what
they believed, not believed what they saw. This would help
explain why Willard Custer had a tendency to lose his patience.
2. The channelwing is capable of hauling very heavy
The following NACA report shows the ability of a Custer channel
Angle of Attack 0° 20° 46°
Velocity lift (lb) thrust
(lb) lift (lb) thrust (lb)
lift (lb) thrust (lb)
0 mph 340 800
580* 635* 770 350
4 mph 360 795
650 600 840 280
11.5 mph 385 745
735 540 980 190
470 590 940
395 1375 -210
Data recorded at Langley Aeronautical Laboratory, Langley field
VA. and published in a 1953 NACA Research Memorandum RM L53A09.
These tests were conducted at less than full power (2450 rpm)
for reasons unknown, so they do not show full capability of the
CCW-2 lift capability. Because the CCW-2 was a "tail dragger",
the angle of attack while sitting on the ground was 20°. This
gives 580 pounds lift and 635 pounds thrust on the channel.
No channelwing ever built was configured for vertical flight,
and consequently, on the CCW-2 the pilot had to hold back on the
power until enough speed was gained to make the controls
effective. Because of this, the CCW-2 required a ground run
of 60 feet. Curly Custer did say that he flew the -2
without any form of aileron, and rolled with differential
throttle. He also has pictures to back up this claim.
I have heard that a modern flap system with leading edge flaps
and slats coupled with trailing edge slotted flaps can generate
a coefficient of lift of 5.5. But, I have never heard or even
imagined a co-efficient of lift of 23!
3. The unaided Custer channel is capable of generating
8.4 pounds static lift per horsepower.
In 1944, the USAAF issued a technical report on the
channelwing. The data was compiled in the 5 foot wind tunnel at
Wright Field in Dayton, Ohio. The tests were on the CCW-1 and
scale model channels driven by electric motors. remembering that
the CCW-1 was the first real attempt at flying the channels and
was in a rather embrionic state of development. It weighed 1375
pounds and was powered by two 75 HP engines. Even so, it was
reported the channels lowered the takeoff speed to 36 mph from
51 mph if the the channels were replaced by normal wing
sections. The experimental channel models, which were much
better finished, showed 8.4 pounds lift per horsepower. For some
reason, this was compared to a helicopter rather than other
fixed wing aircraft. Helicopters of the time could generate 15.2
pounds of static lift per horsepower, so the conclusion was that
the Custer channelwing was, "markedly inferior to the
helicopter, but superior to other wing-propeller
arrangements in producing both static lift and lift when
forward velocity exists". Consequently, the final
conclusion was, "The present device does not show sufficient
promise of military value to justify further development by the
Army Air Forces". An astounding conclusion! For instance, a
channelwing fighter with the 1500 HP P-51 engine could develop
1500 X 8.4 = 12,600 pounds static vertical lift. The max TOGW
for the Mustang was only 11,600 pounds. A bomber with four 1200
HP B-17 engines would have had 1200 X 4 X 8.4 = 40,320 pounds
static vertical lift. Max TOGW for the B-17 was 65,500 pounds. A
channelwing cargo plane with two 1200 HP C-47 engines would have
developed 20,160 pounds static lift. The max. TOGW for the C-47
was 28,000 pounds. Admittedly, there are complications with
using the Custer channel, but to say there would be no use to
the military is beyond reason! At about this point, it is easy
to sense the frustrations Mr. Custer must have felt.
USAAF test stand at Wright Field 1947. (USAAF technical Report
5568) This configuration demonstrated 13.8 lbs/HP static lift
with Custer channel. Note flaps used to further deflect channel
4. The augmented channel with deflector vanes
generated 13.8 pounds lift per horsepower.
The Custer channel with deflection vanes mounted behind the
prop was measured to have a vertical lift capability of 13.8
pounds per horsepower. This was recorded in 1945 at Wright
Patterson, and is documented in an AAF memorandum. It was noted
that the channel and deflectors gave a considerable nose down
pitching moment. This might change the design, but the extra
lift would seem well worth the effort.
5. Cruise speeds are largely unaffected by the
The Custer channel is a wing when unpowered, and
generates approximately 10% greater lift than a straight wing
the diameter of the channel. This was verified by the L.
H. Crook Laboratories report 557, in 1943. As speed increases,
the coefficient of lift increases dramatically, as shown in the
NACA report referenced in section 2 above. This allows less
straight wing, hence less drag. This area has not been
researched to a final conclusion, but it would seem logical that
an aircraft with less wetted area would have equal or less drag.
Harold Custer stated that at high speeds, the extreme
downwash behind the channels did tend to push the nose down.
To obtain high speeds without having to create negative lift
on the elevator, it would appear that the prop / channel
clearance would have to be changed in flight. Several fixes in
this area have already been shown effective.
6. The channelwing is cheaper to build and
maintain than normal STOL.
With modern methods and composites, the Custer channel is much
easier to build than the normal STOL.This is because the Custer
has no trailing or leading edge flaps and associated control
systems. The CCW-5 and other plans from that era show the
channel in the wing. This would be difficult and expensive.
However, if the channel is placed in front of the wing, it is
both simple and cheap. Notice the design for the P-20 Raider or
the P-50 Devastator in the Dreams and Schemes section for an
7. The Custer is far cheaper to build and maintain
than a helicopter.
While a Custer has never been designed to hover, it would not
be as difficult a task as it was to hover a helicopter. (my hat
is off to mr. Sikorsky) If we were to build such a machine, it
would have no moving parts, other than the prop, and flight
controls. In effect, it would be a,"solid state", helicopter.
Because of the simplicity, operating costs would be far lower,
and utilization would be far higher. Added to this, would be
speeds as much as double the speed of a helicopter. An example
would be to replace the V-22 Osprey tilt rotors with Custer
channels. Maximum takeoff gross weight for the Osprey is 47,500
pounds with two 6150 HP turbine engines. If we theorize we could
match the 1944 tests, at 8.4 pounds static lift per horsepower,
or the 1945 Wright Patterson tests that show 13.8 pounds static
lift per horsepower, with augmented channels, the comparison
V-22 Osprey as built // V-22 with Custer channels 8.4
#/hp lift (1) // V-22 augmented CusterChannels 13.8lbs/hp
engines // 2 - 6,150 shp // 2 - 6,150 shp // 2
- 6,150 shp
max vertical lift* // 47,500 lbs // 51,660 lbs //
max. TOGW with short run* // 60,000 lbs // 82,656 lbs (3)
// 115,866 lbs (3)
cruise speed // 240K // 300K+ (4) // 300K+( 4)
(1) USAAF Technical Report 5142 5 September 1942
(2) USAAF TSEAL-2-4586-3-2
(3) 1953 NACA Research Memorandum RM L53A09
(4) Product Development Group, Inc. report of 1988
* Vertical lift allowing for one engine out on takeoff
Because of the higher cruise speed with the same engine, range
would be increased 25% over the Osprey. In reality, because the
Custer version would have far more lift from the wing, range
would be even greater. Ferry range would be vastly improved, due
to the ability of the Custer version to lift more fuel on
takeoff. Notice the Custer version could lift an additional
22,656 pounds (1) or 55,866 pounds (2) of fuel or cargo.And
remember, this is done without tilt engines, folding props, and
the associated cost and maintenance.
Custer Channelwing: Dreams and
Developing a great idea.
Over the years,a number of attempts have been made to revive
the Custer. They are shown in this section to demonstrate newer
ideas, and improvements in Channelwing theory. Only ideas that
have been researched and designed by aerodynamic professionals
are included. These aircraft are capable of being built and
flown to design specifications, according to research data
mentioned earlier, and competent research conducted since.
Designs by Product Development Group
The P-20 Raider and the P-50 Devastator
Sometime in the late 1980s, an international consortium called
the Product Developement Group was formed to design and build
aircraft for the U.S. military. These designs incorperated up to
date technology and new innovation for the channelwing. After a
great deal of searching, I have found the full production
drawings for the two aircraft. It would appear that they were to
compete with the V-22 "Osprey", but the project fell through
with the death of the major backer. I have many pages of "demo"
drawings such as these, along with performance data, tooling
requirements, etc. These are truly impressive aircraft by any
standards. This performance coupled with the reliability would
seem to place them far ahead of the V-22 Osprey. Although they
could be designed for vertical takeoff, it would seem of little
use considering the slow approach and takeoff speeds. You will
note the channel is placed in front of the wing, with a stub
wing behind it. This gives a very strong structure ( +8, -4 G)
that would make the P-50 a real contender for close support or
high speed attack roles. Judge for yourself, keeping in mind
that these designs were the product of University studies,
competent engineering, and hundreds of hours of flight testing,
rather than the wishful dreams of a novice.
The Rhein Flugseubau Company RFV-1
In the 1960's, The channelwing was studied by Rhein Flugseubau
Company of Monchengladbach, West Germany, for the German
government. Their chief engineer, Hanno Fisher, was recognized
as one of the finest engineers in germany. He added several new
concepts to the Custer. On the RFV-1 he placed a single
channel on top of the aircraft, on centerline. This removed
the need to cross shaft the two engines, which drove a single
prop behind the channel, and erased any roll problem
experienced in case of an engine out on takeoff, or during low
speed operation. This was seen as a plus, even though a
single channel creates less static vertical lift. Fisher
also added an annular duct around the prop, which
transitioned into the Custer channel. This added low speed
thrust and channel lift, while helping solve the critical,
prop tip / channel clearance, at various power settings.
He also added an "oberflügel" in front of the duct, and in
the channel, which added extra static lift. Considerable
research was done at the University of Aachon, 20 miles from
their production facility along with 100 actual flight tests.
work on the RFV-1 progressed through these flight tests, but
apparently didn't survive fine tuning. According to one
knowledgeable source, Rhein Flugseubau decided to add extra
width to the fuselage, to gain passenger space. The added
structure interfered with the airflow into the channel,
reducing the efficiency, and eventually dooming the German
project, due to an austerity program by the government in the
60's. This caused a lack of funds to re-work the passenger area
and the project was dropped sometime in the late 60's, in order
to work on more profitable projects, and the aircraft was
parked. I have no knowledge of the disposition of the aircraft
at this time.
Channelwing innovations by Hanno Fisher.
1. Annular duct
3. Custer channel faired to annular duct.
4. Control surface flaps in prop wash for low speed control and
Hanno-Fischer Innovations --
Specifications for the Rhein Flugseubau RFV-1
Empty Wt. -- 3,740 lbs.
Useful load -- 1,573 (less fuel)
Fuel allowed -- 1,507 lbs.
MATOG -- 6,820 lbs
Power -- 2 x 250 HP Lycoming
Max. (two engine) -- 200 MPH
Max. (single engine) -- 154 MPH
Normal cruise -- 194 MPH
Climb: -- 2796 fpm
T.O. distance -- 161 ft.
T.O. over 50' obstacle -- 361 ft
Two Place Civil VSTOL
In the 1950's, The Custers started a design on an aircraft for
the light plane market. The main feature was 50' takeoff and
landing distance, along with 160 mph cruise. With modern design,
this could be improved, particularly in the cruise speed area.
Little more can be done with the takeoff distance, as you need
some forward speed to have enough airflow over the controls to
maintainstable flight. With modern composits, and latest
designs, it should cruise above 180 mph using a 160 HP
Two placed light plane envisioned by Willard Custer, held
by Harold "Curley" Custer.
Custer Channelwing: Data & Video
Speakers, including Harold Custer, are available for interested
groups for a modest fee, plus transportation. Lectures include
the latest in channelwing research, explanations of theory,
problems, and possibilities. Contact us if you are interested in
having a speaker for your group.
The following educational material on the Custer Channelwing is
available. For your convenience, a preformatted order form will
be e-mailed to you for printing. Simply complete and submit the
Channelwing video narrated by Curly Custer
Shows channelwing development and flights .. $25.50
NACA Research Memoranum RM L53A09
1954 NACA study with many pictures, graphs, etc. .. $24.00
Army Air Force Technichal Report 5142
1944 original AAF tests on 1/3 scale model graphs, pictures,
lift data ... $24.00
Army Air Force Technical Report 5568
1947 comparison of two channels with different chord ratio.
graphs, pictures, lift data ... $24.00
P-20 series STOL aircraft brochure
Four page sales brochure for P-20 ... $6.00
Custer Channelwing Corp. booklet
Company explanatory handout. pictures, explanations of the
theory ... $14.00
Power-on Channel Wing Aerodynamics
Edward Blick* and Vincent Homer
Reprint from Journal of Aircraft
* Ph.d. professor, college of Engineering, U. of Oklahama
Theory, diagrams, pictures, equations ... $14.00
More data available on request, for interested parties. Make
all checks or money orders payable to Harold Custer. Send all
7N329 Rt. 31
South Elgin, Ill 60177
Contact: [email protected]
Willard Custer Ideas, Ready for the
Engineers at the Georgia Institute of Technology Research
Institute in Atlanta, U.S.; are quietly researching the
possibility of applying a Channel Wing Configuration technology
to the designs of future aircrafts platforms. The Channel Wing
Configuration, when implemented on the wing design, would give
the aircraft the ability to generate a high volume of lift,
which could open the path to many design possibilities. The idea
of configurating the wing design to be able to generate more
lift has been around since the birth of aviation early in the
1900s. Preliminary studies by aviation engineers on the subject
in the mid-to-late 1910s resulted in experimentation with
various form of wing configurations and settings. Eventually,
advances on airframe deign and a premium on engine performance
took center stage, thus neglecting the concept of wing
modification to achieve greater lift. For years, research into a
greater lift-generating wing design was shaped by traditionalist
aviation engineers and designers. Thus, radical new ideas were
never fully pursued, that was until a brilliant Maryland
inventor came forward with a new concept in 1935.
Willard Custer was one of the first true champions of the lift
principal called aero physics. He stated that the amount of lift
generated by any aircraft is determined by the speed on which
the air flows over the wing, not solely on the speed of the wing
moving thru the air as articulated by many. If a wing
configuration were to be designed with deep channels, dropping
like a couple of "smiles" under the propellers; then the
aircraft would generate more lift with it than a conventional
wing configuration. For Custer, the idea was simple enough. An
aircraft can generate lift with zero forward speed utilizing the
engines to provide the necessary airflow to sustain the plane in
the air, thus achieving an impressive amount of Short Take-Off
and Landing (STOL) capability. Additionally, with the channel
wing shape, the engine thrust is propelled downward, providing
the aircraft with the ability to perform short take-offs, and,
as an added bonus; maintaining air control at relative slow
speeds. The idea that an aircraft can achieve virtually vertical
take-off and landing capabilities by re-designing its wing
configuration a was radical concept in the late 1930s.
Custer CCW-1 (photo, via author)
The aviation community did not think much of Cluster's Channel
Wing Concept. That's the risk someone takes when propelling
radical new ideas. Nevertheless, Custer marched on. In the
summer of 1943, his first aircraft design with channel wings was
demonstrated to the United States Army Air Forces in Maryland.
Immediately, the CCW-1, as the plane was designated, was a media
darling. Stories of this strange-looking aircraft, nearly
hovering over the ground, fascinated many in the country.
Unfortunately, the US Army was not one of them. They branded the
CCW-1 impractical because of the extreme nose-up attitude
requirement for landing. There were also issues about the
survival of the CCW-1 in combat. Test flights showed that if one
of the engines were to be lost, the pilot could not maintain
effective control over the aircraft.
Despite the setback, Custer persevered, and in the fall of
1959; he presented his new aircraft, the CCW-5 to the Marine
Corps. Again he was turned down. Mainly, for the reasons stated
in 1943; the concept seemed to hit the wall. No major research
was invested on the channel wing concept until 1995, when the
idea was resurrected by Dennis Bushnell, Chief Researcher at
NASA's Langley Research Center in Virginia. For years Bushnell
mulled over how to fit an aircraft in tight locations off the
ground. He researched the accepted principals of direct thrust
and rotary wings, but they were not able to produce the desired
results, as recent experimentation had shown. But Bushnell had
an ace. For years he had known and studied the work of Custer on
the Channel Wing; and he wondered if a combination of control
circulation, a method from which lift is generated utilizing jet
of air to improve the aerodynamic characteristics of the wing,
and the channel wing, could be the answer. Either of these
systems, by themselves, could not provide the aircraft with the
necessary characteristic he desired, but combining the two was
seen by Bushnell as the way forward.
Custer CCW-5 (photo, via author)
A new research program commenced in 1999 and lasted until 2004.
The research focused on the Coanda Effect, named after its
founder; Romania researcher Henri Coanda, who in August 1910
discovered that hot gases exiting a jet followed the contour of
the plates installed to deflect it. Circulation Control follows
a different path. Simply put it, circulation control works when
compressed air is directed over a curve or leading edge to
generate greater lift capacity. Researches believed that
circulation control, could in the future rend obsolete moving
surfaces on aircrafts. The next step in the developmental
process for Circulation Control is the replacement of mechanical
lift augmenters with air hoses to make the aircrafts lighter,
quieter and maintenance friendly. For all of this to take
effect, surface system needed to be introduced, and here is
where the Channel Wing Concept comes into play.
Current computing systems used to measure fluid dynamics of
aircrafts' surfaces had established the feasibility that a
Channel Wing configuration with enhance Circulation Control,
could produce a serviceable and stable super-STOL platform. Wind
tunnel testing and computer animations had confirmed the Channel
Wing design payoffs in ways that Custer could not have done in
his time. Custer clearly understood the airflow needed to
generate lift could come from two different sources: the engine
or the airframe forward motion. What he lacked was an
understanding of what happened to the air stream once it hit the
channel. The end result is turbulence. This is why both the
CCW-1 and 5 failed to achieve major air-control properties. At
low speeds, the flow of air is detached form the traveled
surface; leading to the aircraft to lose differential pressure
that is the cause of lift. At Custer's time, there was no method
accurate enough to calculate when this effect comes to play or
how to design an aircraft that used this effect in its
advantage. The solution: Circulation Control. One of the most
challenging arenas for the CCWs models were the high angle of
attack that the aircraft needed to be flown, a dangerous
proposition because the pilot will temporarily loss the ability
to see over the plane's nose. Another problem was the lost of an
engine. If the aircraft were to loss the use of one of its
engines, the aircraft will be subject to high stall degrees and
rolls, without the necessary energy to compensate for them.
Circulation Control can solve this problem.
At present, Bushnell and his team had been pressing for
sometime to design and aircraft that incorporates both concepts,
but like Custer before, without much success. Skeptics' rapidity
pointed out that all the research data done during the past five
decades had failed to produce a serviceable aircraft design,
thus leading them to the conclusion that the concepts are
incompatibles. The Channel Wing concept may need to wait until
advances in technology can undisputable show that an airworthy
aircraft can be achieve; but the Circulation Control concept is
already been use by various countries in the design on unmanned
air vehicles. Even a naval application was found for the
concept. Submarines could use jets instead of conventional dive
planes and rudders to change aspect ratios.
Custer's idea was years ahead of his time, and seems today,
that is still ahead of us. Further research and data collection
maybe needed, but with the current military situation, a premium
is been place on the ability of aircraft to perform short
take-off and landing s procedures, its only a matter of time
before the next great engineering breakthrough comes along and
Custer's idea will probably be at the center of it.
- Raul Colon
"Official Custer ChannelWing Website"
THEORY OF CHANNEL WING AIRCRAFT:
SPEED OF AIR
Willard R. Custer
Most headlines about aviation have been concerned with new
speed and altitude records. Today’s airplane, already faster
than yesterday’s, will fly still swifter tomorrow. Forgotten is
the fact that it is much more difficult to make planes fly
slower than it is to make them go faster. For example: Take a
medium-size stone in your hand, and throw it with all your might
up through the air for perhaps a hundred and fifty feet. As long
as the speed lasts it stays in the air; however, when the speed
is lost it falls quickly. Now take the same stone and try to
throw it the same distance in slow motion.
It is tougher to keep a heavy plane in the air in slow motion
also, yet retain its high-speed qualities. It requires a new
concept in wing design, making it an aircraft. This new design,
the CUSTER CHANNEL WING, is a versatile airfoil, and utilizes
the forces of “Aerophysics,” atmospheric pressure, gyroscopical
action, impact pressure, thrust, etc., to sustain flight at low
speeds, yet retains the very desirable features of low drag and
boundary layer control for high speeds.
The airplane industry has been handcuffed to the word
“airspeed” since the beginning of flight because airspeed was a
requirement before lift could be thought of. Perhaps the term
“Lift” needs a new cant for the directional “lifting force”
induced aerophysically when dynamic airflows are passed through
a channel wing. Accordingly, the induced pressure differential
can be manipulated by angle of attack of the wing and deflection
of Thrust velocities to provide a varying angle directionally
for the resultant force.
Airspeed vs. Speed of Air
To clarify, let us compare the principles of aerodynamic flight
with what I shall call aerophysics “flight”: Airspeed versus
speed of air. The importance of understanding velocity relations
and the directional forces and reactive forces associated with
aerodynamic flight cannot be overstressed. A better than average
knowledge is necessary before one can delve into the basis of
aerophysics “flight” which, in some cases, requires complete
rejection of aerodynamic results to obtain full conception.
Essentially, aerodynamic lift occurs due to differential
velocities of air above and below a wing. Direct Thrust forces
are used to propel the airplane only, and it is a secondary
effect of the resulting motion that provides any air movement
whatsoever to the air envelope. The only motion imparted to the
air surrounding a lifting airfoil is a disturbance resulting
from friction as the wing separates the air mass while being
“pushed” through it. The shape of the parting surfaces induces a
differential air velocity of low magnitude no matter what the
relative velocity of the moving airfoil. The salient belief that
the air is in motion, and with a high velocity, is the greatest
misconception of the aircraft industry.
It is necessary to understand that the air is static, and that
the airfoil alone moves, before the connotation of aerophysics
“lifting force” can be perceived and evaluated. For this reason,
applications of aerophysics “flight” have remained long in
infancy, and improvements in aerodynamic flight have been
contingent on power plant development.
Basically, for air to move there must be changes in pressure;
therefore, if air moves, there IS a change in pressure. The
higher the velocity of air in motion, the greater the pressure
differential becomes normal to a surface of a structure which
has this surface—only—exposed to the airflow. In correct
application of a channel wing the dynamic impact pressure of
high-velocity airflow is applied normal to the inside periphery
of the channel. (No “airfoil section” is needed.) At zero
forward velocity of the aircraft (relative to the stationary air
envelope or the ground) the resultant impact pressure
differential across the channel is greatest because of spread in
relative velocities (mass air in motion within the channel
relative to still air outside of the channel). This force—a
force that is not truly aerodynamic lift—is a collapsing force
on the channel and is counteracted by structural design and
strength to transmit the pressure as a directional force.
(Additional controls provide deflection of this force away from
the mean normal. When deflected as a “lifting force” the
differential pressures provide an “impossible” Lift when viewed
in an aerodynamic concept. For correct appreciation this force
must be viewed in the light of a hydraulic force upon a sealed,
evacuated half-cylinder submerged in a dense liquid.)…
The “Lifting Force”
In front of the propeller the available “Lifting force” is
determined by the mean airflow velocity (relative to the inner
surface). A correctly designed propeller provides high-velocity
air through a channel and a denser than normal air mass with
propulsive efficiency behind. However, some Thrust force will
always be exerted normal to the propeller disk. This is a
secondary effect of the power applied to “lift” rather than
“push” the aircraft, and it provides a satisfactory method for
directional control in the longitudinal plane by forming a
resultant force with a “Lifting force”.
The resultant of the “Lifting force” and the residual Thrust is
always of greater magnitude than the basic Thrust force of the
propeller when out of the channel. At low aircraft velocities
(relative to the ground or the air envelope) the resultant is in
excess of 45 degrees. Thus a “hovering” condition can be reached
when the resultant force equals the weight, the power applied
being proportionate to the rate of Thrust dissipation. In this
condition, with other forces remaining constant, rearward flight
results whether from an increase of power or an increase in
angle of attack.
As the angle of attack decreases the resultant force vectors
toward Thrust augmentation and the supporting forces require
redistribution to remain aloft. This can be accomplished by the
addition of power with a Thrust dissipation or, more simply with
existing developments, the establishing of aerodynamic Lift on a
supplementing wing or a channel wing with an airfoil section.
With the channel wing the “Lifting force” of the velocity
differentials across the channel approach “unity” as aerodynamic
flight is approached…
The aerodynamic Lift coefficient of a channel wing is always
positive with some negative angle of attack. This means that …
there is a negative pressure at the advancing edge which tends
to nullify the normal compression of air at this point. Thus a
velocity increase is permitted for a power application that is
not normal in aerodynamic flight… Terminal velocity is not a
function of “sonic compressibility” due to this laminar air
being displaced during a velocity relationship above “unity”
across the channel. Acceleration, however, will vary according
to the rate of improvement in propulsive efficiency—once again
contingent on propeller design.
It is comprehensible that only a small percentage of the power
used in aerodynamic flight is converted into the work of
“Lifting”. The greatest percentage of power is used to overcome
the “drag” caused by the airplane velocity. On account of its
“push” application power must be increased on an exponential
curve of the “cube” to provide aerodynamic flight of doubled
velocities. The power requirements curve, which does not follow
the aerodynamic “cube” scale (in the channel wing), is also
flattened toward a straight-line increase because of the unique
Lift coefficient of the wing. The “overpower” requirement at the
compressibility area is thus also avoided, or rather obviated,
for velocity calculation in design in the Channel Wing…
Summarizing, although both applications are affected by the
same basic laws, aerodynamic and aerophysics principles follow
different paths to attainment of “flight” results that are
In aerodynamic flight, power is applied exponentially to
produce singular velocity relationship of the airfoil to the
stationary air. Any deflection from the plane of optimum flight
reduces the supporting forces. High velocity is restricted by
air compressibility, low velocity is hampered by the
relationships required to obtain Lift and “Hovering” requires a
Thrust force of greater magnitude than the weight. Best “flight”
results are obtained by power plants providing maximum mass
acceleration of a Thrust force with low propulsive efficiency in
conditions of greatest fuel consumption (high energy less
In aerophysics, Channel Wing omni-directional “flight”
velocities are of fourth-dimensional relationship; associated
with aircraft attitude, Thrust dissipation, and amount of energy
invoked. Power is applied to “induce” the major supporting force
and to provide deflective forces which augment either the
supporting force or the propulsive force; thus increased power
requirements approach the straight-line curve. Low relative
velocity of the aircraft is possible and “hovering” occurs for
weights much greater than Thrust force by conversion of
“horsepower” energy. Velocity “spread” is contingent upon
induced vacuum as attained by an efficient propeller, efficient
power plants of reliable design to “flight” power requirements.
Popular Mechanics, May 1947
The Wing that Fooled the Experts
Front cover With stubby wings like half barrels flanking its
fuselage, it resembles nothing you've ever seen in the air
before--but it flies! Some experts who've seen it still can't
And before you recover from the shock of watching this strange
craft perform in the sky, its inventor, Willard R. Custer, has
many more surprising claims to make.
He says planes of this design can: take off and land at 15
miles an hour in 50 feet of space, lift twice the pay load of
present transports using the same power, and either hover
overhead or pierce the supersonic speed zone. And he supports
his claims with results from Army and privately conducted tests.
Heart of this remarkable performance is the Custer Channel
Wing, adaptable to any type air craft, and to piston, jet or
rocket power. In contrast to conventional airfoil, it is shaped
like the lower half of a tube and has an adjustable-pitch
propeller at the rear. The propeller's tips sweep almost the
entire trailing edge.
Simply stated, an ordinary wing is fashioned so that air
pressure--or lift--is built up beneath it and decreased above
it, as the plane speeds along the runway, until a slight
difference between the two makes the plane fly. Atmospheric
pressure on all sides of the wing is a constant 14.7 pounds per
square inch at sea level when the plane is standing perfectly
still. Modern transports like United Air Lines' DC-6s reach 125
miles an hour and use over 3800 feet to get off the ground, and
they touch down at 100 miles an hour, using more than 1800 feet
Custer's method is to relieve the pressure above the wing by
the propeller's sucking action at the rear of the channel, and
let the undisturbed pressure of nearly a ton per square foot
below the wing lift the plane off the ground. Instead of moving
the plane to achieve lift he gains it by moving air masses
through the channel. To better control the lower- pressure air
thus created within the wing, he has built up its sides above
the propeller's center and slightly constricted the forward
This funnels more air to the propeller, increasing its
efficiency so much that the pitch has to be adjusted to absorb
unused horsepower, Custer says. In contrast to this concentrated
lift, a conventional wing's efficiency varies along its entire
length. Since the channel will lift 75 percent of the plane's
weight without forward motion, very little movement is needed to
add the 25 percent that makes the craft airborne.
His unorthodox attack on the problem of finding more efficient
ways to fly can probably be laid to the fact that Custer is an
auto mechanic by trade, not an aeronautical engineer. So he cut
a simple and unique path to his solution. He uses familiar
objects instead of complex technical formulas to illustrate his
Although the invention has many important potentialities if
completely successful, Custer prefers to stress the safety
features that will further popularize flying. With it, he points
out, planes can sink slowly to earth through the worst weather,
onto deep snow or any other kind of unprepared ground, at
landing speeds slower than a man runs. The crash hazards of
blind approaches and landings at high speeds are eliminated.
Another important item is that all the controls of the channel
wing plane are conventional and it is easy to fly.
While most experiments with the wing have been in wind tunnels
and laboratories, a test ship flew more than 100 hours over a
government field at Beltsville, Md., using 75-horsepower
engines. The engines and propellers were mounted on metal
supports extending across the upper rear section of each
During those first test flights the plane, weighing only 1785
pounds with pilot, was held to a top speed of 60 miles an hour.
Take-offs and landings were made under 100 feet, upwind,
downwind and crosswind.
Two series of nonflying tests were later made by the Air
Materiel Command at Wright Field. One fact established during
trials was that the wing's lift increased even when its chord,
the distance from front to rear, was reduced by half. Frank D.
Kelley, president of the National Aircraft Corporation, said
that a second flying model using a short channel wing will be
ready for flight tests this fall at Hagerstown, Md.
Photo 1 Caption
Inventor Custer points to the channel wing design he says makes
possible safe 15-mile-an-hour take-offs and landings and doubles
planes' lifting capacity. Test version, below, has flown more
than 100 hours.
Photo 2 Caption
Front view of experimental plane with 75-horsepower engines. It
made take-offs and landings in less than 100 feet--upwind,
Photo 3 Caption
Readying six-foot channel for plywood skin prior to full-scale
laboratory trials to determine lift, drag, thrust and pitching
moment. Below, rear view of test model flown during Maryland
trials. Despite its unusual appearance, the plane's controls are
conventional and it is easy to fly.
Photos 4&5 Caption
Supersonic speeds are expected from the channel wing on fighter
Inventor's rough sketches showing adaptations of channel wing.
One has large propeller sweeping rear of channel formed by
section joining twin-fuselage fighter. Transport design uses
double channels on both sides.
Air Progress Magazine, October/November, 1964
Custer's Production Model Takes Bow
[ Original Article with Photos : Page 1 ... Page 2 ]
As a fitting example of independent, persevering aeronautical
pioneering, on July 4th, the first production model of the
5-place Custer Channel Wing CCW-5 was rolled out of the
company's hangar at the Hagerstown, Md., Municipal Airport,
christened and demonstrated. Although this gleaming all-metal
aircraft is entirely new, its design is not. Designer Willard R.
Custer has spent 25 years in full-time research and development
of the vertical lift concept embodied in the half-moon wing
configuration. Single and twin engine scale models were built
and test flown from 1927 through 1940; these were followed by
test flights of single and twin engine aircraft from 1943 to
1952. The prototype twin-engine all metal CCW-5 (photo, upper
left), flying since 1953, took off in less than 100 feet on its
maiden flight, and the company reports that it has hovered at
from zero to 11-mph, although the craft is actually a STOL type.
(Remaining six photos here are of the latest Custer).
The Custer Channel Wing is unconventional in that its wing
channels replace such conventional aerodynamic devices as wing
flaps and slots to provide required lift. In the Custer system,
the powerplant is suspended in the channel with the propeller at
the trailing edge. By drawing the air through the semi-circular
wing at high velocities, pressures over the curved lifting
surface are decreased to a much greater degree than in
conventional wing configurations.
Actually, the Custer utilizes a Boundary Layer Control effect
to create its greater lift. According to Custer, conventional
aircraft obtain lift coefficients of approximately 3 through the
use of high lift devices, whereas the CCW craft have
demonstrated power-on lift coefficients of 5. Designer Custer
further claims 100% greater lift capabilities for his, using
standard airfoil section, propellers and engines, compared to
other aircraft using similar airfoil/power combination.
Depending upon climatic conditions and weight, the Channel Wing
configuration permits the craft to take-off in distances ranging
from a 75 foot ground roll, down to virtually zero length,
although high temperatures (95 degrees) will boost the run to
The CCW-5 is conventional all-metal aircraft construction,
except for the compound curve section of the channels, which are
of fiberglass. The channels, a web and rib structure using a
main and secondary spar, are made of aluminum. Each spar has two
sets of cap angles. The ribs tie into the cap angles and web.
The metal and fiberglass channel skins are wrapped around the
ribs. The engines are suspended in the channels by tubular steel
frameworks with the frames mounted to the wing spars and two
pylons supporting each engine.
The Custer Channel uses a 4418 airfoil section with a 6 foot
chord. Each propeller, located slightly aft of channel trailing
edge, is a thin tipped 7 foot diameter full-feathering Hartzell
with its tips trimmed. In rotation the tips deflect forward to
close the channel; at 2625-rpm air is sucked through the channel
over the airfoil at 115-mph to provide the BLC and lift effect.
Designer Custer reports that about 15 years work went into his
The prototype CCW-5 utilized a redesigned Baumann "Brigadier"
airframe; the production prototype was constructed from the same
jigs. Later craft incorporates only a few changes, mainly
fairings at the channel roots and configuration change of the
wing trailing edge, plus latest version of the 260-hp
Continental engines which were moved slightly aft. The CCW
Corporation reports initial orders for 40 CCW-5's, 20 of which
are slated for a West Coast distributor to be sold as executive
and utility craft. FAA Type Certificate is expected by July
1965, at which time 6 to 8 craft should be well along on a
production line. Price is $65,000 with standard equipment.
CUSTER CHANNEL WING CCW-5
Wing Span 41 ft. 2 in.
Length 28 ft. 8.5 in.
Height (at rudder) 10 ft. 10 in.
Height (at cabin) 6 ft. 9 in.
Cabin interior height 4 ft. 5 in.
Cabin interior width 4 ft. 5 in.
Empty weight 3675 lbs.
Gross weight 5400 lbs.
Top speed 200 mph.
Cruise speed 180 mph.
Minimum sustained level flight 35 mph.
Take-off run 50 to 250 ft.
Landing run 300 ft. (approx.)
Ceiling 22,000 ft.
Accommodation 5 seats
Engines--two 2600 hp. Continental O-470 with fuel injection
Airpower Magazine, Volume 7 No. 3 May, 1977,
The Custer Channel Wing Story
After forty years of pioneering a new concept of lift, Willard
R. Custer's ideas are now in the forefront of STOL aviation!
Author's Note: The Custer Channel Wing concept, after almost 40
years of public attention, is still so controversial that this
article must be written in two sections. The first covers
development of the idea from 1927 to its present (and possibly
future) status. The second presents the engineering opinions,
pro and con, which have fueled the controversy. Both are
included in this article.
Willard R. Custer is a man straight out of American folklore.
He is the prototype Yankee inventor, smart, tough, resourceful,
unafraid of the machinations of big government, big business, or
fate. He believes in his invention, and he knows that sooner or
later he will prevail.
Custer is a genuine innovator, for although designers have
tried virtually every combination of wing size and shape, and
every permutation of engine/propeller placement, it remained for
him to combine the two elements in his famous Custer Channel
The inventor is a friendly, persuasive, energetic,
single-minded man who has pursued a fifty year dream with a
charming tenacity that has weathered many disappointments. He is
convinced that he is right, that he has been right, and that the
world has been denied the benefits of the Custer Channel Wing by
a combination of misfortune, lack of vision, and, sadly in some
instances, simple bad faith.
It's hard to be around Willard Custer for more than a few
minutes before beginning to believe with him in his concept. He
is so evidently sincere, so deeply convinced and so determined
to win through that he converts even the most ardent
skeptics--at least temporarily. He's an exhausting man to
interview, for he flies around his small but neat
shop/laboratory, grabbing an air hose to "fly" a screwdriver
around to prove a point, picking up a 40 year old model, shoving
a jet engine test rig out of the way, all the time telling you
of the fundamental simplicity of his own
science--"aerophysics"--and never tiring of the main topic of
conversation, his patented channel wing.
In some respects these engaging characteristics may work
against him in today's cold business climate. Times have
changed, and aviation is more sophisticated, more finance
oriented, and its engineering requirements are vastly more
extensive. It may be that the very qualities which have
sustained him in his battle against convention are not the ones
which persuade a modern businessman to put up the necessary
No matter; he has succeeded in his own mind, and the minds of
many qualified engineers. More important, there are current
indications that his concepts may yet be recognized.
The Custer saga began during a near-hurricane. A young Willard
R. Custer was taking shelter in a barn, when suddenly the roof
sailed off. Instead of being frightened, Custer wondered where
the power to lift the roof came from. He'd been fascinated with
aircraft for a long time, and knew that they had to accelerate
down a runway to generate enough lift to take-off, while the
barn roof, a poor airfoil, had lifted off "just sitting there".
Custer soon came upon a distinction that has eluded other
inventors, postulating that when air passes over an object, as
in the barn roof incident, that is "speed of air". However, when
an aircraft flies through the air, that is "air speed".
Now don't just put this off as semantics, not just yet, anyway.
It is Custer's theory, and it is best to use his words to
describe it, i.e., "it is the speed of the air and not the speed
of the object which counts. The airfoil was designed to obtain a
reaction from the air mass through which it is moved. The
[channel wing] was designed to obtain a reaction from the air
mass moved through it."
In essence, Custer says that in his wing, the movement of air
through the channels reduces air pressure in those channels, and
that atmospheric pressure on the bottom of the channel airfoil
then creates lift in direct proportion to the pressure reduction
in pounds per square foot.
Other engineers have said it better for Custer, but his
comments are essentially correct. In stricter terms, an engineer
might say the following: "Greater lift coefficients are obtained
by the effect of the propeller slipstream deflecting the air
mass through which the wings are moved, by suppression of the
boundary layer with high velocity slipstream, allowing the wing
to fly at higher angles of attack before stalling, and by the
vertical lift component due to the inclination of propeller
thrust at high angles of attack."
I like Custer's way better.
Speaking again in more practical terms, the pusher propeller,
tailored closely to almost touch the lip of the channel (Custer
has experienced "channel shaving" in some of his experimental
propellers) sucks air through the big half-tubes, and the
Bernoulli principle does the rest. The propeller/channel
juncture is critical, and some of the most efficient results
have been obtained when the juncture was temporarily sealed by
pouring water into the channel.
All of the above might sound like double talk, if Custer had
not demonstrated his principle in several models and four full
size aircraft, or if extensive testing had unquestionably
refuted his claims.
Unfortunately, the testing has been ambiguous, and while it has
not refuted the claims, neither has it sustained them. Part of
the problem is that the tests have not been designed to prove or
disprove the ability of the channel wing aircraft to perform,
but have only investigated certain aspects which were amenable
to contemporary testing techniques. Another part of the problem
is that Custer has understandably maintained such tight control
over his patents that full development programs have not been
There are additional factors which will be covered in the
second part of the article, which are more difficult to define.
On the one hand, government reaction to Custer through the years
has ranged from mildly patronizing to outraged; it is fairly
evident from the correspondence that Willard Custer's native
engineering talents didn't receive the same respect that would
have been accorded an established manufacturer.
On the other hand, Custer has probably been too optimistic
about the potential of his invention, while soft-pedaling some
of the real problem areas. His reluctance to let professional
engineers tell his story in objective, conventional, engineering
terms, clearly delineating both advantages and disadvantages,
has undoubtedly cost him some credibility.
As we shall see, engineering opinion is still divided about the
Custer Channel Wing, but that doesn't inhibit an examination of
the airplanes themselves, which constitute a unique line in
Custer had translated his hurricane/barn roof idea into a
working model by 1928, and obtained his first patent by 1929. At
about the same time he coined the word "aerophysics" to use
instead of "aerodynamics" to emphasize his concept that it was
the removal of air pressure above the channel rather than the
movement of the airfoil that was important.
In 1937 Custer built a single engine model which demonstrated
vertical lift, and by 1939 he had formed a corporation, the
first of many business entities which would sustain the idea
over the years.
In 1940 he was able to demonstrate a twin engine model to his
potential stock holders, obtaining enough financing to begin
construction of the first full size channel wing, the CCW-1.
This aircraft was (is, actually, for it is now in storage at
the National Air and Space Museum's Silver Hill facility) an
amazing combination of futuristic lines and modest test
expectations. Custer is an excellent woodworker, and the
streamlined fuselage and carefully built channels are beautiful
examples of art.
The airplane had 202.5 square feet of wing area, spread over a
most unusual surface. Semi-circular, detachable wing tips were
added to a straight chord wing surface, for a total length of
nine feet four inches; a six foot channel was attached between
the outer panel and the fuselage. Total wing span was 32' 10-1/2
". The channels were hung under the wing like huge nacelles, and
the two Lycoming 75 horsepower four cylinder air cooled engines
were mounted about midway in the channel. The 19'11" egg shaped
fuselage seemed disproportionately short, and the aesthetics
were not helped by a low aspect ration "T" tail empennage. The
stumpy landing gear looked like an afterthought.
Yet, despite this unusual appearance, the CCW-1 logged more
than 300 hours in a restricted test program aimed primarily at
learning how the channels worked, and what new flying techniques
were required to guide the channel wing.
Curiously, Willard Custer made the first flight in the aircraft
on November 12, 1942, entirely inadvertently. Custer is not a
pilot and his test pilot, E.Kenneth Jaquieth was not in town on
the day financial backers came to the Custer laboratory to see
the airplane. The backers wanted to take some pictures, and
asked Custer to taxi the CCW-1 to the small field where Jaquieth
had been conducting taxi tests.
The field was only about 200 feet away, up a slight hill, and
Custer felt qualified to move the plane. He applied power to get
up the incline, and was somewhat disconcerted to see the trees
on the horizon disappear. He was airborne, a non-pilot on the
first flight of a brand new kind of aircraft. Custer throttled
back abruptly, and the CCW-1 settled in, bending the landing
Instead of being upset, the backers were delighted, for their
dark horse had flown, albeit briefly.
The CCW-1 was later flown in demonstration for the military at
the Beltsville, Maryland airport, where gruff, hard boiled
Brigadier General W.E. Gilmore was excited enough to authorize a
test program. Custer recalls that Gilmore actually phoned
Orville Wright, who was in Washington at the time, urging him to
come out and witness the new development.
The testing program which began on June 6, 1944 proved to be
typical of the entire series of government tests of the channel
wing concept, generating mixed conclusions, controversy and zero
satisfaction to anyone.
The report, Army Air Forces Technical Report No. 5142,
concluded that the lift generated by the channel was similar to
the increment of lift generated by normal slipstream velocity in
conventional wing/propeller arrangements. Unfortunately, the
conclusion did not restate the point made in the report that the
channel generated more lift than the conventional arrangement.
The report stated further that while the channel was markedly
inferior to the helicopter in producing static lift, it was
superior to conventional wing-propeller arrangements in this
regard, and in producing lift with forward velocity.
The final conclusion was startling, for the report stated that
"the device does not show sufficient promise of military value
to justify further development by the Army Air Forces." This is
surprising in that the improvement of static lift over a
conventional wing/propeller arrangement was marked, and high
lift devices in the form of flaps, slats and slots were under
intensive development at the time.
The verdict typified the testing process, which never validated
or denied what should have been the basic question, i.e. "Did
the channel wing have characteristics of sufficient value when
compared to other high-lift devices to warrant further
Instead, the government denied that the channel wing was as
good as a helicopter, and Custer compounded the problem by
insinuating that it was indeed as useful as a helicopter.
Custer made an immediate and lasting interpretation of the test
results, one which did not help the controversy. He inferred
that perhaps the results were too good, and might actually be a
threat to the then infant helicopter industry, which was
receiving tremendous backing from the armed services. The first
reaction to a charge like this is to shrug -- it sounds too much
like the stories of "60 miles per gallon carburetors" which have
allegedly been suppressed over the years. Yet there are other
arguments which makes (sic) one wonder.
The most cogent of these are the later Wright Field tests,
conducted in 1945 and 1947, which were cautiously optimistic
about the channel wing concept. There is also the testimony of
the Wright Field engineer, Don Young, who conducted the tests.
He testified in a Washington Federal District Court that the
channel wing was in fact capable of direct lift, and largely as
a result of his testimony, Custer was granted additional
These optimistic test interpretations are reinforced by results
from independent engineering firms which have investigated the
theory and found it to have merit.
The brutal truth is that the government would have rendered
itself and Custer an inestimable service if it could have
structured tests which would have proved, beyond any shadow of a
doubt, that the channel wing was either (A) absolutely worthless
or (B) had some promise. In case A, Custer could have turned to
other things, while in case B, he might have been able to obtain
Custer persisted, however, and built an engineering test
vehicle, the CCW-2, in response to some of Young's test report
recommendations. The CCW-2 used a Taylorcraft fuselage and
empennage, with six foot channels in lieu of wings. Conforming
to Young's suggestions, the channels were shorter, and the
propeller was placed at the extreme rear edge of the surface.
The CCW-2 was extensively tested in free flight, tethered, and
in the NACA wind tunnel.
Harold R. Custer, Willard's son, made the first flight of the
new aircraft on July 3, 1948. Harold is as energetic and
determined as his father, and has the unique distinction of
having logged more channel wing time than any other pilot --
over 1,000 hours.
Harold totaled 100 hours flying time in the CCW-2, which could
take off in 45 to 65 feet, and land in the same distance.
Lateral control of the wingless vehicle was obtained by
differential use of the throttles.
The flight test program was not without its highlights. The
C.A.A. insisted that wings be mounted on the test bed, and short
stubs were attached. These neither helped nor hindered the
aircraft, according to Custer, but did mollify the bureaucracy.
More significantly, the aircraft demonstrated vertical lift in
zero wind conditions, while tethered.
The wind tunnel tests at Langley are another subject of
controversy and conjecture. Apparently the tests were well
conducted, and the results reported accurately. Unfortunately,
the conclusions seem to have been erroneous. Critics of the
report indicate that it stated the primary increment of lift
came from thrust, rather than from the lift, due to increased
velocity in the channels, even though the plotted data showed
this not to be the case. Also, the conclusions dealt with the
CCW's hovering characteristics, rather than its STOL (short
takeoff and landing) characteristics, an unfortunate view, for
the channel wing was clearly designed for STOL, not helicopter
During the whole arduous process of design and testing, one
thing definitely was not happening. Willard R. Custer was not
making a lot of money; he has invested not only all of his own
funds, but all of his life in the channel wing, and the material
rewards to date have been small.
But Custer's tenacity and his almost messianic belief in his
program enabled him to secure backing from a series of
investors, and in 1951 he was able to employ the Baumann
Aircraft Corporation to modify a Baumann Brigadier to the CCW
The result was the prototype CCW-5, a rather handsome five
place aircraft using two Continental 0-470-A engines of 225
horsepower each. The standard Baumann fuselage and empennage
were retained, and a 41 foot span wing with two seven foot wide
channels was attached.
First flight of the CCW-5 took place on July 13, 1953, with
Walker J. Davidson at the controls. The company then spent the
next several years performing demonstrations for civil and
military audiences. These demonstrations consisted largely of
maximum performance take-offs, with a terrifyingly steep climb
out, using steep turns to the right and left at a high angle of
attack and at speeds well below the stall speed of conventional
aircraft to impress the audience. Motion pictures of the
demonstration are heart-stopping; you know that an aircraft in
that attitude, at that altitude and airspeed, is going to crash
-- but the CCW just keeps on turning.
Slow flight was a specialty, with speeds as low as 22 mph being
measured. On August 27, 1954, in a remarkable display of the
channel wing's ability, the CCW-5 was actually hovered against
an 11 mph wind. The demonstrations would then conclude with a
steep approach, an incredibly short landing, with the turnoff
not at the first intersection, but at the approach end of the
Take-off technique is critical in the CCW-5, for the pilot has
to rotate sharply after the first 100 feet of roll, so that the
high velocity air stream passing through the channel does not
bounce off the runway against the undersurface of the horizontal
stabilizer. If this happens, the nose can be forced down,
considerably lengthening the take-off roll.
The prototype CCW-5 had an empty weight of 3,000 pounds, a
gross of 4,925 and a maximum gross of 5,400 pounds. Maximum
speed was quoted at 200 mph, with a 180 mph cruise; these were
probably optimistic figures, but most manufacturers tend to err
on the bright side.
The performance of the aircraft was good enough to interest
several firms in its manufacture, the most promising being the
Custer Channel Wing (Canada) Ltd., which secured the rights from
the Custer-Frazer Corporation, and planned to build the plane in
association with Norduyn Aircraft, Ltd. The short field
capability of the CCW-5 was attractive to bush pilots, and an
initial production run of 100 was planned.
This operation, like several other potential manufacturing
plans, fell through, primarily because of problems synchronizing
FAA approval of the aircraft, and SEC approval of the financing.
Custer undertook to build the 1st production CCW-5 at
Hagerstown, Maryland, using Baumann Brigadier drawings as a
start point, but modifying these as necessary to suit the
different construction and stress requirements of the channel
wing configuration. This, the fourth channel wing airplane, was
rolled out on July 4, 1964, and it appeared that after many
years of struggle the Custer Channel Wing was finally going to
Fate intervened in the form of the Securities Exchange
Commission, which took exception to the manner in which the
corporation stock had been issued, and the rug was pulled out
from under the venture.
The production aircraft was outwardly similar to the prototype,
but building it from the ground up rather than converting
existing Baumann components permitted considerably reducing
drag. There were external differences, too; the engine nacelles
were more streamlined, the wing had 2" less span, the nacelle
strut bracing was simplified, the ailerons were moved further
outboard on the wings, and rudder and aileron travel were
In the steps toward FAA certification, it became necessary to
raise the position of the horizontal surfaces, which impaired
STOL performance. As previously noted, the CCW pilot has to get
the horizontal surfaces below the vectored slipstream as soon as
possible for optimum results.
Performance figures for the production CCW-5 varied little from
the prototype, with the same top end performance indicated.
Power on stalling sped was listed as 22 mph, initial rate of
climb, 1,600 feet per minute, with a 22,000 foot service
A single engine service ceiling of 5,000 feet was obviously a
limiting factor in performance, and Custer planned to remedy
this with improvements in the propeller/channel trailing edge
juncture. A moveable sleeve has been suggested, one which would
maintain a tight seal at the channel for STOL work, but which
could retract for higher cruise performance.
And so engineering rears its ugly head again -- if the
advantages of the CCW derive from the channel section with its
relative lack of mechanical complexity compared to slots and
flaps, what does the addition of a moveable surface do? This and
other similar questions will be covered in the engineering
Where does the program stand now? It still has its advocates,
and there have been proposals for putting channels on everything
from Curtiss C-46s to Fairchild F-27s, to executive jets. As
indicated above, there is a real possibility that full scale
testing of the production aircraft by a government agency will
take place. But whatever occurs, you can be sure that Willard R.
Custer will not stop fighting the good fight, nor will he be
convinced that his discovery is not, still, the coming thing.
The engineering side -- pros and cons.
It is surprising to discover how sensitive and controversial
the Channel Wing is today, after all its many years in the
public eye. In an attempt to determine what the real merit of
the Channel Wing is, I went to several engineers who had taken
part in the testing and development of the device over the
years. I interviewed participants of three different kinds.
First were government representatives who had either conducted
or interpreted the official tests. Second were engineers who had
seen development potential in the channel wing, and who had a
positive interest in its commercial success. Third was an
engineer who was completely objective having neither a financial
or government interest.
Because of the controversy, and because some legal action may
still be forthcoming, I've been asked not to name the
The Pro Side: Proponents of the Custer Channel Wing say
that first of all, the true capabilities of the idea were not
susceptible to test because appropriate engineering theory had
not yet been developed to explain its demonstrated performance.
In other words, the airplane was physically doing things for
which aerodynamic formulas had not yet been evolved at the time
of testing. Such formulas have since been developed, and further
testing could be done in a more scientific manner.
Because adequate theory was lacking, the CCW was tested for
features and modes of flight which were in reality peripheral to
its main intent, simply because there was theory to test these
The pro-channel wing faction maintains that the channel wing's
pusher propeller arrangement not only forces circulation over
the airfoil, taking advantage of the Bernoulli principle, but it
also minimizes loss due to reverse flow around the trailing
edge, and then across the ventral surface.
The NACA tests, they say, erroneously failed to reveal that the
principal source of lift is due to the increased velocity in the
channels, and further, the tests should have highlighted the
findings that the static lift exceeded the weight of the test
vehicle. This basic fact underlies the most salient advantages
of the Custer Channel Wing, its simplicity, and its ability to
operate in the STOL mode without expensive high lift devices.
The initial cost of the Custer Channel Wing aircraft -- of any
type -- is probably slightly greater than a conventional
aircraft of the same type, but less than that of an equivalent
STOL aircraft using flaps, slots, etc. Its advantage stems from
the fact that its operating costs would be far lower than those
of the STOL alternative, because of the lessened maintenance
requirements. Its greatest advantage, however, lies where it can
be employed in helicopter operations, i.e. those situations in
which helicopters are presently used, but with suitable
modification in technique or landing area, a CCW could be
substituted. The reason for this, of course, is the horrendously
high cost of helicopter operation.
The CCW could be used for larger aircraft, and with jet
engines. The increased drag of the channel wing undoubtedly
would reduce top speeds, but in the correct application this
would not be significant. "Correct application" includes STOL
airline transportation over relatively short stage distances;
use in remote areas where the ability to fly in and out of short
airstrips with heavy loads is important, or where revenues
dictate against the use of a high operating cost vehicle.
In sum, the CCW proponents say that the concept provides an
economic, efficient solution for certain applications that are
not presently met by conventional STOL aircraft or helicopters.
The Con Side: Those who take a negative view of the
Custer Channel Wing do so persuasively and authoritatively.
Their initial reaction is usually: "Oh no, not the channel wing
again," followed by a measured explanation that there is not and
never has been any emotional or economic prejudice against
Custer's ideas. If, they say, the Channel Wing had shown merit,
it would have been in the national interest to develop it, and
it would have been so developed.
In more specific terms, the opponents of the Channel Wing say
that it has less potential now than when it was first evaluated
and found to be inferior to other approaches to STOL
performance. The anti-Custer group concedes that the channel
wing's lack of flaps results in a simpler, more economic system,
and that the generation of high lift by the configuration is
accompanied by a low pitching moment compared to conventional
The virtually viceless stall of the CCW is also acknowledged,
as is the fact that ground personnel are exposed to less hazard
by the propellers shielded by channels than by conventional
These are about the only advantages, however, and there are
more than off-setting drawbacks, which the anti-CCW view lists
(a) Higher aerodynamic drag due to the increased wetted area
resulting from the channels and, particularly, from the juncture
of the channel to the fuselage.
(b) The very simplicity of the CCW arrangement prevents
changing the wing aerodynamic characteristics to different modes
of flight. In other words, the thicker CCW is optimized for one
mode, while a wing with a well designed flap can be optimized
for both slow flight, cruise and higher speeds.
(c) The requirement for cross shafting of the engines to avoid
uncontrollable assymetric forces in the event of an engine
failure when at a low airspeed, high angle of attack flight
condition. (Custer concedes that production aircraft will be
cross shafted.) This requirement of course dilutes the CCW claim
(d) A substantially lower power-off lift coefficient compared
to a conventional wing/flap arrangement. This means a higher
landing speed is required for a power-off landing. (Custer
counters this with statistics on the relative infrequency of
power off landings in any multi-engine aircraft.)
(e) Increased structural complexity of the wing.
(f) A lack of engineering investigation into the problems of
stability and control, and the use of rather simplified
engineering theory to explain the complex CCW aerodynamics.
(This, as noted above, is a two-edged sword.)
So there you have it. Two points of view, both basically
honest, and both vulnerable to the assertion that the CCW has
not been amenable to really valid testing.
What is the answer? Perhaps the future will tell, for more
testing, presumably this time with more sophisticated
techniques, is in the offing. It will be interesting to see,
after all these years, who is really right.