"Naw, you don't wan' no three-blade Ivo on that bird cause you kin get
goin' a lot faster with a two-blader..."
"...Yeah, and not only that, the two-blades is more e-fishent cuz they
dawn havta go runnin' thru no turblin air from the blade right in front
of 'em."
So I left. Now I want to know whether or not I'm going to be wasting my
time putting a 3-blade Ivo or Warp drive prop on my Challenger, and why.
Anybody out there a rockit signtest?
(NOTE: Good answers may get published on UUCP, the Unofficial,
Unauthorized Challenger Page. Names will not be used unless permission
is given in your submission.)
-Jon-
Bruce A. Frank
What do the rockit signtists say if you run a six-bladed Warp Drive
prop? More efishent or quiter?
They came out with 2 bladed razor and they worked great.
So I vote 2 blades better than 1 blade....cuts my beard great!
Havent found a 3 bladed one yet.
So I vote 2 blades better.
U can use this in the UUCP.
Email for fast reply.
vic...@pacific.net.sg
http://home.pacific.net.sg/~vickir/
|\/\/\/|
| |
| |
| (o)(o)
C _)
| ,___|
| \
\___\
Having noticed a German firm, whose name I plainly have forgotten, that
sold fourbladed (fixed pitch) propellors for Cubs and other aircraft
I queried an ex-aircraft designer and aviation writer about this.
He said, as others here has stated, that in theory one blade is best
(as used on quite lot of aircraft in the 30's - Flying Fleas and such-like),
but that in reality four blades might well be better than two!
But he added that the theory is not well proven and that multiple
blades can well be better (using other formulas to back it up).
So who can't say science doesn't go forward? In this particular case it
was one fixed pitch propellor replaced by another, so the complications of
constant speed does no come into it. A user, a Cub part-owner, said the
replacement was a great advance, especially at low speed acceleration.
Take off distances were maybe 20% less, but cruise was marginally
affected. Another guy, using a speedier plane, found no advantage at all.
So if you're using something like a Cub (with little difference between
cruise and take off speed) go for as many you can afford (the new prop
cost around $5000 ... if remember correctly).
Isn't there a Larrabee out there who knows? :-)
--
Tord S Eriksson,
amateur in everything ...
Everything I have ever read on the subject states that the more blades
you have tends toward less efficiency. I have never seen any type of
formula or percentage that states how much less efficient each 'extra'
blade is though. Typically you add more blades for ground clearance
(you've never seen a two blade on a Corsair cause it would be so big it
would hit the ground). As far as smoothness(lower vibration), more
blades are better. As far as using the 3-blade Ivo or Warp being a
waste, (it's one of those age old compromises) do you want a little bit
more efficiency (2 blade), or more ground clearance and less vibration
(3+ blades)? The Ivo/Warp might be more efficient than a 2 blade wood
anyways if they use a more efficent blade design.
--
Mike Griffin
VP - Network Administrator
Plains National Bank of WT
***It's not a sport if it can't kill you!***
Here's my 2 blade vs 3 or more blade question-
Most aero engines run max ower at about 2500 RPM,
because of propellor efficiency.
Most auto engine's develope peak power at greater than
2500 RPM.
To match auto engine efficiency to propellor efficiency,
auto engine conversions are either de-rated or have a PSRU
that adds weight, cost, complexity and exacts an effiency penalty
of it's own.
Why not a smaller diameter prop, possibly 3 blades or more, and
run the engine rpm higher? I seem to remember pictures of late WW2
fighter planes with 6 blades.
-Dave
Has anyone told you that you can't speel? (grin)
Maybe you could use a chell specker? (more grin)
L8R G8R :)
>-Dave
Here is some info on props that I think will address both questions
that have been posed:
Given a constant HP, tip speed and "prop blade area" (actually solidity
ratio but that is similar), the larger the diameter of the prop, i.e.,
the prop disk area, the more efficient it will be. This is why people
do not go to small diameter props. This relationship hold regardless
of the number of blades among which the blade area is divided. In
fact, with these things held constant more blades are typically more
efficient than fewer blades.
Having said that... these things are very rarely held constant. When
people chose to use a 3-blade prop they typically do it because they
want to reduce the size of the prop disk and therefore 3-bladed props
are (in practice) often less efficient than the 2-bladed props they
replaced -- however they are often more efficient than a 2-bladed prop
of the same diameter.
To the best of my knowledge, the claim that 2-bladed props are more
efficient because they fly in "calmer air" than 3-bladed props is
without basis. People who have studied inter-blade interference
(particularly for turbofan engines) find that these effects are
minimal until the space between blades is approximately equal to the
chord width of the blades. No GA props I have seen are that dense...
the average prop has a couple orders of magnitude more space between
blades.
Moral of the story... If you replace a 2-bladed prop with a three
bladed prob of approximately the same total blade area and diameter,
you will probably get a small increase in efficiency. If you replace
it with a 3-bladed prob of smaller diameter you will probably get
lower efficiency. This assumes that both props were optimized for this
application, it is certainly possible to replace a particular large
diameter prop with a smaller one and incease efficiency is the larger
one was not optimized for this application, e.g., the tip speed was
more than about mach .8, it had too much/little twist, it had too
much/little area, etc.
Terry
Terry has provided an excellent summary of some pertinent propellor
theory, which you can only ignore if you do not care about efficiency.
The only thing that I would add, is the fact that the ability of the
propellor to convert horsepower into thrust, or thrust efficiency, goes
down as the delta V imparted to the airstream by the propellor disc goes
up. In other words, you get better thrust conversion by moving a LOT of
air a LITTLE than you do by moving a LITTLE air a LOT. You can improve
the efficiency of a high delta V propellor somewhat by enclosing it in
a duct. This is exactly what you see with a "high bypass turbofan"
engine. The duct eliminates the tip losses, which become really
significant with high delta V props. It is like induced drag, which
increases as the specific lift per unit area increases, tip losses
increase as the specific velocity change across the disc increases.
John
<snip>
It is normally quite easy to see who should get the big award in the
sci.electronics.design group; but here in rec.aviation the egos too
often seem to get in the way.
That's why I'm particularly pleased to award the first (but I hope
not the only) Shiny Cybernickel to Terry Schell and John Johnson
for their measured and balanced accounts of prop design considerations.
<grin>
brian whatcott <in...@intellisys.net>
Altus OK
Paul,
Ducted fans aren't around because they are creating too much drag.
They increase the efficiency of a propeller at very slow speed because the
duct act as an airflow guide and cancels bladetip vortex . So I think a
ducted fan has actually an increased thrust at takeoff, but probably not
much.
At flying speeds the duct is only a drag device. And that's the big
drawback.
Airplanes with ducted fans, such as the Fantrainer or the Optica, opted for
that configuration for structural reasons rather than propeller efficiency.
Luc
--
http://ourworld.compuserve.com/homepages/vanbavel/ Since 26SEP96
Mikoyan Design Bureau - Lockheed F-16 - Jet Squalus - FEA and Composites
Jesus Paul...what do you think powers most commercial airliners if it
isn't ducted fans?? Did I miss something somewhere?
I remember a few years back that Boeing or somebody like that was
experimenting with a smaller twin jet plane that had exterior blades
on the engines. not props but small fins that spun around a ring.
Anyone know anything about that?
--
:{) Jim
**************************************
* http://home.earthlink.net/~jaroot/ *
**************************************
³Remember its better to be on the ground wishing you were in the air,
than in the air wishing you were on the ground...²
old VFR saying, source unknown.
I am not an expert on either turbofans or props, but I do know the
awful racket put out by an early 210, Bonanza, etc. at high rpm is from
the prop tips going supersonic. Most of the energy from the last
hundred or so rpm just makes noise, the kind general aviation does not
need. If you want to really hear supersonic blade noise, find a C-185
with a seaplane prop. It's 88 inches, as I remember.
I'm not sure if the Q-tip props actually are more efficient, or just a
gimmick to look like winglets.
I get a laugh out of the guys who will polish and pamper their wings,
while totally ignoring their prop. The wing on a light genav plane
moves at maybe 0.3 mach max, while the prop tips are churning 0.9 to
1.1 mach. Remember, the wing can't provide lift that the prop hasn't
given it. The prop is the critical airfoil, so to speak. Wax your
prop, and you will get extra efficiency and longer life out of it.
RD
>
>> >
>> > Jesus Paul...what do you think powers most commercial airliners if
it
>> > isn't ducted fans?? Did I miss something somewhere?
>> Yea that is the dilemma. I don't know the answer to that. I was
looking
>> into the engine duct on NASA's 747 out at Edwards (open house) today
>> that carries the shuttle and I was pondering that very fact. I
really
>> don't think they have any other choice. When you get near mach one
>>you have to do something about tip speeds.
The big N1 fan turns 3600 rpm at 100% setting, so the tips on a six
foot fan are supersonic, even with the plane stationary. The 777 fan
is bigger than the fuselage of a 707/727/737 so its tips hit mach 1 at
a very low rpm. I leave it to you to do the math.
>>The duct is probably designed
>> as a defuser to slow the forward velocity of the air before the fans
>> get to go to work on it. I could be wrong. It has a very interesting
>> subtle S shape to it. The other thing I noticed was a bunch of
>> perforations inside the duct. Anybody know what those are for?
The perforations probably create a consistent boundary layer.
The CF-6, JT-9 and RB-211 engines on the 747, L-1011 and DC-10 all have
a buzz saw sound if it is not treated. Lockheed drilled and ported the
vanes, etc., to reduce the sound at the source. Boeing and Douglas
engineers weren't so smart; they just put a ton of sheet lead in the
sidewalls of the cabin to keep the sound from disturbing the
passengers.
But, guess what; at cruise the DC-10 and 747 are fairly quiet in the
cabin, while the L-1011 is much noisier.
>> I know GE had a rear engine turbofan airliner set up out at Mojave
to
>> test a turbo prop with a bunch of blades in place of the turbofan in
>> the interest of higher efficiency. I think they gave up because of
>> excess noise and vibrations impinging on the fuselage at the power
>> levels you need to fly an airliner.
>>
>> The new C130's have five blade propellors. Perhaps more power with
>>the same diameter, hence, ground clearance.
>>
>> BTW Rare Bear went faster when they took off the 4 blade and put on
a three blade.
>>
>> I do admit if some way can be found to eliminate the PSRU with high
>>RPM piston engines it should be done. PSRU's with piston engines are
>>heavy, costly and troublesome.
>>
>> Paul Lamar
>
>I remember a few years back that Boeing or somebody like that was
>experimenting with a smaller twin jet plane that had exterior blades
>on the engines. not props but small fins that spun around a ring.
>Anyone know anything about that?
McD installed one on their flight test MD-80 about 8-10 years ago. It
was a UDF UnDucted Fan, made by Allison, maybe. It sounded no louder
than the JT-8D, but higher frequency, so was an attention-getter to
someone used to jet noise. There were some promising attributes, but
lots of compromises on weight, etc. It would have required dual-dual
fail-operational yaw dampers, as I remember.
It didn't sell for the $10M extra asking price, so DAC went with the
IAE V-2500 and called it the MD-90.
RD
Tony
--
"Do not despair -- not even over the fact that you do not despair." --
Kafka
<snip>
>McD installed one on their flight test MD-80 about 8-10 years ago. It
>was a UDF UnDucted Fan, made by Allison, maybe. It sounded no louder
>than the JT-8D, but higher frequency, so was an attention-getter to
>someone used to jet noise. There were some promising attributes, but
>lots of compromises on weight, etc. It would have required dual-dual
>fail-operational yaw dampers, as I remember.
>It didn't sell for the $10M extra asking price, so DAC went with the
>IAE V-2500 and called it the MD-90.
>RD
I heard that the biggest reason that this project has not resulted in
a product was not technical but marketing. People just don't want to
fly in prop airplanes. The only way that airlines would purchase
these engines is if they offered such a large efficiency advantage
that they could absorb the loss of some passengers.
> I am not an expert on either turbofans or props, but I do know the
> awful racket put out by an early 210, Bonanza, etc. at high rpm is from
> the prop tips going supersonic. Most of the energy from the last
> hundred or so rpm just makes noise, the kind general aviation does not
> need. If you want to really hear supersonic blade noise, find a C-185
> with a seaplane prop. It's 88 inches, as I remember.
>
> I'm not sure if the Q-tip props actually are more efficient, or just a
> gimmick to look like winglets.
What about sweeping the blade tips? Part of the "prop-fan" research
alluded to earlier in this thread involved exotic shapes, such as the
scimitar, that radically swept the whole blade. Wouldn't a scimitar
blade designed for GA powerplants reduce the noise and efficiency
penalty of high speed blade tips?
BTW, Rutan has given us innovative airframe options. Why are the
propeller manufacturers still producing 1950s designs for 1990s
aircraft?
GB
>What about sweeping the blade tips? Part of the "prop-fan" research
>alluded to earlier in this thread involved exotic shapes, such as the
>scimitar, that radically swept the whole blade. Wouldn't a scimitar
>blade designed for GA powerplants reduce the noise and efficiency
>penalty of high speed blade tips?
George,
I think that a scimitar shaped blade of sufficient strength would have far too
much blade area for GA type powerplants. The NASA blades were driven
by humongous turbines. New carbonfiber technologies might solve this
problem.
Bob Moore
A big advantage of a scimitar is its ability to twist the prop in
response to load, giving a constant speed effect. That is not a minor
engineering and material selection feat, I would guess.
The Aeromatic prop blades are set at an angle in the hub so they will
rotate in response to air loads and counterweights. It is called an
automatic propeller, and is almost as good as a full constant speed at
a fraction of the weight.
BTW, I have a Sensenich Skyblade prop for sale, complete with oil
control:
C2FM4 74" 6 hole; fits O-235, etc. TC rated at 115hp, but one guy
is flying his on an O-290 at 140 hp.
Oil control gives redline rpm for takeoff, low rpm cruise.
Zero hours since overhaul. Certified airworthy. $1375 obo
RD
(your not in the savings & loan Business are you?)
> penalty of high speed blade tips?
>
> BTW, Rutan has given us innovative airframe options. Why are the
> propeller manufacturers still producing 1950s designs for 1990s
> aircraft?
>
> GB
Because we accept It?
Don't forget, and don't get me wrong I have nothing but the greatest
respect for Burt, but the Wrights had a canard plane a few years before
Burt. I believe the Vari-Viggen is a adaption of a Swiss or Swedish
plane. I don't know if he invented any new airfoil sections or not.
Basicly he made it in plastic. And also as far as homebuilts are
conserned if you building a Rutan design you building a plane designed
20? years ago.
I thinks that Warp and IVO may have made some steps forward in prop
design. and I would hope that they and others like them will continue to
be inovative. But if you take a look around in just this new-group you
will see that some people are un-willing to take advantage of 'new'
things. Remember that the prop people have to produce a saleable
product, in other works an accepted product.
But its a 'Wonderful Life' (TIC) isn't it!
--
:{) Jim
The Blackhawk that my daughter flies has swept tips on the rotor as does
the Apache. The chopper people are always fighting tip speed.
--
ken...@worldnet.att.net (Kenny Danielson)
Try Nagel's in Hesperia or Aero Prop in Hemet.
Mine is disassembled, with broken lag bolts in the blades. New blades
were once NLA, but may be available now.
RD
It has been tried and found to work but the noise was so intense as to
cause people to grab their ears 100 yards away. The noise was said to
have been a problen on the ground even with the plane at several thousand
feet. Approch and departure noise was beyond the threshhold of pain with
most of the noise in the plane-of-the-prop-disk or slightly behind the
prop tips.
Bruce A. Frank
Prop building material is just beginning to become hi-tec enough to
possibly stand up to these new shapes. There is still a lot of testing
going on. Even standard shaped props made out of exotic material(carbon
fiber/epoxy) sometimes come apart. It can be done but I think it will
still be a while before costs make the improvements available to GA.
After all they had to invent fiberglas reinforced epoxy before Rutan
could develop his innovations.
Bruce A. Frank
Yours, Tord S Eriksson, Ovralidsg. 25, S-422 47 Hisings Backa, SWEDEN
.
Bigger, slower turning propellers are always more efficient (until the
speed becomes so low that the propeller goes subcritcal (if that is the term),
Reynold's number-wise. Experience with more than three blades (an odd number
reduces vibration problems) varies. Old theory says 1 is best (you need a
counterweight on the other side then), but things seems not so clearcut
in practice. Modern choppers use multiblade rotors (7 is quite normal
for big ones), while the old theory would prefer two like on all old Bells.
But one has to consider that many blades means lower loading per blade and thus
is it possible to build a thinner, narrower blade that will have improved
aspect ratio and L/D. At the same time the blades interfere with each other,
so you can't be too sure about the ends result!
When the Explorer chopper with it's NOTAR tail the fan within the tail boom
was designed using propeller theory. It was not very efficient, but when it
was redesigned using fan theory they came very near what was theoretically
possible in efficiency. So not even MDC is too sure about how to skin a cat ...
Remember the converted sailplanes that were used as superquiet low altitude
recannaisance planes in Nam? First models had a Wankel above and behind
the pilot and a five-bladed
wooden prop that turned in fron tof the pilot driven by a long extension shaft.
Later versions have (for one is used by NASA to measure aerodynamic noise)
a more conventional engine in the nose driving a handcrafted propeller.
That aircraft is inaudible at 200 feet!
Harwards, old 210s and Convair Metroplitans, all on take-off, are on my top
list of extremely noisy propeller aircraft.
: Bruce A. Frank
The experimental turboprop version of the F-84 had tip speeds well
into the supersonic range. The noise physically debilitated the ground
crew during testing.
The Russian Bear bomber also has subersonic tips (and 32 of them)
at cruise. I've heard anecdotes that the pilots who ran the
phot-intercepts of them were warned of a few positions not to fly in when
escorting them.
--
Life is like a cow.
You get out of it what you put in. cali...@crl.com
But, umm... different somehow.
Also, the q-tips on propellers were supposed to control vortex/drag off of
the propeller and give better efficiency for thrust.
Frank Fichtl
Bruce A. Frank
>
>I have been following this thread with interest these last few days.
>As I understand it, the propeller loses efficiency when it begins to go
>supersonic. The airflow separates from the blade and it no longer "pushes"
>the air. I need to check some docs I have that talk about prop design.
>
>Also, the q-tips on propellers were supposed to control vortex/drag off of
>the propeller and give better efficiency for thrust.
>
>Frank Fichtl
>
>
My original question dealt with the possibility of reducing the diameter
of a prop, say by 50%, turn the engine twice as fast to get a better
power to weight ratio and avoid a PSRU in an auto engine conversion.
To counter the loss of propellor surface area, add more blades.
The replies have centerd on trans-sonic and supersonic prop tip speeds
and ducted fans- while interesting, it isn't exactly the topic I was wondering
about. If you cut prop diameter in half and double the prop RPM, the tip speed
stays the same, so the supersonic prop tip problems don't apply here.
Is a 4 bladed prop more expensive than a 2 blade? Is it more
expensive than a 2 blade plus PSRU? Is there a big efficiency loss with the
faster RPM, even though the tip speed is the same? A faster, smaller,
multiblade prop seeems an obvious answer to the PSRU problem for high output,
high rpm engines....there must be a catch.
-Dave
Yep, I know. I fly the Sikorsky S-76 which also has the same blade
technology as the Blackhawk.
My question still stands, has any prop manufacturer explored any of
these ideas at the GA level? There is research garnered by NASA studies
(with Hamilton Standard), I believe. Is it gathering dust or were all of
the solutions too costly. Where are the prop blade Roncz or Rutan type
innovators.
GB
The catch, so to speak, is the frontal area of the engine, or whatever
is in line with the prop. You won't find many singles with a fuselage
diameter greater than the prop dia. You also won't find much thrust in
the area in front of the engine in a tractor configuration.
RD
>Thrust comes from the momentum exchange obtained by accelerating the air.
>Thrust equals mass flow times change in velocity. But the power required
>increases as the SQUARE of the increase in velocity. So to get the same
>thrust from 1/4 the area (hence 1/4 the mass flow), you have to increase the
>velocity four times as much. That takes 16 times as much power. OUCH.
Oops! I think I squared a square in there somewhere. Takes twice the
velocity and four times the power. I HATE it when that happens!
Anyway, you get the idea - it ain't efficient!
(Red-faced) Dave
>My original question dealt with the possibility of reducing the diameter
>of a prop, say by 50%, turn the engine twice as fast to get a better
>power to weight ratio and avoid a PSRU in an auto engine conversion.
Hi Dave! Your's is the exact logic I started with in 1991 to save weight,
cost, and complexity. Unfortunately, the physics doesn't cooperate.
Here's the short answer: At half the diameter, you're at 1/4 the swept
area (disc). We are dealing with essentially incompressible flow (below
mach .85 or so), and the propeller blades don't see undisturbed air, but
interact. That's where "actuator disk" theory comes in.
Thrust comes from the momentum exchange obtained by accelerating the air.
Thrust equals mass flow times change in velocity. But the power required
increases as the SQUARE of the increase in velocity. So to get the same
thrust from 1/4 the area (hence 1/4 the mass flow), you have to increase the
velocity four times as much. That takes 16 times as much power. OUCH.
>To counter the loss of propellor surface area, add more blades.
Seems logical, but they interact. The blade area divided by the disk area
is the "solidity ratio". You basically want the solidity to go up as the
speed of the incoming air approaches the blade speed. Hence turbofans and
the old, low-speed Aeromotor windmills have high solidity. Newer (more
efficient) windmills have the blades traveling much faster than the
incoming wind, so low solidity. Helicopters same.
If you cut prop diameter in half and double the prop RPM, the tip speed
>stays the same, so the supersonic prop tip problems don't apply here.
And you've increased the blade speed/incoming air ratio, so higher solidity
is going in the wrong direction. Your disk is still 1/4 the area, and
you can't get around the physics.
> Is a 4 bladed prop more expensive than a 2 blade? Is it more
>expensive than a 2 blade plus PSRU? Is there a big efficiency loss with the
>faster RPM, even though the tip speed is the same? A faster, smaller,
>multiblade prop seeems an obvious answer to the PSRU problem for high output,
>high rpm engines....there must be a catch.
The catch is the v-squared problem. Then there's Reg Clarke and his Expresso
dragonfly with a direct-drive Suburu driving the tips supersonic. Reg doesn't
worry about all the physics problems - he just flys it all over the US and
Canada! I like his approach!!
Dave
Paul;
I think the information Terry might be refering to was an article in
KitPlanes about 6 to 8 months ago on prop diameter and efficiency.
The way I remember it, the author was a prop specialist who had decided
on oversizing the prop for his amphib project to get the extra power
transmission. He gave a lot of formulas in the article, which I played
around with just enough to see that they gave similar results to general
physics equations and moved on.
If anyone is heavily interested, they may want to check the back issues
for it.
Mike Johnson
<snip>
>My original question dealt with the possibility of reducing the diameter
>of a prop, say by 50%, turn the engine twice as fast to get a better
>power to weight ratio and avoid a PSRU in an auto engine conversion.
>To counter the loss of propellor surface area, add more blades.
> The replies have centerd on trans-sonic and supersonic prop tip speeds
>and ducted fans- while interesting, it isn't exactly the topic I was wondering
>about. If you cut prop diameter in half and double the prop RPM, the tip speed
>stays the same, so the supersonic prop tip problems don't apply here.
<snip>
Dave I suggest you go back and re-read my reply. The single most
important important design criteria in determining prop efficiency
(i.e., the amount of thrust given your BHP) is the size of the prop
disk relative to the HP... sometimes called the power loading. If you
cut the diameter by half, you reduce the area to 1/4 while keeping
the same HP... power loading goes up into the "inefficient" range.
Once the loading goes up that high, you need to start thinking about
alternative configurations (e.g., fan) to try to minimize the losses
inherent in the high power loading.
Terry
Don't quote me on this, but I think the problem is actually of
a natural logarithmic number. In other words, shortening the blades
by half will reduce your thrust by more than half, and twice the speed
probably wont compensate for the difference. It has something to do with
the availability of air at the blades themselves. More blade area is
essential for more direct contact to the air, and turning a blade twice
as fast probably wouldn't do it, and if I were to wager a simple
guess, I'd say four times as fast is probably more like it. Of course,
don't quote me on this, as I could be wrong, but I know someone out there
has a book on prop design, and can point you in the right direction.
--
__________________________________________________________________
| Curtis D. Levin | Team Os/2 | kd4zkw.ampr.org [44.98.2.22] |
| kd4...@bridge.net ~~~~~~~~~~~~| kd4...@amsat.org |
| http://www.bridge.net/~kd4zkw |kd4zkw@ae4ej.#MIAFL.FL.USA.NOAM|
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To add to this, most props on any given aircraft have to conform to
manufacturers specs for torque and horsepower ratings for a particular
bird. In other words, it's not a question of bigger and better. In a
constant speed design, the prop angle is adjusted by the throttle, giving
the engine a stable rpm setting. Because an Airctaft engine is higher
compression than an ordinary automobile engine, 2400 rpm is about the best
envelope for the operating speed. In prop replacement considerations, one
has to assume the actual amount of torque that is going to be returned by
the prop to the engine, and the horsepower needed to produce enough torque
to drive it. Therefore, even if a prop can fit to a certain aircraft,
there's no guarantee that it won't shorten engine life, if it happens to
overtorque the engine, irregardless of how much thrust it's capable of
generating. This is especially a consideration in fixed prop designs.
If you're thinking of replacing the prop on any given aircraft, check the
manufacturers specs for the amount of torque and horsepower it needs to
operate efficiently. Do not under any circumstances exceed these
recommendations, becuase engine failure becomes a real probability.
Also, you might have the best luck with a csd prop design, as they produce
the least amount of wear and tear on all of the engine components. As the
speed of the aircraft changes,so does the pitch of the prop, allowing it
to operate at almost maximum efficiency, discounting slippage, etc. Also
bear in mind that the faster the aircraft itself can move through the air,
the more pitch is needed to compenstae for the difference. So most props
probably have an optimum speed range somewhere in the manufacturers
recommendations, and if this speed is unobtainable, then so are the
benefits of that particular prop. There's alot of things that come into
play even before considering the engineering aspects of the prop itself,
and there are certain prop designs that will work as efficiently, or even
moreso, than a replacement of dubious specs.
Last, but not least, the best place to find the info needed to replace a
certain prop is in the aircraft etch data, and the data released by the
prop companies, for a particular rating. If you can match the numbers for
all of these things, then you have narrowed down your list of viable
replacements somewhat. After that, word of mouth is probably best to
determine which of those actual viable replacement models will work best
for your particular application. TTyl.
The best way to answer these questions is to measure the thrust
using a bathroom weight scales or similar. I have measured a
72hp McCullough turning a two-blade wood prop at 4000rpm to be
about 270lb. A 40hp Rotax with a 3:1 gearbox and the right
two-blade prop will pull 270lb.
The 4cyl VW engines are operated with direct drive and typical
54inch prop. The Rotax engines with gearbox can turn props up
to 72inch depending on airframe and clearance and proper gearing.
Yes, a smaller diameter multi-blade can be used as an alternate
to a larger two-blade but is not likely to be as efficient.
The Ultraprop 59inch 4-blade does a good job on a Rotax 447 geared
at 2.58:1. Does it generate 270lb? I do not know.
There are a few cases where a 3-blade is slightly better than a
2-blade but if you are not restricted to length, the longest
2-blade possible turning slower usually generates the most thrust.
I theorize that the inefficient portion of the prop blades closer to
the hub are minimized with the longest possible prop. The prop
is also inefficient near the tips so a multiblade may suffer from
this.
My $.02
Norm - Phantom flyer
<snip>
>The best way to answer these questions is to measure the thrust
>using a bathroom weight scales or similar. I have measured a
>72hp McCullough turning a two-blade wood prop at 4000rpm to be
>about 270lb. A 40hp Rotax with a 3:1 gearbox and the right
>two-blade prop will pull 270lb.
<snip>
Of course you can only measure static thrust with a scale. This may
not be very representative of your thrust at speed, which is what you
really care about. (with the possible exceptions of bush and sea
planes) For many design criteria there is a trade off between thrust
on the ramp and thrust at cruise, so *don't* use this method for
determining that best prop set up.
Very true, have been there in the back of a F101 when I was young. So
guess where all that noise energy comes from and what effect that has on
propulsion efficiency.
Nigel Field Subaru vari-eze
There is a serious catch, Dave. Basic Physics bites you. A propellor
is basically a wing traveling in a circle. We line several up behind
one another in a circle and each one travels in the wake of the other!
The "lift" the propellor blade produces is called "thrust." When you
make the propellor diameter smaller you basically increase the "wing
loading" which increases the "lift coefficient" the propellor blade is
operating at. That in turn, increases the "induced drag" which we
call "tip losses" in a propellor. We also have the minor problem
with every rotating set of wings, that the wing is not traveling at the
same speed throughout its length, like the fixed wing that sticks out
the side of the airplane does. The part near the center is actually
moving pretty slowly. Generally, the center third does not produce
any meaningful thrust in any case. If any part of the wing gets too
close to Mach 1 you start building shock waves on the propellor surface
right where the lift ( thrust ) is greatest! This immediately converts
most of the power going into the propellor disc into noise rather than
thrust. Of course the tip vortices produced by the greater pressure
change across the more highly loaded disc become more intense and also
waste great grizzly gobs of power. The MOST efficient prop would be
a single blade infinitely long turning at approximately zero rpm. By
most efficient here, I mean the most thrust per horsepower at zero
forward speed. As the airspeed increases we want the propellor to turn
a bit faster and the diameter can decrease a bit while the pitch can
increase a bit.
As we reduce the diameter of the propellor disc we increase the disc
loading and decrease the thrust per horsepower produced. This decrease
is more or less proportional to the AREA of the disc, which decreases
as the square of the radius of the propellor disc. It is quite a "gotcha"
and is the reason aircraft engines that are direct drive don't turn very
fast. Basically you are always between a rock and a hard place in
propellor design. You get more thrust per horsepower when you make the
diameter bigger, but you have to turn it slower because the tip speed
goes up with the circumference of the disc and you have to try to keep
an efficient prop below about 0.8 Mach so you don't get shock waves
building up on the parts of the propellor blade that are working hardest
for you.
Hope that all helps.
John
Well when I first got on this list I told you guys that my background is
construction, so here is my construction version of what is going on
with smaller prop diameter.
If you take a 1 inch pipe compare it to say a 50 inch prop, which is
basically trying to screw its way through a 50 inch 'hole' in the sky,
Now take a 1/2 inch pipe compare it to a 25 inch prop. I think you can
mentally visualize that 4 - 1/2 inch pipes can fit fairly well into a 1
inch pipe (so it will take 4 - 1/2 inch pipes to move the same amount of
water in a given period of time, all else being equal), hence you need
to move the air four times as fast to move the same amount of air in a
given period of time to produce the same amount of thrust.
So tip speed may be a problem.
These same problems are typical in designing scale aircraft. ie: scale
it down by 1/2 and the area of the structural materials deminishes by 75
percent.
>Paul Lamar <paul-l...@worldnet.att.net> writes:
>>Terry Schell wrote:
><snip>
>>> Given a constant HP, tip speed and "prop blade area" (actually solidity
>>> ratio but that is similar), the larger the diameter of the prop, i.e.,
>>> the prop disk area, the more efficient it will be. This is why people
>>> do not go to small diameter props. This relationship hold regardless
>>> of the number of blades among which the blade area is divided. In
>>> fact, with these things held constant more blades are typically more
>>> efficient than fewer blades.
>>Terry do you have any technical reference numbers on this work?
>>Paul Lamar
>The references for these basic findings are reported in any number of
>aero-texts. I think I learned most of this in an old book by Dommasch
>printed in the 1950's call something like "Elements of prop aerodynamics".
>I am sure that I read most of it again in either "Synthesis of
>sub-sonic airplane design" by Torenbeck or "Fundamentals of
>Aerodynamics" by Anderson. These are both newer (80's/90's) editions,
>however the vast majority of work on these types of props was done
>during wwII when they were trying to get both high efficiency and
>high-powerloading so the older reference is probably not out of date.
>It seems like the bulk of research since the 50's had focused on
>fans and noise levels.
>You consistently doubt my claim that for certain appications
>increasing the number of blades increases prop efficiency.
I, too, am wondering why, given the limits of tip speed and prop disc
size, why more blades would not be more efficient, as long as they
were not so close together that one would be traveling in disturbed
air from the previous blade (someone stated that was approximately 1
chord-length, which would certainly allow 5 blades). I'll try to find
those books, but in the meantime, does anyone have the ISBN number of
a book that _definitely_ explains why or why not more or less blades
are better, given the above limits?
> Why do you
>think that all of those big prop planes of WWII had 3,4, or 5 blades
>if 2 would have given more thrust per BHP?
I would think power loading per blade might be a big factor in adding
more blades, whether or not they were more efficient.
I listened to this debate last year, but I neglected to write down any
references either for or against... I may not agree with everything
Paul Lamar says on auto engines, but he is fond of quoting ISBN
numbers (he must have _some_ library!)... Paul, you got anything?
- jpr
>Terry Schell wrote:
<snip>
>> Given a constant HP, tip speed and "prop blade area" (actually solidity
>> ratio but that is similar), the larger the diameter of the prop, i.e.,
>> the prop disk area, the more efficient it will be. This is why people
>> do not go to small diameter props. This relationship hold regardless
>> of the number of blades among which the blade area is divided. In
>> fact, with these things held constant more blades are typically more
>> efficient than fewer blades.
>Terry do you have any technical reference numbers on this work?
>Paul Lamar
The references for these basic findings are reported in any number of
aero-texts. I think I learned most of this in an old book by Dommasch
printed in the 1950's call something like "Elements of prop aerodynamics".
I am sure that I read most of it again in either "Synthesis of
sub-sonic airplane design" by Torenbeck or "Fundamentals of
Aerodynamics" by Anderson. These are both newer (80's/90's) editions,
however the vast majority of work on these types of props was done
during wwII when they were trying to get both high efficiency and
high-powerloading so the older reference is probably not out of date.
It seems like the bulk of research since the 50's had focused on
fans and noise levels.
You consistently doubt my claim that for certain appications
increasing the number of blades increases prop efficiency. Why do you
> My original question dealt with the possibility of reducing the diameter
> of a prop, say by 50%, turn the engine twice as fast to get a better
> power to weight ratio and avoid a PSRU in an auto engine conversion.
> To counter the loss of propellor surface area, add more blades.
> The replies have centerd on trans-sonic and supersonic prop tip speeds
> and ducted fans- while interesting, it isn't exactly the topic I was wondering
> about. If you cut prop diameter in half and double the prop RPM, the tip speed
> stays the same, so the supersonic prop tip problems don't apply here.
> Is a 4 bladed prop more expensive than a 2 blade? Is it more
> expensive than a 2 blade plus PSRU? Is there a big efficiency loss with the
> faster RPM, even though the tip speed is the same? A faster, smaller,
> multiblade prop seeems an obvious answer to the PSRU problem for high output,
> high rpm engines....there must be a catch.
>
> -Dave
The catch deals with the high velocity of the eflux from a small prop which has to slow down and mix with the
free stream. The higher the eflux velocity relative to free stream the lower the efficiency. A prop delivers
thrust the same as a jet, by accelerating a mass of air. The larger the mass as in a big prop, the lower the
velocity to get the same thrust resulting in less friction loss when decelerating to free stream velocity. On
one extreme are helicopter rotors developing very high thrust by accelerating a large mass only a little. The
other extreme is a pure jet (zero bypass) which accelerates a small mass to supersonic velocity and has poor
efficiency at low speed. There is a bit more to it but thats the jist of the explanation.
One normally adds blades to increase activity factor (area) when the diameter is limited by other factors such
as ground clearance. If you are limited to higher RPM by direct drive, adding blades is a compromise and can
be effective on a fast airplane, but the smaller you go the closer you get to jet efficiency. Also consider
that a larger percentage of a small prop may be shaded by the fuselage or cowling further decreasing
efficiency. You will find that for a slower airplane, the Thrust to Weight ratio including the weight of a PSRU
will be considerably better using a large diameter prop geared down.
I agree with your concept however. The weak link is the PSRU. What the world needs is a really good propfan
design with some reasonable diameter, but now were into the supersonic thing at the tips.
Nigel Field
I have heard from others that this is not the case. (once on this thread I
belive). My understanding is that there isn't any significant
interference between props until the props are closer than an chord width
apart. Part of the reason might be that as the prop moves forward the
wake of the prop ahead of it is now behind it. Do you have any references
that might shed light on this. I have wonder about this for some time.
On another note, I am very interested in quiet props. My max prop is set
by my aircraft (a Saber Trike). The available power is set by my 447
Rotax. So given these two constants how do I achieve the quietest prop?
I think (please correct me if I am wrong) that this would achieved by
adding lots of blades at maximun diameter but gearing the prop down to
turn much slower. The prop loading would go down because of the number of
blades and lower speed. In addition, the stress in the prop would be much
less due to the lower RPM. THis may allow the use of cheaper materials.
WIll it work?
While I am on the soap box. Can anyone out there recommend some good
books on prop design and/or fan design and how they differ?
This is a great thread many thanks to all of you who are willing to share
your knowledge.
In a purely theoretical sense, one would have to figure that the less
surface area a prop has, the more efficient it will be --until you
start getting close to the speed of sound. The other issues that
creep in concern at what point the engines will produce the most
power, and how much of the engine cowl will affect the flow of air
past the prop.
The two versus three blade prop is really a question of engine to
propellor drive systems.
--
Jake Brodsky, mailto:fru...@erols.com
PP-ASEL IA, Cessna Cardinal N30946, Based @ MD24
Amateur Radio Station AB3A
"Beware of the massive impossible!"
> Because they had to take off from short fields and carriers frequently
> over loaded. Climb was also important in fighter planes. Also as I
> recall ground clearence was an issue as well. That's why the Corsair had
> a bent wing. I am not disputing that three and four blades give more low
> speed thrust. After all I have said over and over again that every time
> they are substituted for a two blade they climb better.
That seems to agree very well experiences here with replacing Cub's nornally
2-bladed propeller with a four-bladed (both fixed pitch): Much better
STOL capability but doubtful increase in top speed.
It's worth to note that modern versions of MD 500's have more rotor blades
to absorb the greater power output.
Simple theory would say two blades are more than enough. The momentum along
the crabkshaft might be much milder with multiple blades, but two blades
are optimum for high speed, of course. Preferably a fixed pitch propeller, too!
> I maintain for top speed fewer blades seem to be better. That's all.
> Rare Bear F8 Bearcat WWII fighter went faster with a 3 blade then it
> went with the orginal four blade.
No constant speed propeller have both optimum settings for both high and low
speed (due to the blade's twist), but are they designed for optimum crusie
or take-off? Anybody who knows?
'From your friendly RiscPC, on the other side of the sea ...'
I was stationed aboard the USS Guadalcanal in the early 80's and we
would regularly get buzzed by the Russian "Bear" recconaissance planes.
They sounded like a barrel full of Cox R/C plane engines all going full
blast.
Doug
>That seems to agree very well experiences here with replacing Cub's
nornally
>2-bladed propeller with a four-bladed (both fixed pitch): Much better
>STOL capability but doubtful increase in top speed.
>
>It's worth to note that modern versions of MD 500's have more rotor
blades
>to absorb the greater power output.
>
>
The three blade conversion for my Mooney M20C gives a quieter ride inside
but about a 5 MPH slower cruise. Same engine and airframe. Tested hours
apart.
The added rotor blades on an MD500 are there because with the added power
of the new engine, They needed more rotor area and could not increase the
rotor legnth without increasing the tail boom legnth and reducing rpm.
The rpm is needed for low noise and manuverability. Long tail booms limit
landing areas and make it heavier.
There are always trade offs.
Mark