PSRU design advantages

"Richard Lamb" wrote in message
k.net...
Peter Dohm wrote:

"Bill Daniels" bildan@comcast-dot-net wrote in message
...

The basics:

Piston engines produce more power per pound if they rev higher. (HP =
RPM

x

torque/5252)
Propellers are MUCH more efficient if they turn slow.
This begs for a PSRU.
BUT, a PSRU adds weight, cost and complexity.
Resonances, particularly torsional resonances are a real problem.
Lots of examples of PSRU's on 12, 14 and 18 cyinder engines
Few workable examples with fewer cylinders suggesting PSRU's don't like
power pulses.
If a shaft has a strong resonant fundamental, don't excite it or lower
the
fundamental below the input frequency.
Tuning a PSRU/shaft/propeller system is like tuning a piano - it's an
art
not a science.


The 9 cylinder 1820 and 1840 CID radials used on B-17's were geared
approximately 16:9. However, your point is well taken, and I also am
unable
to name any 4 or 6 cylinder engines that have stood the test of time
with
reduction drives.

I also believe that tuning any drive system, including a PSRU, is a
science--when fully understood. And therein lies the rub: There's
plenty
left to learn--especially if it must also be light. So, in practice,
you
are right--it is still an art. :-(

Peter




Rotax - the 912/914

Jabaru - (but the 6 cylinder will be a better seller - IMHO)

Believe it or not, a few VW's with belts.

And a couple of Subes with Rotax B boxes scabbed on.

The one that DIDN'T work was the Geo Metro 3-banger (broke the crank).

But that issue was already known - don't cut off any flywheel on 3 holers.
With the full flywheel, the 3 cylinder runs fine.


Richard

OK, you caught me fair and square on poor phrasing. I tend to think of
higher power applications, but you are right that some of the more
conservative and lower powered systems with flywheels still in place and a
little looser coupling seem to run quite reliably. I don't know how much
power is lost to friction, but some of the v-belt reduction drives even seem
to work quite reliably without any external crankshaft support!

Peter

"Dan Horton" wrote in message
ups.com...
There is nothing that eliminates a long shaft from the design of a
PSRU. Nonbelievers might be advised to consider ship propulsion; long
shafts, low cylinder counts, propellers operating in uneven flow, often
via a gearbox. Sound familiar?

The important issue is torsional stiffness of the shaft, not
length. A long shaft can be torsionally stiff or soft, depending on
diameter and material. The engineering process will tailor torsional
stiffness of the shaft (along with a number of other factors) to adjust
natural frequency.

The information you need is found in engineering texts, not on RAH.
The subject can be complicated, but there are no unknowns. You will
find most of the torsional vibration classics listed in the
bibliography of Taylor's "Internal Combustion....". Some texts, like
Wilson's "Practical Solution.." (the ultimate reference) will be
difficult to locate. Try a large university library. The best readily
available text (sort of the ultimate primer on all matters vibrational)
is JP DenHartog's "Mechanical Vibrations". You can buy it for less
than $15 at Amazon. Here is a short list:

CF Taylor, "The Internal-Combustion Engine in Theory and Practice",
1966 (vol. 1), 1968 (vol. 2), MIT Press

W Ker Wilson, "Practical Solution of Torsional Vibration Problems", 3rd
Ed, 5 Vols., 1956, 0412091100, Chapman & Hall

JP Den Hartog, "Mechanical Vibrations", 1956, 070163898, McGraw-Hill

My compliments to Mr. Christley, whose comment (re frequency) was a
sole beacon of accuracy.


Dan Horton

You are very probably right--and it won't be the first time that I believed
that something was still a "black art" until I found out otherwise. For
years after I first became an electronic technician, I believed that about
grounding problems--and then I read a book titled "Sheilding and Grounding
Techniques in Instrumentation." Even 20 years ago, that book was long out
of print; but could still be obtained by special order from University
Microfilm. Almost miraculously, the problems went away!

After reading your post, I decided to look for the books you mentioned and
found that you were correct about the difficulty of locating W Ker Wilson's
book. That could indicate that it is the true source, as the dates
mentioned for earlier editions suggest, and therefore a custom reprint could
be worth every penny and more if a source is known.

The other two books seem to still be available, although I have no idea when
I might find time to read them...

Peter

Peter Dohm wrote:
"Richard Lamb" wrote in message
k.net...

Peter Dohm wrote:


"Bill Daniels" bildan@comcast-dot-net wrote in message
...


The basics:

Piston engines produce more power per pound if they rev higher. (HP =

RPM

x


torque/5252)
Propellers are MUCH more efficient if they turn slow.
This begs for a PSRU.
BUT, a PSRU adds weight, cost and complexity.
Resonances, particularly torsional resonances are a real problem.
Lots of examples of PSRU's on 12, 14 and 18 cyinder engines
Few workable examples with fewer cylinders suggesting PSRU's don't like
power pulses.
If a shaft has a strong resonant fundamental, don't excite it or lower

the

fundamental below the input frequency.
Tuning a PSRU/shaft/propeller system is like tuning a piano - it's an

art

not a science.


The 9 cylinder 1820 and 1840 CID radials used on B-17's were geared
approximately 16:9. However, your point is well taken, and I also am

unable

to name any 4 or 6 cylinder engines that have stood the test of time

with

reduction drives.

I also believe that tuning any drive system, including a PSRU, is a
science--when fully understood. And therein lies the rub: There's

plenty

left to learn--especially if it must also be light. So, in practice,

you

are right--it is still an art. :-(

Peter




Rotax - the 912/914

Jabaru - (but the 6 cylinder will be a better seller - IMHO)

Believe it or not, a few VW's with belts.

And a couple of Subes with Rotax B boxes scabbed on.

The one that DIDN'T work was the Geo Metro 3-banger (broke the crank).

But that issue was already known - don't cut off any flywheel on 3 holers.
With the full flywheel, the 3 cylinder runs fine.


Richard


OK, you caught me fair and square on poor phrasing. I tend to think of
higher power applications, but you are right that some of the more
conservative and lower powered systems with flywheels still in place and a
little looser coupling seem to run quite reliably. I don't know how much
power is lost to friction, but some of the v-belt reduction drives even seem
to work quite reliably without any external crankshaft support!

Peter



You didn't follow the link that blueskies posted, didja Peter.

The BD-5 story - in all it's glory! And a few other odds and ends,
That was not a high powered setup, but kicked a bunch of engineers around.

http://www.prime-mover.org/Engines/T.../contact1.html


Richard

"Richard Lamb" wrote in message
nk.net...
Peter Dohm wrote:
"Richard Lamb" wrote in message
k.net...

Peter Dohm wrote:


"Bill Daniels" bildan@comcast-dot-net wrote in message
...


The basics:

Piston engines produce more power per pound if they rev higher. (HP =

RPM

x


torque/5252)
Propellers are MUCH more efficient if they turn slow.
This begs for a PSRU.
BUT, a PSRU adds weight, cost and complexity.
Resonances, particularly torsional resonances are a real problem.
Lots of examples of PSRU's on 12, 14 and 18 cyinder engines
Few workable examples with fewer cylinders suggesting PSRU's don't
like
power pulses.
If a shaft has a strong resonant fundamental, don't excite it or lower

the

fundamental below the input frequency.
Tuning a PSRU/shaft/propeller system is like tuning a piano - it's an

art

not a science.


The 9 cylinder 1820 and 1840 CID radials used on B-17's were geared
approximately 16:9. However, your point is well taken, and I also am

unable

to name any 4 or 6 cylinder engines that have stood the test of time

with

reduction drives.

I also believe that tuning any drive system, including a PSRU, is a
science--when fully understood. And therein lies the rub: There's

plenty

left to learn--especially if it must also be light. So, in practice,

you

are right--it is still an art. :-(

Peter




Rotax - the 912/914

Jabaru - (but the 6 cylinder will be a better seller - IMHO)

Believe it or not, a few VW's with belts.

And a couple of Subes with Rotax B boxes scabbed on.

The one that DIDN'T work was the Geo Metro 3-banger (broke the crank).

But that issue was already known - don't cut off any flywheel on 3
holers.
With the full flywheel, the 3 cylinder runs fine.


Richard


OK, you caught me fair and square on poor phrasing. I tend to think of
higher power applications, but you are right that some of the more
conservative and lower powered systems with flywheels still in place and
a
little looser coupling seem to run quite reliably. I don't know how
much
power is lost to friction, but some of the v-belt reduction drives even
seem
to work quite reliably without any external crankshaft support!

Peter



You didn't follow the link that blueskies posted, didja Peter.

The BD-5 story - in all it's glory! And a few other odds and ends,
That was not a high powered setup, but kicked a bunch of engineers around.

http://www.prime-mover.org/Engines/T.../contact1.html


Richard

Actually I did, some months ago following an earlier post, and subsequently
also learned that the Contact! article is quite famous. One of the more
interesting points was that trying to make the shaft and/or transfer drive
more rigid was not helpful on the BD-5. Softening the system eventually did
resolve the breakage problem within the drive train; but IIRC the drive
system to airframe resonance (evidenced initially by loosened rivets) was
not fully resolved during the author's tenure. That was the article that
really convinced me that I didn't necessarily know enough to design a
clutchless system with a high degree of confidence--even by leaving the
flywheel in place.

However, the set of books mentioned elsewhere in this thread, by Mr. Horton,
could prove to contain the necessary formulas and explanations to reduce
this problem to a cookbook science. A quick web search confirmed his belief
that one of the books may now be virtually unobtainable.

I am willing to entertain his book suggestion because, in my earlier career
as an electronic technician, a technical tome entitled "Shielding and
Grounding Techniques in Instrumentation" made previously insurmountable
grounding problems easy to solve. It is probable that work on mechanical
resonance, done for World War II, may have been covered in books published
during the succeeding quarter century.

Peter

"ADK" wrote in message
news:WlXYf.4880$4S.2741@edtnps82...
Actually I do work in aviation. I am an aviation machinist and aircraft
mechanic, I also work on Allison turbines (hercs and convairs) that drive
a gearbox via a shaft. My experience is mostly helicopters but being a
fixed wing pilot I want to have my own plane for cross country flights. I
don't believe any one person can ever learn everything there is to know
about a subject and therefore I am was soliciting usefull information on
this subject.
Thank you!

Good for you. Read all you can, talk to some others that have been there,
done that. They are not on this group, though. Find the author of the link
that was posted on the subject. Then, if your heart is set on it, start
experimenting, and be prepared to experiment, a bunch! g

Good luck!
--
Jim in NC

"Dan Horton" wrote

There is nothing that eliminates a long shaft from the design of a
PSRU. Nonbelievers might be advised to consider ship propulsion; long
shafts, low cylinder counts, propellers operating in uneven flow, often
via a gearbox. Sound familiar?

I think you will find that they do it on ships, with pure weight. A big,
heavy, solid steel shaft. Very heavy! That is how they get the stiffness.

Also, the shaft turns very slowly, so there are many pulses per revolution;
more than you will get with a 4 or 6 cylinder, 4 cycle airplane engine, in
most cases.

I agree with the rest of your post; dig into the engineering text books.
--
Jim in NC

Richard Lamb wrote:
Ian Stirling wrote:

Peter Dohm wrote:
snip

The 9 cylinder 1820 and 1840 CID radials used on B-17's were geared
approximately 16:9. However, your point is well taken, and I also am unable
to name any 4 or 6 cylinder engines that have stood the test of time with
reduction drives.

I also believe that tuning any drive system, including a PSRU, is a
science--when fully understood. And therein lies the rub: There's plenty
left to learn--especially if it must also be light. So, in practice, you
are right--it is still an art. :-(


I suspect that electronics help.
Instrumenting the shaft, to measure resonances in real time is no longer
prohibitively expensive.
I suspect a belt PSRU - if properly configured could act to decouple the
prop from the engine/shaft somewhat.
Add one or more rotational vibrational dampers - fill the shaft with
oil? And trim.

Best tool available to the amateur is a variable speed strobe - Party Light!

That way you can actually look and SEE what's happening.

That'll spot ordinary vibrations.
Torsional ones are a little bit harder.

Especially if you want, as you probably should, a graph of maximum stress
anywhere in the shaft/PSRU/Prop system vs RPM.

ADK wrote:
IF you had to design a PSRU, to drive a pusher propellor via shaft, what
would your experience dictate? Thinking along the lines of a gearbelt, chain
or gear. Please, I would appreciate the collective experience available on
this group. I have decided on the aircraft, but want to make it the most
reliable and safest it can be.


For the sake of Peter, IT DOESN'T MATTER!!

For the energy to transfer to the prop, you have to attach the engine to
the prop. The engine doesn't produce smooth even power. It produces a
series of pulses. If the frequency of the pulses resonates with the
prop or shaft, it will store a little bit of each pulse as "spring
energy". This type of energy is stored by deflecting (ie, bending) the
prop or shaft. The prop stores it and then immediately tries to release
it by unbending. If the next engine pulse comes along at just the right
time, the new "spring energy" will be added to the previous "spring
energy" and the prop will bend a little more. This continues until the
prop or shaft has as much "spring energy" as it can phyiscally hold, and
then the element just vibrates. Eventually, the prop or shaft gets
tired of all the bending and unbending and just gives up (ie, breaks).

Making the pulses smaller doesn't help for the most part. All that does
is cut down on the amount of "spring energy" added with each pulse. A
smaller pulse will take 2000 pulses to fill the prop with "spring
energy" vs 1000 with a unmodified pulse. Whoop-te-do! What difference
will that make with the engine turning 2000RPM and four pulses per round.

Any one of the gearboxes you mentioned made to work safely, and each has
a set of advantages and disadvantages that are well known and easily
engineered around. The type of gearbox has nothing to do with torsional
resonance. Will not mitigate torsional resonance. Will not
cure/alleviate/lesson or bypass torsioanl resonance. Torsional
resonance is a totally different issue.

You didn't tie gearbox type and torsional resonance together directly,
but many people have in the past, and it's just self-deception. Any of
the gearboxes you mentioned can be as safe and dependable as any of the
others, if engineered properly.


--
This is by far the hardest lesson about freedom. It goes against
instinct, and morality, to just sit back and watch people make
mistakes. We want to help them, which means control them and their
decisions, but in doing so we actually hurt them (and ourselves)."

Ian Stirling wrote:

Best tool available to the amateur is a variable speed strobe - Party Light!

That way you can actually look and SEE what's happening.


That'll spot ordinary vibrations.
Torsional ones are a little bit harder.

Especially if you want, as you probably should, a graph of maximum stress
anywhere in the shaft/PSRU/Prop system vs RPM.

A few fine white lines down the length of the shaft will clear up that
problem.

--
This is by far the hardest lesson about freedom. It goes against
instinct, and morality, to just sit back and watch people make
mistakes. We want to help them, which means control them and their
decisions, but in doing so we actually hurt them (and ourselves)."

"Ernest Christley" wrote in message
...
ADK wrote:
IF you had to design a PSRU, to drive a pusher propellor via shaft, what
would your experience dictate? Thinking along the lines of a gearbelt,
chain or gear. Please, I would appreciate the collective experience
available on this group. I have decided on the aircraft, but want to
make it the most reliable and safest it can be.


For the sake of Peter, IT DOESN'T MATTER!!

For the energy to transfer to the prop, you have to attach the engine to
the prop. The engine doesn't produce smooth even power. It produces a
series of pulses. If the frequency of the pulses resonates with the prop
or shaft, it will store a little bit of each pulse as "spring energy".
This type of energy is stored by deflecting (ie, bending) the prop or
shaft. The prop stores it and then immediately tries to release it by
unbending. If the next engine pulse comes along at just the right time,
the new "spring energy" will be added to the previous "spring energy" and
the prop will bend a little more. This continues until the prop or shaft
has as much "spring energy" as it can phyiscally hold, and then the
element just vibrates. Eventually, the prop or shaft gets tired of all
the bending and unbending and just gives up (ie, breaks).

Making the pulses smaller doesn't help for the most part. All that does
is cut down on the amount of "spring energy" added with each pulse. A
smaller pulse will take 2000 pulses to fill the prop with "spring energy"
vs 1000 with a unmodified pulse. Whoop-te-do! What difference will that
make with the engine turning 2000RPM and four pulses per round.

Any one of the gearboxes you mentioned made to work safely, and each has a
set of advantages and disadvantages that are well known and easily
engineered around. The type of gearbox has nothing to do with torsional
resonance. Will not mitigate torsional resonance. Will not
cure/alleviate/lesson or bypass torsioanl resonance. Torsional resonance
is a totally different issue.

You didn't tie gearbox type and torsional resonance together directly, but
many people have in the past, and it's just self-deception. Any of the
gearboxes you mentioned can be as safe and dependable as any of the
others, if engineered properly.


--
This is by far the hardest lesson about freedom. It goes against
instinct, and morality, to just sit back and watch people make
mistakes. We want to help them, which means control them and their
decisions, but in doing so we actually hurt them (and ourselves)."

Let me try a different idea.

Suppose the prop shaft is to be just long enough for the gear belt pulley
and the neccessary bearings - say 10 inches. But the engine flywheel pulley
is to be 4 - 6 feet below the prop shaft. The idea is to use a very large
multi-blade carbon fiber prop turning 800 - 1000 RPM driven by a 4 cyl Soob
turning at best power RPM. The idea is to get best thrust in the 0 - 60
knot range. The airframe configuration is a prop over tail boom pusher - an
ultralight on steroids. (BTW, I'm not looking for a long engine life under
these conditions. I'll treat the Soob as a throwaway power plant.)

I'm thinking there isn't too much torsional vibration concern with very
short shafts, high reving engine and a stiff carbon fiber prop. The prop
will be seeing 6 - 7 power pulses per rev from the high reving Soob.

Bill