Torsional Vibration and PSRU Design

Dan Horton wrote:
Dan, based on my studies, I don't think they (clutch springs) are
there to drop total
system resonance below idle speed......I am pretty sure that the
frequency of the driveline system is already low.

Good argument. Consider me corrected, with a caveat. I think
you're right about driveline frequency already being low, even if there
were no clutch. The caveat? The clutch springs are one of the
stiffnesses in that system, and contribute to that low frequency. Take
them out, frequency goes up. Design them in, frequency goes down.

I wasn't disputing the effect you had described, it would be hard to
disagree. My point was more along the lines of --- "are you sure that
is why they are there?" Crankshaft counterweights also lower the
system frequency, but thats not why they are on the crank. It is also
why I am tentative in this. I don't know for a fact, so I may
speculate, but will label it as speculation. I had researched it
before because I had heard the story about them being detuners which I
never found any support for in the literature and your numbers put to
bed. The only things I have found attributed to them in engineering
texts ( not websites or enthusiast pubs like Hot Rod) was shock loading
and the excerpts I quoted before.

Note the use of the term "damper" in the quoted text. Are you sure
the text wasn't speaking of something a bit larger than our light duty
clutches? Not much sign of a frictional damper in the Subaru clutch.


I am pretty sure, given that the image next to the text was the classic
clutch disc we are talking about. For further insight look up patent
2,674,863 It talks about the limitations of the friction mechanism in
standard clutch disc dampers. I have found though that in automotive
practice they seem to take a lot of liberties with the terms dampers and
absorbers. Look at the term "shock absorber". And they call the
detuner a balancer. I don't have a clutch disc around here at the
moment to look at. I have one in a storage building across town, so
may go look at it in more detail.


Regarding the rubber elements someone mentioned in their driveline,
"Automobile Engineers Reference" makes mention of these as well, saying
that they can provide similar damping to the clutch damper

Rubber elements do have a damping value, although it is very, very
small. We got the actual value from Lovejoy when we were doing the
modeling, but logic alone tells you it ain't much. If it had much
damping value, it would melt g

If you stick one in the driveline of an auto it cant help but get hit
with torsional vibrations and they can be found there. We are not
talking about damping at system resonance. The damping is dependent on
the hysteresis of the ruber which obviously creates heat, but it is
exposed to ample cooling air in a drive shaft if the amplitudes and
frequencies are mild I would expect. The rubber is going to heat up if
you are using them to correct for axial misalignment as well, since you
will be doing the same thing to it, alternately stretching and
compressing it. In any event, not my suggestion, just right out of the
manual "Autombile Engineers Reference Book" by Molloy (the book does not
use these terms, but the pictures make it obvious that the Layrub is a
rubber in compression unit and the Rotoflex is a rubber in shear.):

"A normal rear axle with a hotchkiss drive has probably adequate overall
flexibility in the drive shaft aft of the gearbox, and any flexibility
provided by such couplings as the Layrub and Rotoflex is probably
desirable only for local effect, i.e. to reduce gearbox chatter in some
cases. There does, however, in some cases seem to be a marked and very
welcome quieting effect in the car from their use and this may arise
from their effect in reducing the transmission of road noise. Their
torsional flexibility also becomes more desireable overall where more
positive and therefore more rigid drive and braking torque resisting
means are provided on the axle, and even more when independent rear
suspensions or a De Dion axle is used and the flexibility of the
half-shafts is largely subtracted too. This aspect, torsional
flexibility, can therefore be of increasing importance in the future"

and

"In both these cases, the axial flexibility is sufficient to make it
possible to dispense witha sliding joint in the propellor shaft..."


The comment about local effect is interesting and goes back to the
original discussion about the springs in the clutch plate.


Charles

Dan Horton wrote:
Charles,
I said: I suspect that eliminating transmission noise (with selector
in neutral, clutch engaged, mainshaft spinning) is the purpose of the
654 ft-lbs/rad spring rate found at less than 3.5 degrees
displacement.

Got curious and ran numbers for a simple two element model. I used
what I think are reasonable guesses for the inertias, 0.07 slugs-ft^2
(crank, flywheel, and most of the clutch assembly), and 0.01 slug-ft^2
(transmission mainshaft and gearset). The connecting stiffness is of
course 654 ft-lbs/rad.

No joy. The above yields an F1 of 43.5 hz. That would make the
mainshaft rattle like hell at 1305 engine RPM with the selector in
neutral, so my guess about the purpose of the 654 spring rate does not
appear to be true. 654 isn't soft enough.

You got an idea about the 654 rate?

Dan


Two things. To get that spring rate(654), there would have to be
enough resistance in the transmission to absorb that level of torque. I
think that would be a pretty inefficient transmission. I doubt it. I
expect that in neutral, clutch engaged , idling you would have to be at
the lower limit of the spring since by definition you are at the lower
limit of torque, so look at the rates there. Once the shaft was
acclerated up to speed, the only torque that could be transmitted
through the springs would be the result of friction.

Second thing, dyno charts show the maximum torque the engine can provide
at a given rpm, doesn't mean that the engine has to produce that torque
at that rpm ( you do have a throttle right? the engine speed isn't
controlled solely by load). Every car is different, but I have heard
quotes of 25-30hp to cruise on the highway, so your 40ft lbs of torque
is almost enough at 3600 rpm for highway cruising, which means you could
see much lower torque values in high gear crusing at thirty miles an
hour. Finally, the torque on the dyno chart is the mean torque, not an
absolute ( the torque variations are how we got here in this
conversation) so I would expect it to have a range. But a definative
answer I can't give you, just my very unqualified thoughts. By the way,
I think I figured out why the auto engineers call it tip in and tip out.
Must be slang for TP in and TP out, the TP being throttle position.

Charles

After sleeping on it, I will bet that the lower rate is there for tip in
tip out discussed from cruise and or initial engagement. I do remember
reading about using multiple spring rates for overdrive vs non overdrive
states, but don't recall where I read that. Realized after I posted
that the rate you quoted was probably from 0-3.5 not starting at 3.5

Charles

"Dan Horton" wrote in message
oups.com...
To get that spring rate(654), there would have to be enough
resistance in the transmission to absorb that level of torque.

Uh oh, terms confusion. That's per radian. Torque vs deflection
is 0 to 40 ft-lbs across 3.5 degrees. Convert that to ft-lbs/rad (or
an equivelent term like N m/rad) to do a natural frequency calculation.
(40/3.5) = 11.428571 ft-lbs per degree. Multiply by 57.29578 = 654
ft-lbs per radian.

Second thing, dyno charts show the maximum torque the engine can
provide
at a given rpm, doesn't mean that the engine has to produce that torque

at that rpm

Good point! You're right, duh on me. Entirely possible to be
cruising part throttle at torque levels that do not deflect the clutch
center beyond the first 3.5 degrees and it's softer rate.

Uhhh, we're getting way sidetracked from a PSRU topic.

Well, we may be getting way sidetracked from a PSRU topic, and then again we
may not be.

I think that the clutch springs may also be a way of providing a more
constant torque, and eliminating the torque reversals that would otherwise
be inherent in a 3 or 4 cylinder engine. That would provide far less
excitation to the prop and be a lot easier on the PSRU. It may also be the
original reason for their use in a car.

The greater question, by my way of reasoning, would be how much additional
mass (flywheel) or damping (viscous/hydraulic disk) on the PSRU side of the
clutch/spring assembly would provide enough additional smoothing to justify
the added weight. I doubt that the decision is really "open and shut" since
the PSRU, prop, and prop shaft would not need to be as strong.

Peter

Dan

Dan Horton wrote:
The JN-4C project had several goals. A friend wanted to proof
custom software written to model torsional vibration in a complex
aircraft drive system (a pusher with with long shafts, might I add).

Makes me sweat just thinking about it. What was the pusher - an Imp?

Peter Dohm wrote:
"Dan Horton" wrote in message
oups.com...

To get that spring rate(654), there would have to be enough
resistance in the transmission to absorb that level of torque.

Uh oh, terms confusion. That's per radian. Torque vs deflection
is 0 to 40 ft-lbs across 3.5 degrees. Convert that to ft-lbs/rad (or
an equivelent term like N m/rad) to do a natural frequency calculation.
(40/3.5) = 11.428571 ft-lbs per degree. Multiply by 57.29578 = 654
ft-lbs per radian.

Second thing, dyno charts show the maximum torque the engine can
provide
at a given rpm, doesn't mean that the engine has to produce that torque

at that rpm

Good point! You're right, duh on me. Entirely possible to be
cruising part throttle at torque levels that do not deflect the clutch
center beyond the first 3.5 degrees and it's softer rate.

Uhhh, we're getting way sidetracked from a PSRU topic.


Well, we may be getting way sidetracked from a PSRU topic, and then again we
may not be.

I think that the clutch springs may also be a way of providing a more
constant torque, and eliminating the torque reversals that would otherwise
be inherent in a 3 or 4 cylinder engine. That would provide far less
excitation to the prop and be a lot easier on the PSRU. It may also be the
original reason for their use in a car.

The greater question, by my way of reasoning, would be how much additional
mass (flywheel) or damping (viscous/hydraulic disk) on the PSRU side of the
clutch/spring assembly would provide enough additional smoothing to justify
the added weight. I doubt that the decision is really "open and shut" since
the PSRU, prop, and prop shaft would not need to be as strong.

Peter

Dan






I believe that brings the thread full circle.

Text is a hard medium.
A photograph with circles and arrows and a paragraph on the back comes to mind.
But in text, it's sometimes hard (spelled a lot of work) to be really articulate.

Well...
If someone needs a clue, this thread ought to give them plenty to think about.


Richard

Somewhere along the way, something strange clicked in somewhere and I think I
better understand how microwave ovens work now - and we didn't even go there!

-----------------snip---------

I believe that brings the thread full circle.

-----------------snip----------

You're right, and discovered just now that I had missed Dan Horton's post
dated 4/15/06 at 7:16pm where he stated in part:

" The JN-4C project had several goals. A friend wanted to proof
custom software written to model torsional vibration in a complex
aircraft drive system (a pusher with with long shafts, might I add). I
wanted a new and improved PSRU. So, we modeled the old drive and then
altered the model for optimum predicted results. Then I designed a
drive to match the model inputs, built it, and ran it with telemetry to
check the accuracy of the predictions. Along the way we developed a
damper and tested it, ran two different props while I had the
telemetry, played with strobing linear vibration, and a whole bunch of
other stuff. End result was proven software and a pretty good PSRU,
plus an education."

So it appears that there is a solution, which may or may not be for sale or
rent...

Peter

Peter Dohm wrote:
I think that the clutch springs may also be a way.......

Classic homebuilder's disease.

Why adapt a component (with unknown properties, clunky packaging,
and a least one severe drawback) when you can purchase products
engineered for the task? Consider the Goetz, Lord, or Lovejoy soft
couplers. Smaller, lighter, and (the really important part), they come
with complete engineering information.

A quick glance at the Centaflex application data shows 16 different
physical sizes with 34 different torsional spring rates, all with
complete data including nominal torque, max torque, allowable
misalignments, nominal and max twist in degrees, weight and mass moment
of inertia. You also get multiple mounting methods.

If nothing else, it is rather nice to be able to model for
predicted results, then select the required torsional stiffness from a
list.

The greater question, by my way of reasoning, would be how much
additional mass (flywheel)......

You really want to say "inertia", not mass. There are two good
reasons. First, using accepted, correct terms greatly improves
discussion. The subject is complex enough without everyone inventing
or misusing terms on the fly, and drilling yourself on correct terms
will help you with correct thinking. Second, as a practical matter, it
possible to build two flywheels of equal mass moment of inertia, but
unequal mass.

or damping (viscous/hydraulic disk) on the PSRU side of the
clutch/spring assembly would provide enough additional smoothing to
justify the added weight.

In the case of the Suzuki drive, the experimental damper added about
5 lbs and lowered steady-throttle resonant vibratory torque from 180
ft-lbs to 115 ft-lbs. Some portion of that 5 lbs also served as
flywheel inertia, a nice design bonus. A torsional damper does not
operate "on the PSRU side of the spring". It operates in parallel with
the spring, or to be more precise, in parallel with a connecting
torsional stiffness.

Dan

I think that the clutch springs may also be a way.......

Classic homebuilder's disease.

Why adapt a component (with unknown properties, clunky packaging,
and a least one severe drawback) when you can purchase products
engineered for the task? Consider the Goetz, Lord, or Lovejoy soft
couplers. Smaller, lighter, and (the really important part), they come
with complete engineering information.

A quick glance at the Centaflex application data shows 16 different
physical sizes with 34 different torsional spring rates, all with
complete data including nominal torque, max torque, allowable
misalignments, nominal and max twist in degrees, weight and mass moment
of inertia. You also get multiple mounting methods.

If nothing else, it is rather nice to be able to model for
predicted results, then select the required torsional stiffness from a
list.


Yes, it is, and thanks for a pretty good initial list.

The greater question, by my way of reasoning, would be how much
additional mass (flywheel)......

You really want to say "inertia", not mass. There are two good
reasons. First, using accepted, correct terms greatly improves
discussion. The subject is complex enough without everyone inventing
or misusing terms on the fly, and drilling yourself on correct terms
will help you with correct thinking. Second, as a practical matter, it
possible to build two flywheels of equal mass moment of inertia, but
unequal mass.

or damping (viscous/hydraulic disk) on the PSRU side of the
clutch/spring assembly would provide enough additional smoothing to
justify the added weight.

In the case of the Suzuki drive, the experimental damper added about
5 lbs and lowered steady-throttle resonant vibratory torque from 180
ft-lbs to 115 ft-lbs. Some portion of that 5 lbs also served as
flywheel inertia, a nice design bonus. A torsional damper does not
operate "on the PSRU side of the spring". It operates in parallel with
the spring, or to be more precise, in parallel with a connecting
torsional stiffness.


Dan

This is mostly a nomenclature issue. In most areas of electronics, where I
worked until a few years ago, we would have regarded the flywheel portion as
capacitance (and therefore the flywheels are parallel) while the springs and
dampener would be modeled as inductance and a small resistance (in series
with the capacitance of the flywheels).

In short, you appear to have designed/developed a pretty good solution which
can also be scaled for more or less power. I will add your list of
manufacturers and nomenclature to my notes for future use.

In the meantime, is the system or software you designed/developed available
for sale/rent to the guys who are currently trying to build flyable
aircraft? If so, from whom?

Peter

I may have started typing, and pressed send a little too quickly, so let me
flesh the question out a little more...

Here is a simplified electrical equivalent circuit, in an effort to provide
a little more precision that words alone.

(Note: This will display correctly only if your newsreader is displaying
fixed width font--probably Courier New--so I will also add a version spaced
to display in a readable fashion using Arial)

___________ ________
| DC | | DC |
| Generator |---------\/\/\/\---OOOOO-----------------| Motor |
| (Engine) | | R1 L1 | | | (Prop) |
|___________| | | \ |________|
| | /
=C1 =C2 \R2
| | /
| | |

Where C1 is the engine flywheel
C2 is the flywheel mass of the hydraulic disk dampener
L1 is the dampening effect of the disk dampener
R1 is the heat energy loss of the disk dampener
R2 is the friction loss of the PSRU and other bearings



Note that a true electrical equivalent is extremely difficult to draw, even
when one does not consider the limitations of text as graphics. However, in
my opinion; the system would be much more effectively damped, as seen by the
propeler, in the case that the hydraulic disk dampener is placed after the
soft coupling--and this would be more important in the case of the pusher
with a long driveshaft, where propeller excitation by the disturbed
slipstream is a practical issue.

So, what did you actually do?



(Note: The following is re-spaced to be readable on Outlook Express. It is
not very good, but you can sort-of make it out.)
___________
________
| DC |
| DC |
| Generator |---------\/\/\/\---OOOOO-----------------| Motor |
| (Engine) | | R1 L1 | |
| (Prop) |
|___________| | | \
|________|
| |
/
=C1 =C2 \R2
| |
/
| |
|

Where C1 is the engine flywheel
C2 is the flywheel mass of the hydraulic disk dampener
L1 is the dampening effect of the disk dampener
R1 is the heat energy loss of the disk dampener
R2 is the friction loss of the PSRU and other bearings



Peter