> I am reposting my last msg from a couple days ago;
it
> seems to have been lost.
>
>
> Rolf, Bill and Kenneth
> Thought I would give one more try on this subject.
If
> it is possible to operate lobes and journals with
a
> thin wall of .06" min, then the min dia of the
spline
> approaches 1.47" [1.69-2(.06)-.098] which meets
Rolfs
> stress criteria at 415#', without the torque
ripple
> factor. My question is simple; is this wall too
thin?
>
>
> Kenneth
> Have you addressed the rather large tapped holes
> required for the oil squirt jets? Seems they would
> take away a lot of the shaft cross section area.
>
> regards, Robert
>
Wait a minute...
Is it possible that you meant to spline the
eccentric lobe onto the
e-shaft?
Is that what you had in mind???
If so, please don't even try. It will not work.
In order to make the narrow side of the lobe strong
enough, the e-shaft
will be too weak to carry the load.
You have now two members that each must carry the
same torque within the
same space restraints.
Don't forget there is also bending on the e-shaft,
not just torque. The
highest bending stress is under the lobe.
Rolf
Rolf
That is certainly my understanding from the initial
description. However Kenneth has pointed out that the
eshaft is supported on each side of the lobe by
bearings. It would seem to me that the bending forces
would appear to the short supported eshaft section as
shear forces. Disregarding the torsional forces for
the sake of discussion.
This brings me to ask you about a comparison of axle
shafts vs this splined eshaft in particular. Axles of
similar size to this proposed eshaft will handle large
amounts of torque, but it is my understanding they do
so by allowing windup. This would be undesirable for
any eshaft. So does the stress criteria you stated
earlier apply directly to axles? Or is it an entirely
different situation to allow windup?
regards, Robert
An axle will not accept any more torque due to its
length compared to any other shaft. Angular deflection in
a long shaft has nothing to do with maximum stress. The windup
stores torsional shock loads and returns them to the system when
the shock load is over with. Peak torque is peak torque.
http://library.wolfram.com/examples/Problem3/
The worst feature of this design is the stress concentrations
in the root of the splines. That leads to cracks and fatique.
But by all means build it and test it. I suspect it will break
long before you get around to flying it.
Paul Lamar
Robert,
My comments were meant to raise awareness and concerns.
By all means, build it and see what happens.
Technically, you are talking many different things here.
Looking at shear, the load of the piston at maximum gas pressure is near
27,000 to 32,000 lbs.
245.5 cm^2 * 50 to 60 bar = 12275 to 14700 kg * 2.205 = 27,060 to 32,480
lbs total.
At a high charging rate it is even more.
In the case of the e-shaft, we have combined stresses of shear, bending
and torque.
As a good approximation, two forces are added quadratically, meaning
sqrt ( A^2 + B^2)
Or in this case sqrt ( A^2 + B^2 + C^2 ). But that is not necessarily
correct.
To be precise an analysis of the time and angle of these three forces
must be made. Not that easy.
Regarding storing energy through wind up, yes that is occurring.
Heat will be introduced and most importantly, the stiffness or rather
self frequency of the shaft must be known
together with the input cycle of the forces. But this applies to longer
shafts only.
Axle shafts just carry torque, no bending and no shear. Different
situation to an e-shaft.
The e-shaft when weakened as proposed, should also be checked for its
own frequency.
Being much weaker, more deflection is being introduced which is a key
enemy of the Wankel engine.
Any vibration may lead to the so called chatter marks. That may be
encountered with a weak shaft.
Enough said, try it anyway.
Regards
Rolf
PS. For 2000 inlbs of torque I get 377 psi for the example you send. He
calculates 296 psi.
The low number is unrealistic for steel or practical engineering. The
shaft may be wood or plastic.
The length is irrelevant if no material and its G-module is given.
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
Copyright 1998-2006 All world wide rights reserved.