Looks like it is strong enough to take an 850 lb load on each of the engine
attach points. (You must be the judge, of course.)
Here are the FEA results:
Bill Dube'
Thanks Bill. Nice FEA program you have there. Good work. Does it run on
Windows? How much does it cost?
Paul Lamar
Bill if you don't mind and it is not too much work could you rerun
it with two 1/4 square key stock steel added to the top flanges 12 inches long
and centered? Consider the additional material homogeneous. Which would
be an approximation for welded on pieces of course.
I like a safety factor of two.
Paul Lamar
If you look carefully at the factor of safety plot and the stress
plot, you will notice that you are having more trouble on the ends than you
are in the middle.
I will try to find time to add the extra "meat" in the middle
tomorrow. It may be Monday before I get to it, however. I would scab the
sides with a large thin doubler rather than placing a concentrated thick
section at the top. You also would want a doubler in the side mount area. I
will try to do this.
I should note that the load used is a "maximum" type load that
would not happen with any regularity. Thus, fatigue is not a consideration,
so the SF can be less than 2. Also, you should be aware that I selected the
yield strength of 4130, rather than the ultimate strength, for the factor
of safety calculations. This is because the load is not truly a "one time"
crash load, but something the beam would experience multiple times. The
beam might bend (permanently deform) a little, but it is doubtful that it
would fail. If, instead, I had selected the ultimate strength, then the SF
would have been closer to 2.
About the FEA software <<<<<
All the software I use runs under Windows XP.
I used AutoDesk Inventor to build the solid model. I then used
ANSYS Design Space to do the structural simulation. Inventor costs about
$5000.
ANSYS costs about $6000 (per year.) The price depends on what
capabilities you buy. You can pay less and have fewer modules (or work with
a limited model size) or you can pay more (like $12000) and get everything
they make. In addition, you really need a $3000 or $4000 computer to run
ANSYS on geometries of any complexity.
Here is a link to ANSYS DesignSpace
http://www.wildefea.co.uk/software/software3/
Here is a link to Inventor
http://usa.autodesk.com/adsk/servlet/index?siteID=123112&id=4291647
If you are a student or a faculty member, you can get both of
these packages for a fraction of the commercial price. Probably worth the
cost of tuition in your local community college just to get the price
break. Also, you can't just drop the geometry into ANSYS and get valid
results. You need to take a few classes in FEA to understand how the
software works inside so you can make intelligent choices on meshing and
boundary conditions. Thus, enrolling in your local college is not a bad
idea if you want to do FEA.
Bill Dube'
Thanks Bill. What I am worried about is buckling where the flange
is cut away and bent over near the center. At a compressive stress
level of 55,000 psi buckling cannot be too far away. The 1/4 square
key stock will serve to stiffen the flange.
As far as yielding at the rubber bushing ends IMHO that is not very
important. A bit of yielding out there won't hurt a thing.
Hard to predict that anyway because all the load is fed into
the beam through the rubber bushings. There is a 1/2 inch thick
2.25 inch by 2.25 inch alloy plate out there that can be screwed or
riveted to the bottom of the beam if necessary.
What method did you use to apply the rubber bushing loads?
Paul Lamar
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
Copyright 1998-2005 All world wide rights reserved.
