Paul, try three. the export tab in autocad is where to save to many file
types. I chose autocad14(dfx.)  I clicked the file and it auto opened in
true-view2009 which must be installed on this computer. I tried rhino
version 1. The file symbol in upper left corner then select file type to
autocad14dxf from the list and it opened. the view " top" seems to be the
plan view and since this is a twoD working drawing that seems right. text
doesn't translate in other views. I didn't set the circle performance
numbers from the default and it is readable, just not as smooth on the
lines.What is left is the various cross sections up through the core
engine
in true scale. I will be extruding, revolving etc these into 3D solids
and
joining/filliting the solids for the solid model shortly.As usual no
rights
reserved,  anything helpful to anyone is available. this drawing has
most of
the mazda engine parts to scale as well as RWS  RD-1B mounting plate with
the coordinates in decimal degrees translated from polar coordinates from
the newsletter last month.The stock side housings were deleted off this
drawing and the rotor was never drawn. Also the e-shaft between the
eccentrics is approximate not to scale.The tension bolt locations were
measured by dial caliper and the expected tolerence is on here somewhere.
Enjoy oshkosh, I have family visiting and can't get there.

george grimes

The dxf file did not load. It appears it was not a vector dwg.
The error message was un supported data type raster image.

I am afraid the 3D world has moved away from autocad.

Paul Lamar

 Paul...

 This probably isn't any help as you've already identified it as a raster
 file, but I was able to open it in Irfanview.  Irfanview will attemp
 several different formats if it doesn't recognize the extension.

 The image appears to be a very low resolution shot of an upside down
 program...with the drawing, whatever it is, scattered around.  It
 appears to be a bmp or some such, as Irfanview had no trouble
 translating it.

 It may be helpful if it can be used to show George what he's sending...

 Harley

 Paul the drawings are twoD at this point because I need to  relate all the
 cooling passages, oil circuit , bolt locations  and mazda parts into as
 compact a size as I can get, and keep relocating and removing metal here
and
 there  to get the weight out of it without  violating the integrity of the
 separation of oil and water defined by possible o-ring locations.


Simultaneously the draft of the patterns both inside and outside  from top
 and bottom must allow the pattern to be pulled from the molding sand and
 avoid transitions in section modulus which could weeken the casting as it
 solidifys or produce hot tears during the metal solidification. For me it
is
 easier to look through all the layered sections in a stack leaving out
 layers not in the area being worked on and when the puzzel is worked out
 then just extrude the layers with the pattern draft and join the parts
into
 the solid.


The pattern solid can then have the machined surfaces removed
 with the slice command and bolt holes added to the remaining solid which
 will be the actual part. Then the g-code translater can machine the
finished
 parts. I get to the stage of a rendered solid but rarely in the design
 stage. I have tried working with the 4 panes in 3D but entering 3
cartesian
 coordinates and editing points from the 3 windows  with filters  is much
 slower and to me confusing.


Many times I don't have the third coordinate
 because there is not a finished part to measure or digitize. When I get to
 the rendered views like your sample I will send them along. I expect to be
 finished with the core engine in twelve weeks and started on the patterns
on
 my cnc router. making patterns is quick, figuring out how to run the
router
 is unexplored territory. Anyway, glad we got  something to
 transfer,



george-grimes


You need to try Rhino George. It does everything you want it to
do. http://www.rhino3d.com/ The user interface is very much like ACAD.

Here is an end housing I did in Rhino.

---------------------------------------------------------------------

Here are some jpgs I think you will find useful.


Paul Lamar


Paul, I saw your idea with the iron wear surface combined with an alloy
housing on your website about a year ago. Since it was not available to
purchase I ust started sketching, not to scale , just a sketch. As best I
can remember the B2jpg equivalent had a central web like an I-beam with two
loose side plates. I didn't come up with a sure way to isolate the water and
oil where the  center core( integral with the central web) contacted the
iron wear plates, and the ports both had to be periphial to avoid leaks from
intake and exhaust to the water jacket.Periphial is obviously not a problem,
inexpensive horsepower, and who doesn't like that.My idea on the core
sealing was to cast the two wear plates and core as a separate piece, then
put that in he mould while casting the rest in aluminum, possibly milling
some grooves in the contact areas with the aluminum. This has been done for
years in auto engine cylinders.


In the end however I didn't want to preclude
the use of side ports. Then with the RX8 parts being current production, and
therefore much cheaper, and the water seals moved back to the rotor housings
as with the earlier engine pre 1984? I decided on the two piece center
housing split down the center leaving the port passages  like an open fae
sandwich for cnc porting. This allows getting away from the 90 degree sharp
turn from the side housing to the rotor housing since I don't need a
straight path for the tooling through the port. Also I can relocate metal
around the port which is not available in a stock housing.

Picture 2jpg is really interesting to me . I had no idea that so
much space is available for the oil drainback inside the housing,but more
interesting is the connection from the oil fill through the center to the
pan.If I was looking to make a stock fuel injected engine aerobatic. A
fitting to the fill tube, one to the top of the acceessory case near the
counterweight and one section of a dry sump savange pump piped into the oil
cooler may take care of the oil return.

A swinging pickup on the pump could
be done without too much trouble. Then it's just bolt it upside down on the
engine stand and see how it does. My thinking is that most of the  rotor
cooling oil return has to be through the center housing because the e shaft
bearings and rotor to side housing fit  contain the rotor cooling.Also  half
of the pressure lube to the e shaft ends up on this side of the bearings,
only 1/2  the rear main  oil would have to be addressed to run it upside
down.The stock intake system may not be in the way if it was upside
down.Some way to let trapped air or steam out of the high point of the
waterjacket would also need attention. Might be an interesting project at
some time in the future.

I have watched other people use alternate programs like
Rhino,Catia, and solid works. I think solid works has some wonderful
features such as extruding an irregular shape along a 3 dimensional path
with a smooth transition to a different irregular shape within a single
command--- think about porting. It also allows something like a mechanical
linkage or device to run through it's motion rendered in 3D----think about
retractible landing gear.Unfortunately only larger machine shops and
industrial fabricators can afford the price of the base software. Then all
the add in packaces like computer aided machining and finite element
analysis literally bring the price to more than a new house costs here where
I live.

Its just not doable for an individual.We don't have any junior
college courses on cad here where I could use their software on a term
project for my ideas, and just be a perpetual student. I figure if engineers
in1940 could take celophane and india ink with a slide rule  and build the
rolls royce merlin, the daimler benz 600 , the 36 cylinder curtis wright
radials,and the 18 cylinder turbo-compound radials  and many more examples,
then build them with an engine lathe and milling machines which only cut
straight lines. Well then I am fortunate indeed for a workable cad progran
and shops with cnc everything willing to make the parts. My best hope for
now is some machine shop  being auctined off and buying  some of their
computers with software on the hard drives. Well back to reading the
newsletter posts.

george grimes
aerobatic