AIXRO had a customer CAD step file on their web site so I downloaded
         it into Rhino. Here are a few 3D's. Still need a rotor
         and a rotor housing.

         Paul Lamar

    I noticed that that engine uses a caged needle roller bearing for the
    rotor bearing. I understand that they have to since it is a charge
    cooled engine but if it works for them running up to 10k rpm then maybe
    a similar setup could be used for a mazda type engine. There shouldnt be
    any "weirdness" in the bearing such as you described earlier posts
    since
    the cage holds it together. For that matter they are used on every 2
    stroke connecting rod out there and nave no problems even with high
    rpms. It would seem to me that there are benefits to having a
    antifriction bearing vs. a journal bearing in terms of assembled
    stiffness, oil supply issues and reduced friction. The only downsides I
    could see would be cost and surface hardness requirements on the needle
    to rotor since it is cast iron.

    Rob Woods

    That would be a big mistake. Those bearings are the Achilles heel of the
    charged
    cooled rotary engine. The charged cooled rotor and the e-shaft run way too hot
    for those bearings. They are right at the life limit at 10,000 engine RPM.
    Remember, they have to really zing RPM wise. If it were not for the fact that
    the rotor was also turning on the e-shaft they would not work. As it is they
    are
    turning only 6000 RPM which is the absolute upper limit for that size roller
    bearings.

    Paul Lamar


Wrong type of bearing to use.

http://www.timken.com/en-us/products/Documents/Needle%20Roller%20Bearings.pdf

Look on page C19-20 and you will find the 65mm stuff. Notice that these
weigh far less than the type you pointed out. A couple other things to
consider. Assuming the e-lobe is 65mm OD, which I think it is, you are
constrained by the ID of the rotor bearing housing. A caged needle
roller bearing is thicker than the stock journal bearing so you would
need to actually reduce the OD of the e-lobe in order to fit a bearing
in there. The catalog says the shaft has to be 58RC to run on which isnt
hard to do but the cast iron rotor would not be able to take that.
Therefore you would have to insert a hardedened sleeve into the rotor
which furthur reduced the available diameters. This would most likely
put you in the 55mm-60mm e-lobe diameter range which has oiled rpm
ratings 6500 to 7500 rpm. Also these are really light. You might need to
put two of them side by side if the full width of the rotor journal
bearing is wider than 1.25 inches. I know there are arguments this way
and that way but with light bearings of a proper rpm range will be much
stiffer, much less suseptible to damage/failure and less friction which
would tax the oil system less.

Rob Woods


You put me through a lot of work for 500 RPM on the RPM limit :)
Next time you do the cut and pasting.

You cannot assemble the engine if the rotor bearing diameter is not tangential with the e-shaft diameter. With a given eccentric radius there is no choice of rotor bearing dimensions. Unlike a piston engine and crankshaft there is a fixed
ratio for these rotary engine dimensions.

What will be "stiffer"?

The e-shaft or rotor will not be stiffer. Both the AIXRO and the Rotamax use caged rollers with outer rings. That helps reduce the heat transfer into the roller bearings from the hot rotor.

Paul Lamar