Subject: 3 rotor renesis....It lives!!!
From: Rotary Engine
Date: 7/21/2008, 3:08 PM
To: AAA Put this in the To box



    Hey all,

         Came across this reading up on the rx7club.com and rx8club.com website.
         Here is the link. This is exactly what I want to do with my car but with
         lightweight housings. It would put me at the power and weight level I
         would need for my platform. I am sure Paul will rave about doing a
         P-Port which would make life easier in the long run and I will consider
         it when I feel it there is supporting data to back the claim. (Note: Not
         trying to be mean Paul just stating what I did before because I know
         what your response would be.) Its worth getting on and seeing some of
         the conversation. It brings up an important question I had about port
         balance and the engine.  I know why they have different
         shape/flow/timing ports but for a performance automotive aspect and for
         that matter a constant rpm aerospace aspect such as your it would seem
         to me that you would want a symmetric port configuration. OEM
         driveability seems to be of little use to us even if we have different
         platforms to put them in. I I am going to port a lengthy response to the
         incredible information that Steve posted two days ago and this will be
         part of the reference to what I am talking about. Just though you guys
         would like to see a working, proven setup even though it is still in its
         tuning stage. I heard 260lbs bare block for a 20B. Lightweight housings
         should put this at about 210. With all accessories and such we can say
         300lbs would be reasonable foe a N/A three rotor with I imagine around
         375hp and 225 ft-lbs. For you plane guys as long as you could fit the
         length then it would be a wash with a two rotor turbo weight wise and
         you would have better reliability over a turbo setup and with less cost.

         http://www.rx8club.com/showthread.php?t=141000

         http://www.youtube.com/watch?v=JZLFUCqdMZ4

         http://www.youtube.com/watch?v=40Hohpsjq2E

         Rob Woods
         585-615-8474


         That engine would be lucky to see 300 HP :)

         Paul Lamar ...No rotor no motor.


         So I assume this engine would have to have a custom eccentric shaft, as it
         doesn't have the thick center housing of the factory 20B. It would be
         interesting to see a picture of the shaft.
         Larry

    Larry,



    The guy who did the engine isnt really forthcoming because the people
    on the message board did stupid messageboard things and I think he got
    gunshy about putting info out due to being hasseled. I am going to email
    him soon to ask some legitimate questions and that was on my list so
    hopefully I can get some out of him.


    If I did the shaft I would do a three piece shaft. The first piece would
    be the shaft and the lobe for the center lobe. The second and thrid
    piece would slide onto the ends of the shaft. By doing this you can get
    a bearing, journal or anti-friction, on either side of a lobe to make
    one hell of a rigid setup. Also I never really liked that tapered
    section and keystock method. That really doesnt induce much rigidity
    into the shaft assembly because you are not engaging the slider lob as
    much as possible. I dont have one in front of me to see if there is room
    to do a full length taper but I do know that taper fits work best in
    terms of assembled stiffness when you engage as much as possible. I
    would even go as far as use a taper 3 lobe stoffel spline and toss the
    key all together in order to increase the center lob shaft thickness. I
    used stoffel splines on my formula car and they work great. Neat thing
    is with a 3 lobe spline you could orientate the lobes to make a even
    fire 3 rotor or a "big-bang" 3 rotor. Now that would be interesting
    to
    test. As far as the centerplates and bearings are concerned I sent a pic
    of that stationary gear 2 weeks ago that showed the modified stationary
    gear in the plate. Do that for the first plate and for the second plate
    do the same but just cut the gear off. Then you would have a journal
    bearing on each side of each lobe.

    Rob Woods

    The bottom of the eccentric lobe is the same as one side of the shaft.
    If you make a 3D dwg of the Stoffel arrangement that will become apparent.

    I think these are pictures of the Guru 3 rotor shaft using the thin center
    housing.
    Stoffel splines are unnecessary. A taper and a key way with a long tension bolt
    holding them together works. Stoffel splines are expensive to machine.
    http://stoffelpolygon.thomaswebs.net/ They also allow sliding in and out which
    is not what one wants in this application.

    Paul Lamar

    The Rotary Engine NewsLetter. Powered by Linux.
    ACRE NL web site. http://www.rotaryeng.net
    Youtube key word UTUBPLEASE
    Copyright 1998-2008 All world wide rights reserved.

Paul,



I understand that the lobe is tangent to the 1.5"-ish bearing diameter.
The stoffel splines dont slide anymore than an ID slipping over and OD
such as a Guru shaft. They both require the same tension bolt to clamp
the pieces together. Stoffel floater splines (P4) are only used in
transmission like sliding setups or where the size dictates. I dont
understand your comment about making a 3D drawing of the stoffel but
like I said before I did use one on my race car with great results of
which I did have a 3d model of it. The point I was getting at about
using them is this. Everyone comments about shaft stiffness being very
important in order to reduce deflection that causes all sorts of
problems. As you know the only way to do this is to increase section
width. Also slide fits have horrible stiffness because to all of the
takeup. Put a shaft inside a shaft and put in a deflection rig and
you'll be suprised how much it will deflect. By not have homegenous
material and less than long taper you give up alot of stiffness. So on
the guru shaft the front snout will undoubtly bend more than you would
want it to.



I assume that they made their shaft out of something that can be
hardened so they will need a heat treat and post grind. The shaft on our
formula car was 4340 that was premachined, heatreated to 45RC and final
ground spline included. The first one we did we ground but the second
time we did one of our guys took to his co-op and we machined the whole
thing with no grinding since the shop had the machines and tooling to do
a single op machining up to 60RC steels. The second one turned out
better to be honest and didnt have to worry about localized heating from
grinding and turned out the part faster. Setup costs in grinding is
where the money is and not the time to grind really and if you can avoid
it then better for you. Then again I dont know what surface finishes are
capable and/or required so that might not apply here.  If you look at
the Guru shaft you can see discoleration on the lightening hole side
which looks like grinding "burn" to me.



I am not saying that the stoffel would be best but I am saying there are
other options. I guess you could reduce the taper angle and let the
taper do the work it should. I personally hate keys, straight or
woodruff, and if a mechanical locking fit can be used that attributes to
higher stifness and strength for not too much more money then I would go
for that. Comes down to machine time vs. job operations(machine
switching) and these days with the equipment out there it tends to be
cheaper to minimize ops. So Stoffels arent really that bad price wise if
engineered and machined right but they allow for much stronger
mechanical connections then the old school ways of doing it. Most turbo
compressors uses stoffels and hirth couplings to hold the units together
on the main shaft. In fact Cameron Compressor does this on almost all of
their units. One of my sponsors does their machining. I find it funny
that mazda tried a curvic coupler and said it sucked but then again
everyone uses them now on high speed turbo machinery.  They needed to
add an additional axial constraint was their downfall. Grannys speedshop
double 13b engine went through a complicated method of attachment
instead using a taper stoffel connection but I am sure it is because of
the garage nature of what he was doing. What he did is what is common in
industry right now.

http://www.amcprecision.com/index.html

Not saying it is the best way but it is a possible way. I would lean
toward a 3 piece shaft with the center one have taper stoffel splined
ends and the end pieces with the female connection which would allow
them to have whatever end diameter they wanted and then a retention bolt
to hold them together and a bearing on each side of the three lobes.
Stiff as shit and supported like hell.


Rob Woods

Do a sketch of what you have in mind and I will do a 3D when I get back from
OSH in about a week.

The main problem with e-shaft bending deflection in the two rotor is there is
no center main bearing. However there are  ways of dealing with that. Even with
a one piece shaft. Here is one of them. Curtis Wright kept this a secret until
somebody got one of their engines and took it apart 50 years later. Notice it is not
shown in the side view dwg. They were holding out on Mazda :)
-- 
Paul Lamar ...No rotor no motor.

The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Youtube key word UTUBPLEASE
Copyright 1998-2008 All world wide rights reserved.