Subject: exhaust problems
From: Rotary Engine
Date: 8/25/2006, 10:53 PM
To: AA-me



    > >     Yep, that's who's design I copied - I even bought my 6"
   inconel tube
    > >     from
    > >     him because he had some left over from his muffler.  The
   welder used
    > > the
    > >     inconel welding rod that I bought. So I know it was the
   right stuff.
    > >
    > >     The exhaust flange joint is welded both on the outside
   joint and on
    > > the
    > >     inside joint. There is no possibility of weld failure there
   - the
    > pipe
    > >     itself would have to tear/crack.
    > >
    > >     Like I said - Inconel doesn't live long when there is
   flexing. So,
    > > maybe
    > >     Atkins has his muffler supported and mine wasn't - although
   I've
    > >     seen his a
    > >     few times and didn't note any supports close to the
   can.  It does
    > >     look like
    > >     he has a tab for support down the line a ways - that may
   damp out
    > > enough
    > >     vibration to keep the flexing under control. For sure
   something was
    > >     different!
    > >
    > >     Joe Hull
    > >     Cozy Mk-IV N31CZ (65 hrs - Rotary 13B NA)
    > >     Redmond (Seattle), Washington
    > >
    > >     -----------------------------------------------------
    > >
    > >     The exhaust always breaks just next to the weld for some
   strange
    > >     reason so the flange actually supports the tube.
    > >
    > >     Here is some more info from Marks Wrathal site on welding
    > >     tube to exhaust flanges. Many many racers do it all the time.
    > >
    > >     http://www.wankel.org/
    > >
    > >     " Weld only the engine side of the flanges. Four spot wells
   on the
    > >     header side seems to be the way to go. Belt sand the flange
   faces
    > >     then lap them on a piece of glass and a piece of oiled wet or
    > >     dry paper."
    > >
    > >     "Archie backed up this crucial point as follows:"
    > >
    > >     "In hard core racing, the same problems (as occurred on N214KW)
    > >     arose a number of years ago, mostly in funny cars, but
   dragsters
    > >     also. The common denominator, other than various grades of
    > stainless,
    > >     was outside welds. Now you may see some tack welds on the
   outside,
    > >     but the actual welding is done on the backside.(inside) Also in
    > >     current use is a tube within a tube. I noticed the weld
   bead was
    > >     quite wide, increasing the hazard, and may have contributed
   to the
    > >     failure."
    > >
    > >     Archie builds 7000 HP AA drag engines for a living and has
   several
    > >     forums at OSH.
    > >
    > >     This last trick is expensive but Everett Hatch came up with the
    > >     idea. A ball flex joint is used right next to the block.
    > >
    > >     Paul Lamar ...No rotor no motor.
    > >
    > >
    > > Paul and Joe,
    > > The more I learn about Everett Hatch the more impressed I am.
   The ball
    > > joint he uses would almost double as an anti-reversion cone.
   Besides
    > > relieving the pipe to block stresses note that the pipes just down
    > > stream also have inside slip fits. Very nice work.
    > > Bill Jepson
    > >
    >
    > Were the welds heat normalized afterwards??
    >
    > Gary Buster
    >
    >

   I don't know if mine were heat normalized. Does that help the
   surrounding
   metal? Because again, it wasn't the welds that broke - just the
   inconel all
   around the joints.

   Joe Hull
   Cozy Mk-IV N31CZ (65 hrs - Rotary 13B NA)
   Redmond (Seattle), Washington


I don't know if it's applicable here, but I have heard that stainless is
welded, the back side of the weld needs to be in an O2 purged
environment.  I wonder if/how that's done when the inside of the pipe is
being welded to a flange..


Matt-

That is true for tube butt joints such as headers. You need a
separate supply of nitrogen gas behind the weld. Not so for a
flange welded on the end of a tube as the slag, if any, can be
mechanically removed.

Paul Lamar ...No rotor no motor.

The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Copyright 1998-2006 All world wide rights reserved.