" You could always make that coupon sample I was describing.
Put it in the freezer for 24 hours. Take it out of the freezer
and bake it in the oven for 2 hours at 200 F and then repeat
500 times. Get back to me with the actual count :)"
"There is zero tensile load on the tube."
Notice I did not say anything about applying a load to the
sample. So boil it in antifreeze solution instead of sticking
it in the oven. The intent was to check the seal integrity and
not the strength.
BTW the radiator tubes you may have in your car are more likely
sealed with epoxy to the header flanges.
Sorry I should have been clearer. The bond to the aluminum was
exactly what I was talking about. If the bond fails it starts
leaking. I was not purposely side stepping the issue.
I am not against a press fit it is just difficult for the
average builder to accomplish. First the hole must be bored
to the EXACT size in the rotor housings +/- .001 and then
the aluminum trimmed to fit the inner trochoid surface exactly.
Not easy to do for the average fabricator.
In the beginning we used an NC punch to make all the slide
throttle parts. The flanges and all plates were punched out
square along with all the bolt holes. 3/16th thick with a 2 inch
hole was at the limits of the NC punch. The intent was to
mass produce the parts and keep the cost down.
If the tubes are cut first with the flange welded on them then
both flanges must be aligned so the holes in the flanges fit
the back plate of the slide throttle and the cut on the end
of the tube fits both trochoid surfaces. Are you going to
heat both rotor housings, cool the tubes and press fit
both tubes in simultaneously? Good luck.
That is why I used a slip fit for the tubes with square ends.
We placed both rotor housings side by side, inserted the tubes
with the flanges welded to them and bolted up the slide
throttle back plate to align the holes in the flanges
with the slide throttle back plate. The exact angular
position of the tubes was then marked on each rotor
housing. The back plate unbolted and the tubes bonded
in.
Think it through to final assemble. Perhaps you can
make a change in the design of the slide throttle that
can accommodate press fits with this alignment problem.
Sorry I did not explain all this stuff again when
the subject came up again. I assumed everybody had this
in their mind when I explained all this stuff way back when.
I'll dig it up. I will also add it to:
http://www.rotaryeng.net/fund-of-intake.txt
Paul Lamar ...No rotor no motor.
Paul & Doug ,
Ye Gads - what have I started, I was only trying to be helpful, as Doug is
planning some serious long distance flying and I didn't want to see him
POSSIBLY stuck in a strange country with a POSSIBLE leak of this kind.
Paul you are definitely right, this is much more than the average builder
can accomplish and indeed it may be terribly overkill.
However Doug has the capacity and the desire to try this method and I
couldn't think of anyone better to be able to accomplish it.
I was just looking at the outer housing and it's mighty thin perhaps only
4mm around some of the cast depressions - there may not be enough room for a
outer taper even a small one. Paul's suggested method of welding this may be
the answer but as Paul also suggested, there is a concern with warping. I
think the answer here is to have the outer tube tig welded ( very slowly,
i.e. a bit at a time). I am only thinking out loud here guys, so until I get
some more answers from my welder expert, I will leave it there.
My apologies - we're all trying our best!
George (down under)
--------------------------------------------
There is no question in my mind that this will NOT fail
in a mechanical sense. Therefore welding it won't help.
Yes you might someday develop a pin
hole leak but that is not catastrophic. You will just wonder
where the coolant is going. Worst case you don't have a coolant
level sensor installed and your engine starts over heating.
If I were flying anything for five hours over water short
of an airliner I would be very nervous. In that case I would have
a couple of gallons of spare coolant I could high pressure
pump into the engine a little at a time :)
Using the technique and tools I just outlined you could fix it
anywhere. Carry some spare JB Weld.
Paul Lamar ...No rotor no motor.
Paul, everyone,
We are on a three day holiday now but the shop owner did confirm yesterday
a press fit is possible. I want to double check that I heard him correctly
but I think he said only the very edge of the insert pipe steps need to be
tapered slightly to aid in alignment but the actual interface is not
tapered. Think about gears and bearings pressed on shafts.
Both need different amounts of force to press them on because they each have
different amounts of operational forces trying to pull them off. A bearing
needs a small friction to hold it in place compared to a pressed on helical
gear.
The insert force in trying to loosen it is mainly heating cooling cycles.
The main purpose a tight press fit will accomplish is a tight water seal.
In short the inner housing would have a seat (like you advocate but the
press interface would be between the pipe and the inner housing where it
passes through.
The outer housing would also have a step bore but one of a larger diameter
then the inner one. Put some silicon sealer on the inserts or epoxy on the
seats if you want but it will scraped off by the tight fit. Put JB Weld in
the whole cavity as well if you like. Tape in a thin stainless shield to
protect the inner liner and grind the protruding aluminum insert down.
Finish it with a slight radius to make sure the aluminum insert is below the
stainless housing.
Yes this work needs a good end mill and a digital counter is nice but it the
mill doesn't need to be CNC. It is not that hard to do this or have it done.
You just need accurate measurements. Heat and freeze the appropriate parts
have them pressed in with a hydraulic press. After its in it ain't going
anywhere:) If you like you can still pot the pipe with JB Weld. I
personally suspect Mazda wanted to keep the intake port a little warm to
keep fuel in a vapor form.
Yes I absolutely agree that this is not the cheapest or simplest way to do
it. Hold on it gets worse:)
I wish to have a thick round flange up next to the housing on the end of the
inserted intake pipe with a seat for an O- ring on the face and a step to
align the pipe flange that will be bolted to it. The port flange needs to
have four studs taped into it (they could go into the housing) which would
be positioned and drilled after the insert is pressed in. I would agree
that the flange could be TIG welded on a thick wall pie and the lathe work
done after the welding to true up the face and create the proper step
diameters.
Why a flange at the port opening? For the test engine I suspect I will not
get the perfect manifold configuration on the first try. I may have to try
different set-ups even though in operation they will all be slide manifolds.
After the first engine runs and I know my preferred lengths I can go back
and adjust the pipe design to place the pleumn in the most space saving
location. I think most of cowls have very limited space on top.
For instance I could bolt on a short or longer extension and orient the
slide throttle in a vertical position or as a design that has some merits,
bolt on a cast aluminum elbow which orients the slide throttle in a
horizontal attitude. The reason I would consider the latter is that
injector shoot straight in and can be enclosed in a rectangular box fairly
easily.
I know this is difficult to catch what I am saying so I am working on a
Rhino drawing right now. I still have trouble making a bent manifold pipe.
Doug Fir
--------------
It is very very risky pressing anything into a rotor housing
or welding on it.
The exact shape of that wear surface is critical to the functioning
of the apex seal. The slightest amount of distortion will
cause the apex seal to fail to seal. I am talking .001 inch here.
This has been in the back of my mind all along. It was whispering
in my ear and I finally heard it. I personally would never try it.
I am close to a simple fix. I'll upload something later to
day.
Paul Lamar
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
Copyright 1998-2006 All world wide rights reserved.