Bill Freeman wrote:

          BIG SNIP

BTW I was so impressed with JB Weld as used on my water pump that I think
we can use it to hold the wiggins couplings on the coolant system aluminum

tube

plumping. The coupling just needs enough overlap for a safety factor of

three or

four.

The JB Weld bond is claimed to be good for about 3000 psi and 600 degrees F.
If the tube is 1.5 inch OD and the length of overlap .25 inches we will
have a bond area of 1.178 square inches or 3500 pound pull apart load.
We could easily make the bond area .5 inch wide so that would be a
7000 pound pull apart load. Sounds good to me and easier than welding
or swaging.

Paul Lamar

Wait a minute.  Barb hose fittings with double hose clamps with safety
wire is dangerous and fancy glue is OK for critical coolant lines???  Yikes.
I agree on AN fittings being the best, and would not use hose barbs for
critical lines in an A/C engine compartment, but GLUE?  Going to the
trouble of a wiggins coupling and gluing it on?

Not with my butt in the seat . This strikes me as unsafe.  Gluing coolant lines
together regardless of ANY specs on the glue will never make it in
any engine of mine.   Call me old fashioned but I'd run a stock pump or
a welded one, but never glued up one.

I have recommended JB Weld for plastic radiator tank repairs and it has
worked *extremely* well with glass cloth reinforcement (Mazda 929) but no
way for a critical aircraft engine system.  The bead fractured off the inlet
neck about 18 months after the top tank crack, still causing a cooling
failure.
 The plastic was deteriorated,  but the JB Weld/glass still held.    However -
not on my airplane.  Might last forever, might not.  I'm not that brave.

One coward's opinion.

Bill

The trouble with hose clamps is they are steel and barbs are aluminum.
When the barb expands it permanently squishes the rubber hose. When the barb 
cools off it leaks or blows the hose. Spring type OEM hose clamp expand and contract
as needed but nobody considers those.

I was thinking about doing a bond test today while I was driving back from Riverside.
I was thinking of gluing two pieces of aluminum together with about 
0.2 square inches of bond area. I would then boil it for an hour and test the bond
for tensile strength. I would also test a non boiled sample. I could also put it in
my industrial oven and cook it a 250 for 24 hours and then test it again. 

Paul Lamar
 
The Aircraft Rotary Engine Newsletter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org