I was the OP in the Jan 26 thread "Any tips for finding air compressor
leak?". I'm here to report that I've had complete success ... YES!

One by one, I removed the potential leak-ers: the pressure guage, the
drain plumbing, the quick release outlet, the pressure switch, the
pressure relief valve and the Christmas tree (manifold) itself. As each
item was removed, I plugged it's port and checked for continued leakage.
If the leak had stopped, I would know that piece was the problem. I
also rigged pressure to that item and held it under water, looking for
bubbles. And with them all off, I checked the tank alone.

It was the last piece - the manifold itself that was leaking. Well not
itself exactly - it was the check valve built into it inside the tank.
I had soap bubble tested for this leak without finding it. Poor
technique, I guess. The valve was just a spring loaded disc against a
seat. The disc was phenolic like, so I tried making ones of nylon and
soft rubber. They both leaked worse than the original.

So I cut off the stub that the valve was built on and drilled and tapped
an NPT hole for a real check valve, which I got from McMaster-Carr. I
reinstalled everything, pumped it up, and 2 days later there has been
ZERO drop in pressure!

Thanks for all the responses. The posters who said "It's the check
valve" were right.

Bob

Bob Engelhardt wrote:

It was the last piece - the manifold itself that was leaking.

It's always the last piece.
The explanation is very simple:
You stop searching when you have found the cause! :-)


Nick
--
Motor Modelle // Engine Models
http://www.motor-manufaktur.de
DIY-DRO - YADRO - Eigenbau-Digitalanzeige

Yeah, but sometimes, like the one just described, it really IS the last
piece. I was rebuilding a Austin 1100 gearbox years ago (was stuck in two
gears at once -- something wrong with the shifter interlocks). It turned
out to be a loose setscrew (British: "grub screw") that held a shift fork to
its rod, allowing the fork to move on the rod. To get at it and tighten it,
absolutely EVERTHING else had to come out of that gear case first. The
shift rod for that fork was was the very last thing you could remove. Once
it was out, it was a simple matter of tightening the screw, and then putting
the whole damn thing back together.

--
Bob (Chief Pilot, White Knuckle Airways)


""Nick Müller"" wrote in message
...
Bob Engelhardt wrote:

It was the last piece - the manifold itself that was leaking.

It's always the last piece.
The explanation is very simple:
You stop searching when you have found the cause! :-)


Nick
--
Motor Modelle // Engine Models
http://www.motor-manufaktur.de
DIY-DRO - YADRO - Eigenbau-Digitalanzeige

In article , Bob Chilcoat says...

Yeah, but sometimes, like the one just described, it really IS the last
piece. I was rebuilding a Austin 1100 gearbox years ago (was stuck in two
gears at once -- something wrong with the shifter interlocks). It turned
out to be a loose setscrew (British: "grub screw") that held a shift fork to
its rod, allowing the fork to move on the rod. To get at it and tighten it,
absolutely EVERTHING else had to come out of that gear case first. The
shift rod for that fork was was the very last thing you could remove. Once
it was out, it was a simple matter of tightening the screw, and then putting
the whole damn thing back together.

LOL. So they started by assembling the gearbox: "place shifter fork
on shift shaft no. 1."

This is one reason those setscrews should be locktited in place, which
is what I did when assembling my neighbor's jeep gearbox, using red
locktite. I thought it might have been a JC maneuver, until we both
realized that the aftermarket shafts he had bought were machined incorrectly,
allowing one of the forks to overtravel and have mild gear interference.

He purchased NOS shafts, and after removing the gearbox from the car,
I realized that the setscrew was in there like it was welded in place.

I had to apply the flame of an air-acetylene torch to the head of the
screw, whereupon it was trivial to back it out. I felt a lot better about
the locktite option when re-assembling it the second time.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

Yeah, red Loctite is overkill. IIRC, Harley Davidson was assembling
slip-fit gears onto smooth shafts with it without the need for any splines
or interference fit. Just cleaned everything up, put red Loctite on both
parts, and slipped the gear on the shaft. Was effectively a permanent bond.
I presume they're still doing it that way, but don't know for sure.

--
Bob (Chief Pilot, White Knuckle Airways)


"jim rozen" wrote in message
...
In article , Bob Chilcoat says...

LOL. So they started by assembling the gearbox: "place shifter fork
on shift shaft no. 1."

This is one reason those setscrews should be locktited in place, which
is what I did when assembling my neighbor's jeep gearbox, using red
locktite. I thought it might have been a JC maneuver, until we both
realized that the aftermarket shafts he had bought were machined
incorrectly,
allowing one of the forks to overtravel and have mild gear interference.

He purchased NOS shafts, and after removing the gearbox from the car,
I realized that the setscrew was in there like it was welded in place.

I had to apply the flame of an air-acetylene torch to the head of the
screw, whereupon it was trivial to back it out. I felt a lot better about
the locktite option when re-assembling it the second time.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

On Mon, 6 Feb 2006 18:05:35 -0500, "Bob Chilcoat"
wrote:

Yeah, red Loctite is overkill. IIRC, Harley Davidson was assembling
slip-fit gears onto smooth shafts with it without the need for any splines
or interference fit. Just cleaned everything up, put red Loctite on both
parts, and slipped the gear on the shaft. Was effectively a permanent bond.
I presume they're still doing it that way, but don't know for sure.
Was it really the red? That's for threads. I'd think it would be 609.
The green forever Loctite. For cylindrical fits.
ERS

According to Bob Chilcoat :
Yeah, but sometimes, like the one just described, it really IS the last
piece. I was rebuilding a Austin 1100 gearbox years ago (was stuck in two
gears at once -- something wrong with the shifter interlocks). It turned
out to be a loose setscrew (British: "grub screw") that held a shift fork to
its rod, allowing the fork to move on the rod. To get at it and tighten it,
absolutely EVERTHING else had to come out of that gear case first. The
shift rod for that fork was was the very last thing you could remove. Once
it was out, it was a simple matter of tightening the screw, and then putting
the whole damn thing back together.

IIRC, the fork setscrews for the MGA transmissions were
square-headed, and drilled for safety wire -- as was a flange on the
fork to accept the other end of the safety wire.

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

No safety wire on the one I was in. Headed (can't now remember if square or
hex) machine bolts that just screwed in against the shaft. The one I needed
to get to was aimed toward the bottom wall of the casing, and completely
inaccessible. As far as I could tell, the only thing keeping it tight was
the proper amount of torque. The MGA you refer to probably used gear oil as
a lube. The 1100 gearbox (like the Mini's) ran in engine oil since it
doubled as the sump for the engine. I don't know if that would make a
difference to screw retention.

--
Bob (Chief Pilot, White Knuckle Airways)


"DoN. Nichols" wrote in message
ervers.com...
According to Bob Chilcoat :
Yeah, but sometimes, like the one just described, it really IS the last
piece. I was rebuilding a Austin 1100 gearbox years ago (was stuck in
two
gears at once -- something wrong with the shifter interlocks). It turned
out to be a loose setscrew (British: "grub screw") that held a shift fork
to
its rod, allowing the fork to move on the rod. To get at it and tighten
it,
absolutely EVERTHING else had to come out of that gear case first. The
shift rod for that fork was was the very last thing you could remove.
Once
it was out, it was a simple matter of tightening the screw, and then
putting
the whole damn thing back together.

IIRC, the fork setscrews for the MGA transmissions were
square-headed, and drilled for safety wire -- as was a flange on the
fork to accept the other end of the safety wire.

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

In article rs.com, DoN.
Nichols says...

IIRC, the fork setscrews for the MGA transmissions were
square-headed, and drilled for safety wire -- as was a flange on the
fork to accept the other end of the safety wire.

That's a nice feature. Unfortunately ford jeep transmissions
use simple slotted setscrews, with the heads below the fork
surface once they're tight. I think the theory was the jeeps
in question would wind up destroyed before the gearboxes came
apart from vibration. g

But I was interested in a long-term restoration, so I went with
the red locktite, and I'm glad I did. If those shift folks
come adrift, really bad things can happen, and the T84 gearcases
are terribly fragile.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

I work on a lot of Minis. Your description of the fork bolt is
accurate. They are 7/16" hex. You can get to them without removing the
engine from the gearbox. You have to remove the drivetrain from the
car and take off the diff housing. Then you need at least two 7/16"
box end wrenches that are clocked slightly differently since you can
only move the wrench about 1/24th of a turn.

The good news is that when torqued properly the fork bolts don't back
off. The one I had to fix was just a poor rebuild that a friend
bought. He pulled the drivetrain, I owned the wrenches. This also
proves that you can never have too many tools.

Cheers,

Kelley


On Mon, 6 Feb 2006 23:02:33 -0500, "Bob Chilcoat"
wrote:

No safety wire on the one I was in. Headed (can't now remember if square or
hex) machine bolts that just screwed in against the shaft. The one I needed
to get to was aimed toward the bottom wall of the casing, and completely
inaccessible. As far as I could tell, the only thing keeping it tight was
the proper amount of torque. The MGA you refer to probably used gear oil as
a lube. The 1100 gearbox (like the Mini's) ran in engine oil since it
doubled as the sump for the engine. I don't know if that would make a
difference to screw retention.

On Mon, 6 Feb 2006 23:02:33 -0500, "Bob Chilcoat"
wrote:

No safety wire on the one I was in. Headed (can't now remember if square or
hex) machine bolts that just screwed in against the shaft. The one I needed
to get to was aimed toward the bottom wall of the casing, and completely
inaccessible. As far as I could tell, the only thing keeping it tight was
the proper amount of torque. The MGA you refer to probably used gear oil as
a lube. The 1100 gearbox (like the Mini's) ran in engine oil since it
doubled as the sump for the engine. I don't know if that would make a
difference to screw retention.
You are not the only one to have to deal with that "chinese puzzle"
Square headed grub screws installed by dis-satisfied union workers
during a "job action" were DEFINITELY a bad combination. If the design
had been more "repair friendly" it would not have been as much of an
issue.

Told you so!
Tom
"Bob Engelhardt" wrote
in message
...
I was the OP in the Jan 26 thread "Any tips for
finding air compressor leak?". I'm here to
report that I've had complete success ... YES!

One by one, I removed the potential leak-ers:
the pressure guage, the drain plumbing, the
quick release outlet, the pressure switch, the
pressure relief valve and the Christmas tree
(manifold) itself. As each item was removed, I
plugged it's port and checked for continued
leakage. If the leak had stopped, I would know
that piece was the problem. I also rigged
pressure to that item and held it under water,
looking for bubbles. And with them all off, I
checked the tank alone.

It was the last piece - the manifold itself that
was leaking. Well not itself exactly - it was
the check valve built into it inside the tank. I
had soap bubble tested for this leak without
finding it. Poor technique, I guess. The valve
was just a spring loaded disc against a seat.
The disc was phenolic like, so I tried making
ones of nylon and soft rubber. They both leaked
worse than the original.

So I cut off the stub that the valve was built
on and drilled and tapped an NPT hole for a real
check valve, which I got from McMaster-Carr. I
reinstalled everything, pumped it up, and 2 days
later there has been ZERO drop in pressure!

Thanks for all the responses. The posters who
said "It's the check valve" were right.

Bob

On Wed, 08 Feb 2006 05:08:35 GMT, Ignoramus11278
wrote:

Would anyone have any suggestions on how to make the most bubbly soapy
solution for detecting leaks. Thanks

i
I use a tablespoon sized squirt of dawn dish soap in a large spray
bottle of water, give it a shake and you're in business. Put he water
in first or you won't get it all in for foam coming out,
Gerry :-)}
London, Canada

On Wed, 08 Feb 2006 00:52:33 -0500, Gerald Miller
wrote:

On Wed, 08 Feb 2006 05:08:35 GMT, Ignoramus11278
wrote:

Would anyone have any suggestions on how to make the most bubbly soapy
solution for detecting leaks. Thanks

i
I use a tablespoon sized squirt of dawn dish soap in a large spray
bottle of water, give it a shake and you're in business. Put he water
in first or you won't get it all in for foam coming out,
Gerry :-)}
London, Canada
Add a bit of glycerin to keep the bubbles from popping - you find the
tiny leaks better that way - or simply go to the "dollar store" and
buy a bottle of kids' bubble soap (for blowing big bubbles from a
bubble pipe or ring) That's what some gas-fitters I know use.