I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.

In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?

Thanks,
Bob

Bob Engelhardt wrote:

I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.

In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?

Thanks,
Bob



Got a buddy in the HVAC business?

If you do, maybe you could empty the tank and get him to squirt a little
refrigerant gas into the tank and then use his leak detector to see if
he can spot where it's leaking out.

Jeff

--
Jeffry Wisnia

(W1BSV + Brass Rat '57 EE)

"Truth exists; only falsehood has to be invented."

In article ,
Bob Engelhardt wrote:

I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.

If my ballpark calculations are correct, you're losing 3.7 gallons of
air (at atmospheric) per hour, so you should be able to see bubbles when
you hit it, unless it's distributed over a lot of joints, each leaking a
little bit.

You can add a bit of glycerine to the soapy water to help the bubbles
last longer, but generally soapy water (and good lighting) does the
trick. Does this thing have piping connected to it? Have you checked all
that? If it has an airhose attached and pressurized, drop the hose in a
bucket of water and look for bubbles.

Other option is just to turn it off when you are not using it.

--
Cats, coffee, chocolate...vices to live by

1. http://www.amgas.com/ltpage.htm This stuff will detect gnat farts.

2. Single malt...and take comfort in the sound of your pump.


"Bob Engelhardt" wrote in message
...
I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I just
measured & timed it: in 17 hours the pressure dropped from 100 psi to 55.
This means the compressor runs several times a day just to maintain
pressure. It is loud, so this is annoying.

In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL leaks
(but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?

Thanks,
Bob

If my ballpark calculations are correct, you're losing 3.7 gallons of
air (at atmospheric) per hour, so you should be able to see bubbles when
you hit it, unless it's distributed over a lot of joints, each leaking a
little bit.

Or.......

Leaking back thru the checkvalve, into the cylinder and out the
crankcase or intake valve?

Did you spray soap in those areas ??

Grummy

I like the check valve suggestion. Not sure how to prove it, perhaps
you could remove the intake filter and put a soap bubble over the
intake. But then if the leak is going through the oil bath of the
pump...... You would need to have a diagram of the pump to figure out
hot to test the hypothesis of a check valve leak.

Richard


wrote:
If my ballpark calculations are correct, you're losing 3.7 gallons of
air (at atmospheric) per hour, so you should be able to see bubbles when
you hit it, unless it's distributed over a lot of joints, each leaking a
little bit.


Or.......

Leaking back thru the checkvalve, into the cylinder and out the
crankcase or intake valve?

Did you spray soap in those areas ??

Grummy



--
http://www.fergusonsculpture.com
Sculptures in copper and other metals

The last time I searched the internet on soap bubbles, Dawn and Joy
liquid dishwasher detergent were recommended for making bubble
solutions (not for finding leaks but for amusing kids ). I used Dawn
and water to find a leak in a tire that was dropping from 35 psig to 25
in a week or so. It is sensitive enough to find your leak.

I would consider putting a valve at the tank inlet and see how fast it
leaks down with the valve closed and preventing air from getting to the
compressor.


Dan




Bob Engelhardt wrote:
I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.

In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?

Thanks,
Bob

Bob Engelhardt wrote:

I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.

In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?

Thanks,
Bob

Well it seems like you've isolated it. If you've plugged the tank
outlet,
and the "christmas tree" checks out, it must be the non-return or check
valve between compressor and tank. Once removed, they're not a biggie to
test.

Tom

Bob,
I have found that small leaks are hard to find with high pressure in
the
system. Let it down to 20# or less, then you will see the bubbles
foaming
at the little ones. Too much press. blows away the suds before you
notice them.
RichD

Bob Engelhardt wrote:
I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.

In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?

Thanks,
Bob

"Bob Engelhardt" wrote
in message
...
I have an IR home-shop compressor (20 gal, 1 1/2
hp) with a leak. I just measured & timed it: in
17 hours the pressure dropped from 100 psi to 55.
This means the compressor runs several times a
day just to maintain pressure. It is loud, so
this is annoying.

In an attempt to find the leak, I sprayed soapy
water on EVERYTHING: the entire tank, the drain
fittings, the "Christmas tree" fittings
(pressure switch, guage, etc). I even
disconnected the pump output line to see if the
check valve was leaking back into the pump. I
found a small leak in the pressure guage
connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find
that leak! Before I start taking parts off to
eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me
the work?

Thanks,
Bob


I'd bet good money on the non-return valve. In my
experience a check valve is a devise that restrict
flow in one direction and restricts it slightly
more in the other direction. If you want to stop
it completely, put a solenoid valve in the line
before the check valve and connect it pressure
switch. Don't forget to tee a pressure relief
valve off the line,upstream of the solenoid valve,
so that the system doesn't blow up when the coil
on the solenoid burns out.

Tom

Bob Engelhardt wrote:
I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.

In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?
Have you checked the regulator? I have a plumbed-in system with several
taps and regulators. ALL of the regulators leak. Not a whole lot,
but together it is enough to make the compressor cycle every 20
minutes or so. (It has a REAL small tank.) Regulators have a bunch
of leak paths, such as the screw thread of the adjusting handle.
It may also have a relieving port that can leak. There's usually an
unloading valve in the pressure switch, too. How about the water
drain?

I don't leave my compressor on except when I am using it (obviously,
with it running that often.)

Jon

On Thu, 26 Jan 2006 20:08:04 -0500, Bob Engelhardt
wrote:

I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.


What you describe is NOT considered a leak on a compressor. If you
are not using the machine for that length of time, simply turn it off.

There is always some air loss through the pressure regulator. You
will notice that when you adjust the pressure from higher to lower,
the regulator vents the excess pressure from the hose into the
atmosphere. The venting is why you hear the hiss. That is where the
leak occurs -- the air seals in the regulator are metal to metal and
are not intended to be perfectly air tight; they are intended to be
functional, safe, and robust during normal usage of the machine. I've
had name brand, top of the line, replacement regulators, straight out
of the box, leak more than that.

In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?


You are wasting your time.

Thanks,
Bob

I'd go for a binary chop and insert a few ball valves in the line
to isolate parts of the system. At least you'll be able to better
pinpoint what parts to check. Turn off half the system overnight
and see if it still leaks.

Bob Engelhardt wrote:

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?

I've had the same problem with lab quality regulators for
nitrogen, argon, CO2, etc. I'm surprised how much gas is
lost through them.

Speechless wrote:

There is always some air loss through the pressure regulator. You
will notice that when you adjust the pressure from higher to lower,
the regulator vents the excess pressure from the hose into the
atmosphere. The venting is why you hear the hiss. That is where the
leak occurs -- the air seals in the regulator are metal to metal and
are not intended to be perfectly air tight; they are intended to be
functional, safe, and robust during normal usage of the machine. I've
had name brand, top of the line, replacement regulators, straight out
of the box, leak more than that.

Tom Miller wrote:
I'd bet good money on the non-return valve.

What they refer to as a "Check Valve" on most of this new JUNK is
scary.

Often it is just a rubber slug down in a hole with a spring pushing on
it. The new air compressors in general are ALL pretty much JUNK.

Grummy

Try rubber banding a balloon onto or over suspected items.

Pete Stanaitis
---------------

Speechless wrote:
On Thu, 26 Jan 2006 20:08:04 -0500, Bob Engelhardt
wrote:


I have an IR home-shop compressor (20 gal, 1 1/2 hp) with a leak. I
just measured & timed it: in 17 hours the pressure dropped from 100 psi
to 55. This means the compressor runs several times a day just to
maintain pressure. It is loud, so this is annoying.



What you describe is NOT considered a leak on a compressor. If you
are not using the machine for that length of time, simply turn it off.

There is always some air loss through the pressure regulator. You
will notice that when you adjust the pressure from higher to lower,
the regulator vents the excess pressure from the hose into the
atmosphere. The venting is why you hear the hiss. That is where the
leak occurs -- the air seals in the regulator are metal to metal and
are not intended to be perfectly air tight; they are intended to be
functional, safe, and robust during normal usage of the machine. I've
had name brand, top of the line, replacement regulators, straight out
of the box, leak more than that.


In an attempt to find the leak, I sprayed soapy water on EVERYTHING: the
entire tank, the drain fittings, the "Christmas tree" fittings (pressure
switch, guage, etc). I even disconnected the pump output line to see if
the check valve was leaking back into the pump. I found a small leak in
the pressure guage connection, which I fixed (tightened). It STILL
leaks (but not the pressure guage)!!

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have any
tips/tricks for finding leaks that might save me the work?



You are wasting your time.


Thanks,
Bob

Richard Ferguson wrote:

I like the check valve suggestion. Not sure how to prove it, ...

I did check this. I disconnected the outlet line/pipe/tube from the
pump and bubble checked it. Negative. bob

wrote:

...
I would consider putting a valve at the tank inlet and see how fast it
leaks down with the valve closed and preventing air from getting to the
compressor.

The problem with that is there is a "Christmas tree" screwed to the
tank. Off of it is: the pump output, the unloader, the pressure guage,
the pressure switch, the pressure relief, and the outlet port. So there
is no isolating the tank, easily. Bob

wrote:

Bob,
I have found that small leaks are hard to find with high pressure in
the
system. Let it down to 20# or less, then you will see the bubbles
foaming
at the little ones. Too much press. blows away the suds before you
notice them.
RichD

Good idea, I'll try it. Thanks, Bob

Jon Elson wrote:
Have you checked the regulator? ...

No regulator - the compressor just has a quick release on it & is not
plumbed around the shop. (The quick release does not leak.) Bob

spaco wrote:

Try rubber banding a balloon onto or over suspected items.

Eh, "great minds" - that's what I did with the pressure guage to check
for an internal leak. Bob

Jeff Wisnia wrote:
Got a buddy in the HVAC business?
...

No. But good idea.

Ecnerwal wrote:
....

You can add a bit of glycerine to the soapy water to help the bubbles
last longer, but generally soapy water (and good lighting) does the
trick.

Good idea.

Does this thing have piping connected to it? ...

No, just the compressor itself.

Bob

Mike Berger wrote:

I'd go for a binary chop and insert a few ball valves in the line
to isolate parts of the system. ...

Well, it's not plumbed into a distribution system. This JUST the
compressor that's leaking. But I am chopping its parts. In stages I've
removed: the pressure guage, the drain plumbing, the quick release
outlet, and the pressure switch. In each case, I plugged their port and
soap checked it. Still leaking! I only have the pressure relief
valve and the Christmas tree itself to go. And the tank itself. Bob

Ecnerwal wrote:
....
Other option is just to turn it off when you are not using it.

It might come to that. I's rather not, as my use is small and if it's
leaked down to 20 psi, it's a nuisance to wait for it to pump up, just
to use the blow gun for 10 seconds. Bob

Tom Gardner wrote:

1. http://www.amgas.com/ltpage.htm This stuff will detect gnat farts.

....

"Thin film bubble juice" - who would have thought?! Bob

Yep - min to max. Half way - yes no - move to the next half point closer or further.
You will be there in 7 or less tests.

Martin Eastburn
@ home at Lions' Lair with our computer lionslair at consolidated dot net
NRA LOH & Endowment Member
NRA Second Amendment Task Force Charter Founder



Mike Berger wrote:
I'd go for a binary chop and insert a few ball valves in the line
to isolate parts of the system. At least you'll be able to better
pinpoint what parts to check. Turn off half the system overnight
and see if it still leaks.

Bob Engelhardt wrote:

Now it has become a challenge. I've GOT to find that leak! Before I
start taking parts off to eliminate them as leaks, does anybody have
any tips/tricks for finding leaks that might save me the work?

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Martin H. Eastburn wrote:

Yep - min to max. Half way - yes no - move to the next half point
closer or further.
You will be there in 7 or less tests.

In 7 tests you could isolate to 1 of 128 possibilities. To find which
of N possibilities, you would need log-base-2 of N, tests. Rounded up.

But that's what you meant, right?

Bob

"Bob Engelhardt" wrote in message
...
Ecnerwal wrote:
...
Other option is just to turn it off when you are not using it.

It might come to that. I's rather not, as my use is small and if it's
leaked down to 20 psi, it's a nuisance to wait for it to pump up, just to
use the blow gun for 10 seconds. Bob

Slight leakage at the tank drain valve is not uncommon. Small pinhole leaks
at the tank seams or even in the tank material might be possible.

The kid's bubble stuff in a spray bottle works great for finding leaks in
air equipment and in tires, wheels, etc.. It is handy for getting the
solution on the bottom surfaces.

Don Young

On Fri, 27 Jan 2006 02:50:21 GMT, "Tom Gardner"
wrote:

1. http://www.amgas.com/ltpage.htm This stuff will detect gnat farts.

Okay, let's try it. Now where's Cliff when we need him... ;-P

-- Bruce --

I've read a lot of the replies (maybe not all) and don't want to
re-iterate what was said.They are all good suggestions. One poster
mentioned the use of refrigerant and a leak detector. That gave me the
idea of canned smoke. It is used to detect drafts. I don't know how
effective it would be at detecting a leak in a compressor. One crazy
guaranteed way of finding that leak is to submerge the whole system
into water. If not the whole system, then maybe part of the system
which wont be destroyed by water. Good to test the not so obvious.
Since the system is 20 gal, that could be a challenge.

Sounds crazy perhaps, but have they ever been tested?

I've read a lot of the replies (maybe not all) and don't want to
re-iterate what was said.They are all good suggestions. One poster
mentioned the use of refrigerant and a leak detector. That gave me the
idea of canned smoke. It is used to detect drafts. I don't know how
effective it would be at detecting a leak in a compressor. One crazy
guaranteed way of finding that leak is to submerge the whole system
into water. If not the whole system, then maybe part of the system
which wont be destroyed by water. Good to test the not so obvious.
Since the system is 20 gal, that could be a challenge.

Sounds crazy perhaps, but have they ever been tested?

I had just such an insidious leak once. I sprayed every threaded
connection to no avail. Know where the leak turned out to be.... the
seat of the pop-off valve. I had not thought to spray the entire
pop-off valve, I only squirted the threaded connection and totally
missed the problem area on my first few searches. Sure enough, bubbles
formed at the side holes of the relief valve.

I guess the moral of this story is always look beyond the fittings.
Even things like pressure switches can leak internally (like if the
diaphragm is old). Your leak is A LOT faster than the one I had, but I
still had very obvious bubbles once I found the spot. My leak also
stopped at around 80 psi, with the pop-off valve spring winning out
over the air pressure. Only at high tank pressure did I have a
detectable leak.