How much pressure loss is acceptable

[Rob Hannon]

I know ideally no loss is what I should shoot for, but what is realistic in a home shop setup?

Right now just from charging my line, valve to my tank is closed, I lose about 40 PSI over a 12 hour period. I have 1 leak in a cast tee that I am not excited about replacing. It is in about the worst spot possible to pull out. That leak is very slight and in the body itself. I retested it several times because I couldn't tell if it was just a bubble I created putting the soap on or one from a leak.

Not finding any other leaks at joints so I suspect I am losing some either at of the couplers or the regulator. The line set has probably 30' of 3/4 black pipe and another 20' of 1/2 drops.

 

[Boilerhouse]

I doubt if I lose 1 psi a day. I kept my compressor off when not in use, so it may go several days before it gets turned on. I don't recall ever seeing the pressure lower than when I shut it off. Go over everything with soapy water, the bubbles will reveal leaks. It may be a pain to repair, but you will be glad you did.

 

[dagofast]

Short answer: The acceptable amount is how ever much you are willing to put up with.

Long answer: How many hours a day/week/month are you in your home shop? How often does your compressor cycle when you aren't using any air? How much does your electricity provider charge per kilowatt hour? How much cash do you have to burn? How bad is your OCD? Where do you fall on the ambition/contentment scale? Is your wife/girlfriend/SO hot and very lonely? Would you rather be doing something else besides chasing air leaks?

 

[930dreamer]

I know a few of my air tools leak so I don't leave them connected i.e air chuck.

 

[Rob Hannon]

Short answer: The acceptable amount is how ever much you are willing to put up with.

Long answer: How many hours a day/week/month are you in your home shop? How often does your compressor cycle when you aren't using any air? How much does your electricity provider charge per kilowatt hour? How much cash do you have to burn? How bad is your OCD? Where do you fall on the ambition/contentment scale?

In my shop probably 1 long day a week and maybe 3 or 4 more days a week that I am in for things that take an hour or less.

Not sure how often my compressor cycles with my current setup. I have a timed drain so every 45 min from 8am to 7pm it opens for about 5 seconds. The compressor is also on a timer for that period as I don't want it kicking on in the middle of the night. I have good neighbors and want to keep them that way.

I think the electric usage has been pretty negligible. Right now my OCD is pretty high, but that will change when I move on to the next project. Once my air system is the way I want it I will be knee deep in setting up some pneumatic controls for dust collection. I suspect that whatever state the leaks are in at that point will remain until I move into a new shop.

 

[545_days]

I would think the timed drain is a far greater loss than a small leak producing a few bubbles.

Opening every 45 minutes from 8AM to 7PM is 14 times per day. At 5 seconds open, that adds up to 70 seconds during the 11 hour period. How much pressure does your system lose if you open the drain for 70 seconds? My guess is quite a bit.

Or did I misunderstand, and the 40 psi loss is overnight? If so, check the drain valve for leaks. With it cycling so frequently, I would not be surprised if it has developed a leak.

 

[Rob Hannon]

I would think the timed drain is a far greater loss than a small leak producing a few bubbles.

Opening every 45 minutes from 8AM to 7PM is 14 times per day. At 5 seconds open, that adds up to 70 seconds during the 11 hour period. How much pressure does your system lose if you open the drain for 70 seconds? My guess is quite a bit.

Or did I misunderstand, and the 40 psi loss is overnight? If so, check the drain valve for leaks. With it cycling so frequently, I would not be surprised if it has developed a leak.

Loss was overnight, but also isolated from the compressor and drain. I was surprised at how little loss I had from a full day of the condensate drain going on schedule with the compressor turned off. Only lost about 15lbs from the tank.

Some of my loss is from the 1 bad fitting and a small amount I suspect is from temp change. I think I now know how to test for leaks in the regulator. No idea how to test for leaks in the quick couplers other than to submerge them. More I think about this, the more I think I am just being a little OCD on this. Just need to decide if tearing everything apart to replace the poorly cast tee is worth it or not.

 

[sberry]

I thought the hearing in my right ear is better than it is, my son said this connector leaks. I stuck it in a cup of water,,, sure as shat. I am working my ambition up to fix it yet. Leaks probably cause me a cycle or 2 a fay, 8 or 10 minutes of run on 3 hp, couple 3 dollars a month maybe?
That drain could be timed to once a week vs several times a day.

 

[The Cobbler]

ideally a non leaking system is what we all would like, but with a marginal leak(s) that is difficult to repair, it's a judgement call I would say.

 

[Showkey]

In industry air leaks are HUGE concern as it costs $$$$$$ over the long haul. Consulting and repair to solve leaks is a big business.

Home leaks......my concern is I do not want the compressor running unattended. This leaving a compressor on 24/7 has been a debated topic in the GJ.

 

[DFB]

My setup also leaks down some. Kicks on and cycles up about once in every 24 hrs right now with no one using it. And of course nothing connected... Ha I have one tire chuck that hangs up sometimes after using it, if that happens will go off every 30 minutes

My thread connections have all been tested too, not obvious where my leakdown coming from maybe it is the regulator valve

 

[Dingleburry]

You could be lazy and weld the leak closed.
Or try peining it closed with a center puch/hammer?
Just my first few thoughts.
Or
Python-Wrap Pipe Leak Repair Kit

 

[MattT]

The line set has probably 30' of 3/4 black pipe and another 20' of 1/2 drops.

That's only about a gallon of volume so that pinholed fitting may be your only leak. And it's definitely not worth fixing from an economic standpoint.

 

[matt_i]

I think that's a lot, especially if you can see a fitting with a crack in it. Get a Ward Mfg fitting to replace it.

If a fitting is hard to get to and requires taking half the shop apart, then sawzall the pipe out, rethread it in a couple places and put in a union.

 

[fang123]

If I was losing anymore than 10 psi a week, I'd be finding and repairing.

 

[sparky 1971]

I have a leak in my line. It takes about three days to empty the 60 gallon tank. I know where the leak is, but am not concerned enough to fix it. I have a ball valve right at the tank and 95% of the time, I close the valve when not in use. The other 5%, I have to wait for it to fill up before I use it. I always turn the compressor power off. I put a switch for it right next to the light switch so it's hard to forget that.

 

[DocsMachine]

Started with a 60 gallon upright in the corner, and connected to nothing but a single hose straight off the side bung. Used it that way for years. The compressor basically never ran unless I was using it. Hose was plugged in and pressurized for years, basically zero leak at either end.

Recent replumbed the shop with Rapidair, and moved the compressor out into an attached shed. Connected, live, and pressurized 24-7. Generally have at least two hoses connected, one just a blowgun whip.

The compressor might run once a week, on it's own. I'm not really even sure. Might be once every two or even three weeks.

If it ran once a day, on it's own, I'd be leak-hunting again.

Doc.

 

[Citation]

Here is how I would think about it.
1. I don't like leaving the compressor on if I'm not around for all the usual reasons. Thus I think about how much leaks out between times I use the compressor.
2. How long are you willing to wait to refill the tank before you use it next? Does waiting for the tank to refill slow you down? If you are the sort who starts your day by turning on the compressor and then doing other jobs it may not matter if the thing has to run for 5-10 min to build up pressure. However, if you find that you are waiting the same 5-10 min just to fill up a leaking bike tire, well that leak is now impacting your work flow.

These days most of my compressor needs are inflating or a quick use of the air gun. So I really prefer to have no leaks so I can use the compressor right away. This would be even more true if I had a larger tank and it took that much longer to build usable pressure.

I would certainly look at having a shutoff valve at the tank so leaks in the rest of the system don't drain the tank itself.

 

[Shadowdog500]

Your 40 psi loss over 12 hours on your isolated 206 in^3 lines will equate to a 0.6psi loss over 12 hours in your total 14000 in^3 system assuming a 60 gallon tank with the valve open. This leak would make your compressor kick on once every couple weeks if you left the tank valve open 24/7. I leave the valve on my tank closed and the power shut off when I’m not using compressed air and don’t really worry about leaks that small in the lines.

If you only kept the valve open on your compressor for the roughly 16 hours per week you are in the shop that means the leak would produce an 0.8PSI drop in your system per week. If you think it is worth it to replace that hard to get at part for a 0.8PSI drop per week go for it. As for me, I really wouldn’t worry about it.

Note This post was updated after realizing that I moved the decimal by one order of magnitude on the lest step of the calculation. Thanks MattT For catching my decimal error.

 

[Milton Shaw]

I put a motorized ball valve on the outlet of my compressor. When the lights are off the compressor is isolated from all the pipe except a 24" piece of hydraulic hose between the tank and the ball valve. Doesn't run ever until I turn the lights back on. Problem solved. Motorized ball valve from Amazon was about $50 as I remember. Ball valve opens on power on and closes when power removed.

 

[finn]

I put a ball valve on the tank outlet when I installed my old, but new to me Champion with an 80 gallon tank. With the ball valved closed, the compressor will go a month or more before the pressure drops enough to be noticeable. I cut the power every time I leave the shop so it’s not an issue.

The piping system past the ball valve is a hydraulic hose to a regulator / filter, to 2/4” black iron and three drops with quick disconnects. That part of the system leaks pretty badly. I am pretty sure the major leak is above the door, near the 16’ ceiling.

Since it’s black iron, with tees and elbows, sixteen feet in the air, it’s not going to be an easy fix, so I choose to live with it until I run out of other things that need fixing.

The ball valve works for me, and the tank fills the line pretty quickly.

The leaks still bother me, though.

 

[Farmall450]

Ball valve on compressor outlet. Turn it off when not in use.

 

[MattT]

Your 40 psi loss over 12 hours on your isolated 206 in^3 lines will equate to a 6psi loss over 12 hours in your total 14000 in^3 system assuming a 60 gallon tank with the valve open.

Your relative volumes are about right but it looks like you've fat fingered the pressure calc. 0.6 psi in theory but since the system will be leaking at higher pressure for the entire 12 hours probably more like 1 psi loss actual.

Differential on the pressure switch is probably 30 psi or more so it'll take a while for that tiny leak to make the compressor cycle.

 

[Shadowdog500]

Your relative volumes are about right but it looks like you've fat fingered the pressure calc. 0.6 psi in theory but since the system will be leaking at higher pressure for the entire 12 hours probably more like 1 psi loss actual.

Differential on the pressure switch is probably 30 psi or more so it'll take a while for that tiny leak to make the compressor cycle.

You are correct. I multiplied 14.7 times 0.4 atm instead of 0.04 atm in the last step. A leak like that is hardly worth worrying about.

Here is the math.

 

[MattT]

You are correct. I multiplied 14.7 times 0.4 atm instead of 0.04 atm in the last step. A leak like that is hardly worth worrying about.

No wonder something went wrong with all that math. I just ballparked it using Ptank=Ppipe*(Vpipe/Vtank)

 

[Shadowdog500]

No wonder something went wrong with all that math. I just ballparked it using Ptank=Ppipe*(Vpipe/Vtank)

That works too and saves with hand cramps!

 

[bsaint]

If the leak prevents you from using a tool to full capacity then it’s too big of a leak. Other than that it’s up to you if you just want to pay for compressing air and releasing it back into the atmosphere


Sent from my iPhone using Tapatalk

 

[Rob Hannon]

I typically do close the valve on the compressor, but the older i get the more forgetful I am. I did find another leak. Pressure gauge was leaking. I replaced that and the bad t fitting. Chasing other leaks down now further in the setup, but that took care of the bulk of it. I like the idea of a motorized valve. May look into that.

 

[cabranch47]

I put a hp rated relay on the line feeding the motor. Wired the coil into the lighting circuit. Lights off, compressor off. Lights on, compressor on. Good piece of mind when leaving for extended period of time. Costs about $25.00.

 

[1320stang]

Ball valve on compressor outlet. Turn it off when not in use.

I too put a ball valve on the outlet to my compressor, then a hose to the regulator, then just to a air hose for now. If I forget to shut the valve, the compressor will probably bleed down overnight, I also shut the compressor off unless I'm using it. Nothing like it kicking on a 3am when you're sound asleep.

 

[nmk_61802]

Jumping on the OP's thread.

I too went thru this exercise. I have been chasing minor leaks for the last week so that I can move on to the next phase construction.

For me I strive for perfect, I cut, threaded and assembled all of the piping with both pink tape and dope. After testing the system with bubble solution twice, and an Amprobe Ultrasonic Tester twice, I can't find any additional leaks (Only 4 to begin with). The ultrasonic tester found 1 bad casting in a coupler that I was struggling to locate. After repairing those, I decided to run the sanity check in this thread for those I can’t locate

Because I like Excel, I set everything up to populate the tables once you enter in the basic info (dia, length, test duration and delta-P). I realize this is skewed slightly to the lower end since the leak rate would reduce as the PSI drops (not linear), but this is the easiest way to quantify the leak.

I only tested the distribution system (120 PSI) from the regulator out, if my math is correct, I am showing around 2 PSI loss across 20hrs or .1 PSI/hr once the tank is included. My guess is these are just to small to locate.

Further taking this out with a 30 PSI cut on, it should take 300 idle hrs before the compressor kicks on. That assumes I leave the system energized, doubtful since I have an automatic valve at the compressor to kill the air unless I call for it.

I think I will call it good and move on.

If I ever get this project completed, I will have to make a gallery post.

 

[Firebrick43]

For me I strive for perfect, I cut, threaded and assembled all of the piping with both pink tape and dope. After testing the system with bubble solution twice, and an Amprobe Ultrasonic Tester twice, I can't find any additional leaks (Only 4 to begin with, so I decided to run the sanity check in this thread. The ultrasonic tester found 1 bad casting in a coupler that I was struggling to locate.

Have we come to the point of saying only 4 leaks is good? Why are we tolerant of crappy Chinese fittings and pipe now? It used to be one leak would get someone joshed on a job. Then to save money everyone started using Chinese fittings and then pipe which threads something awful.

You can still get made in American ward fittings that are great.(supplyhouse.com). Last year I plumbed my pretty complex lp gas system at home and while I couldn’t find in stock USA pipe I did find some Turkish stuff that threaded at least better than the **** we get a work. Tested at 90 psi for two weeks and no leaks.

We now save old pipe we take out when replacing machines instead of scraping it as it threads like butter without tearing.