Industrial Utility Efficiency

Notes From Down Under: Having a Go at a Leakage Survey?

So you’ve purchased an ultrasonic leak detector after a sales person gave you a demonstration on detecting compressed air leaks. You’ve read all those articles on how air leaks are wasteful, expensive and leakage programs provide good paybacks. Perhaps you’ve even had a go at a leakage survey. Either way, by now you’ve realised leakage programs are not as simple as they sound and without an ongoing plan of attack, you will probably never see the results you thought you could achieve. This article is written to illuminate common mistakes made in leak surveys and hopes to provide guidance on how to turn that around.


Start with the Repair Plan

Probably the most critical mistake I find is that everyone thinks leakage surveys starts with detecting the leaks.  Any two-bit fool with a spray bottle and some soapy water can detect leaks. Now that you own an ultrasonic leak detector, this will be faster and less messy, won’t it? But leakage doesn’t start with detection, leakage starts with the plan for repairs, after all, if you aren’t going to do the repairs why are you even thinking about doing the leakage survey? To start with, who is going to do the repairs? When is the best time to make the repairs? How long is it going to take? What parts will you need? How long will it take to get them?


When and Who and What?

I would usually aim for a planned shutdown, ideally one where the compressed air can be turned off across the whole site for around 12 hours or more. Of course this takes some coordination, but it can be achieved. The best ones are those where the power will be off across the site. This means you’re going to need some torches and lights. Working back from that date you want to allow at least six to eight weeks prior to completing the survey, to organise the parts and book the labour. I can almost guarantee you that using one or two technicians isn’t going to cut it for the repairs.

Now that you know when the survey needs to be completed by, who’s going to do it? You have a detector, so you can always do it in-house, but have you ever really tried? Do you have everything you actually need? And what’s worse is the poor individual that’s going to do the detection is pretty much going to listen to varying levels of static for as many days as it’s going to take to do the survey.

In short, it’s a horrible job and your technicians may be better utilised on routine maintenance! You are going to need a tag of some form for each leak, a way to record the details for each leak, how big the leak is, what parts are required to fix the leak, whether the leak can be isolated or repaired during production periods, how long will it take to fix it, the list goes on and on.

A very large notepad might work, but when you try and consolidate everything you wrote down, it will be a disaster. Best off to get someone in that specialises in leakage surveys. The minimum reports you will want to see come out of it will be:

  • A full list of the leaks with an estimated flow and cost (more about this shortly)
  • Details on each leak, its location and a couple of photos to help locate the leak later
  • A list of parts required for each leak
  • An estimate on the amount of time required to repair each leak



Contrary to popular belief, it is possible to have a leak-free compressed air system.
leaktag015 leaktag168


After the Survey – Ordering Parts

By the time a leakage survey is completed you should have all the information you need to complete the repairs. Let’s order some parts…

For your own sanity, stick to like for like parts, if it’s an XYZ regulator, replace it with an XYZ regulator. Otherwise you are going to be drilling new mounting holes and creating a whole lot of extra work for yourself. There will be enough superseded items on your list to keep anyone entertained, so don’t make it worse. That does mean you might need to place 10 or 15 small orders to different suppliers - but it’s worth it. You’ll also be wanting to double check the part you are ordering comes with all the extra bits. For example, some regulators don’t come with a gauge or mounting brackets, some solenoids don’t come with the face-mounting gasket…etc.

Inevitably on your list will be a long series of push fits, hoses and clamps required for simple quick fixes. You know you keep that sort of thing in stock, so you should be good to go. More than likely what you think is a good level of stock will be everything you don’t need and this will bring your entire plan unstuck right at the first gate. You need to buy in all the parts on your list and some extras, I’ve seen one site go through more than a hundred ¼” x ¼” straight push fits on a single survey. You must order everything!

Once the parts start to arrive, you’ll want to sort and bag them by each leak. You’re going to need space to do this… but each set of parts should be bagged and tagged to match each leak, that way when the repairs happen the technician can toddle off with a bunch of assigned leaks, all the parts, their tools and some fresh thread sealant! Don’t forget the THREAD SEALANT! Make sure every technician has at least one fresh tube and you have some spares because someone is going to lose one in the factory somewhere.



One machine can have several leaks.


The Big Repair Day: Target 80-90% of Leaks Repaired to Make a Difference

And now the big day is finally here. By now you are wondering could I have done the repairs in bits and pieces and to some extent the answer is yes, but the reality is they never get completed, other problems take priority and you never see a notable change to the system.

Here we go… time to shut the system down. Did you remember that it takes time for the pressure to drop? You’ll want to shut the system down at least an hour before the technicians are due to start, assign leaks on a per machine basis to each technician or ten at a time if there are only a few on each machine. Let the technicians head off with everything they need and when they return have them confirm each leak that was repaired.

If by some chance they were unable to repair something, make sure all the explicit details are documented so you can come back to it another day. It is best to get as many leaks repaired as possible rather than focus on why one particular leak can’t be done. You need to get greater than 80% of the leaks repaired to have any real effect and preferably more than 90%. That includes all the tiny little ones, they have plans to grow and they are going to grow a lot quicker than you expect.




Leaks are commonly found where older FRL’s are located.


What Happens When Your Repressurize the System?

By the time the repair day is over, everyone will want to go home, but you need to stick around and let the system come back up to pressure. Typically the system pressure will come up around 7-10 psi, after a big repair day, but that means some extra stress on all those components that up until now have had it easy.

Typically several regulator, filter and drip leg drains, along with a few hoses and push fits are going to fail within the first hour. Sometimes these can be so big that it will negate anything you and your armada of technicians just finished doing. The best thing you can do, while everyone is still there, is to send them out into the factory and do their best to repair anything they can hear that failed or document what is needed to fix the problem. Once these repairs have been completed, sometimes over the next week, then and only then might you see a result.

But how will you ever know if you aren’t monitoring your system correctly?


For more information, contact Warwick Rampley, Compressed Air Alliance, at email: 

To read more about Compressed Air Leak Surveys, please visit