leak search in large piping system

[coldseeker2013]

Good evening to everyone. Greetings from a new member. Just out of a refrigeration technical college with almost a year of working experience. This is my first post.
I need to verify the possibility of leaks in a number of sets of large diameter refrigeration copper pipes: 2.5 and 3 inches, according to the printing in the Armaflex-type insulation. The length of each line is between 30 to 45 meters. I have used nitrogen for this purpose in much smaller jobs, but never before in such a large scale. In the same place where these pipes are installed, there is a compressed air system for pneumatic machinery. The air passes through a mechanically refrigerated drier and several steps of filtration. Does anyone has any experience in using compressed air for leak search instead of nitrogen? Thank you for any comments.

[Tesla]

Hi coldseeker and welcome to the forum
I see you are looking for a cheaper way to leak test and I have an old good proven method for you.
Get a hold of some large plastic sheet, cut it into strips wide enough to surround the pipes. Use some duct tape and tape it so it covers and seal the pipes in sections. Go back the next day and cut the lowest point of the bag sections and use a leak detector to detect any leaked gas. This may take time to cover the pipes but is a lot cheaper than loosing any gas and nitro. if you tape the small cuts made for testing you can just rip the tape open again when another leak happens. I have used this method on many chillers even at low pressures. You should not use compresses air as it contains oxygen - oxygen and oil under pressure can be explosive!

[coldseeker2013]

Good evening Tesla and thank you for your comments.
Let me tell you what I understand from your method to see if I am getting it right:
1. I have to surround sections of the pipe with the plastic sheet strips: using duct tape around the circumference of the pipe to hold down the plastic strip tight at both ends of each section, and more taping along the longitudinal cut.
2. Next I put in the the pipe a mixture of refrigerant and nitrogen up to certain pressure: maybe 10 kg/cm2.
3. Wait until the next day and if the pressure did not hold, check each bagged section by making a hole in the plastic sheet and sniffing it with the electronic leak detector until I locate the bag containing any amount of the escaped mixture.
Am I right?

[Tesla]

Hi coldseeker
You are spot on (correct), that is the best method. I don't know why these new gas laws don't make it mandatory. But even better it will usually work without the nitro. You could simply bag it with the system running normally and check it the next day. Really simple stuff - it was an old trick for testing low pressure chillers where the some vessels could only take a maximum of 12psi pressure before the rupture disc would give. Very good method for big chillers with lots of little leaks. Best of luck on this job.
Other members should take note of this procedure too with the price of gas now days and the cost of using nitro.

[coldseeker2013]

Good evening. Thank you for your response. Just one comment.
I talked to other techs about the job I´m going to do and the procedure I selected to follow, and one of them made the following recommendation: forget about using electronic leak detector if the copper pipes are already insulated. The reason: chemistry of the components the Armaflex type insulation is made of, interferes with the detector´s sensor sensitivity, resulting in possible false positives. The pipes are already insulated, but I have cut open and removed about 30 cm of insulation at each joint. I also cleaned the exposed pipe with degreaser and water in order to eliminate any trace of the powder that the Armaflex insulation has in its inside. Did someone hear something about this issue with the sensitivity of the electronic detector?

[Greek_engineer]

Why don't you try an uv lighter?
And please can anyone of you send a photo with that method of using sealed plastic stripes-bags?
It seems interesting enough!

[jittypop]

What makes you think the leak is in the piping, have you completely checked the condensor and evaporator. if that checks out you may want to isolate each line and pressure test, i have done this with difficult leaks to find and had success. Good luck!

[Rob White]

.

On the back of what Tesla said I use a similar method but without the plastic.

I cut the Armaflex every 3 or 4 meters and remove a small section about 10 cm
wide. I also remove a section at each elbow and tee.

Then clean all the exposed sections and wait.

After a day or two if oil is seen at the end of a piece of insulation, then you know
that section is leaking. I also stick the sensor of the leak detector in the ends of
the insulation and that will also indicate the section of leaky pipe.

When you are happy it is all clear then re-insulate and tape or glue it sealed.

It always works because the oil and gas will always show up in one of the sections.

Regards

Rob

.

[coldseeker2013]

Good evening and thank you for your interest in my post.
I have spent the last couple of days getting the N2 bottles, plastic sheets, electronic leak detector etc. When I thought I was ready to go, first thing I did was send a discharge of N2 through the pipe and, to my surprise, I found a large amount of debris coming out the other end. To my understanding, this means that the welder did not have any inert gas flowing through the pipes when welding them. I proposed to the project manager that we better clean the inside of the pipes before beginning to look leaks. He agreed. Once again: for much smaller systems, I´ve used R141 as flushing agent. Is it equally convenient in the case of large pipes?