Instrument Air Leak Detection for Offshore Brownfield. 4

[Hans85]

Hi fellow engineers,

Recently in my company's brownfield platforms, there has more frequent Shutdowns due to instrument air leaks triggering the pneumatic solenoids.

These platforms are operating beyond the design life, thus there are a lot of wear/tear and reliability issues.

There has never been any programs for PMs on instrument air tubings; these are long runs of SS316 tubings, which extends from modules to modules and penetrates the deck gratings.

I would like to seek any advice in terms of integrity (leak) testing of these instrument air lines/tubings. We've ruled out the following:
1) Hydrostatic test as these are instrument air lines; the instruments would be susceptible to water.
2) Ultrasonic leak detection test as this would mean the leak detector used will need to be close to the source; i.e. will have to trace the tubing runs which would not be practical as the tubings can be very high above or hard to reach.
3) Pressure holding test - too time consuming and cannot be performed online. Planned Shutdowns for platforms are usually reserved for other activities of higher priorities.
4) Soap/snoop leak test - too long tubing runs which are not readily accessible.

We are in the midst of discussion of other possibilities. There's been talks about doing something similar to dry deluge testing; we introduce smoke into the system with the hopes that the leaks can be detected visually. For the source of the smoke/vapor, we are thinking of introducing dry CO2 or Glycol+Glycerin (similar to vaping) into the system.

Opinions/Recommendations/Cautions welcomed.

Kind Regards,

Hans

 

[3DDave]

Some gases are relatively opaque to IR and are very visible under high sunshine conditions on Far IR cameras. Methane is one of them, CO2 is another, but there are likely to be others. See

https://www.youtube.com/watch?v=ClLP_Xv1buA

as an example.

I would suggest contacting a camera maker for a suitable list of gases suitable for such work as you aren't limited to either of the above.

 

[LittleInch]

Hans,

You seem to have a long list of things you can't do.

So maybe best to step back and see what you CAN do / test and what you know.

Your assumption seems to be that the tubing itself is corroded / damaged? Why do you think this?
Most leaks come from screwed connectors / regulators / leaking valves. All these should be accessible? Or most of them. Find the leaks in these things and fix them. Control valves can leak air at a huge rate if the diaphragm has a hole or the regulator is pissing out air. Go for the easy wins first.

Challenge the planned shutdown issue. To me if your air system is causing unplanned shutdowns then it has elevated itself to a very high priority.

What can you check in terms of accurate pressure guages at the headers / instruments / valves. There are usually test ports/ vents/drains to tap into. Don't use the small guages on the regulators - most will be broken by now. then it allows you to focus on those branches which are lowest pressure first.

Helium injection and detectors is a way to check leakage around all the instruments / valves/ regulators etc if it's too noisy to listen for leaks.

Can you insert small flow meters on the headers then do a check on "normal" air consumption from all the things hanging off that particular branch? An external U/T meter might work? Then find the branch(s) with excessive flow.

If a lot of the tubing is inaccessible then why are you bothered about inspecting it? Either program in a replacement or live with it and fix all the leaks you can.

Or start to have a portable pack of bottles such that you can hook up a set of valves and isolate a branch while you pressure test that branch. Inserting two valves and a branch on your next shutdown in a dozen places to allow you to isolate and run on bottles without shutting down the system.

DO a check on where the trips have been occurring and install an air tank. I'm puzzled though how the air supply drops then comes back. What sudden extra air flow caused the pressure to fall and trip the platform? Are those pressure trips accurate? Test all the PT's and pressure switches.

But always think - Am I testing to find faults to fix them or am I just wasting time / money and just replace the whole section?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[EdStainless]

You need to focus on areas where there have been issues and find the problems there. You need to know what you are looking for before you go hunting. In a system of this age it could be anything from corrosion damage at fittings to mechanical damage issues, to leaking valves or sensors.
Ultrasonic depends on background noise levels also, but I have seen it done over some distance, and by connecting sensors to lines.
I like the ides of a small amount of tracer gas in the air, maybe SF6 since it is inert and easily detectable.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[0707]

I think tracers are the comum use in these cases.

luis

 

[Hans85]

@3DDave
Thank you for the feedback. We've previously went through the IR camera route for Gas Leak detection via a service provider company as the equipment is too expensive. I'm personally not involved in that but the feedback that it has limited success rates.

@LittleInch
Very long and insightful feedback, I'll try to answer what I can.

"Your assumption seems to be that the tubing itself is corroded / damaged? Why do you think this?" - We've has cases where leaks are caused by corrosion. There are also cases where leak occurs post blasting activities and there are signs of impact damage (tubes out of the holding bracket).

"Most leaks come from screwed connectors / regulators / leaking valves." - That's true. But for these they are likely easier to trace and usually we'd be looking at it first. The issue is that we only do this as reactive measure/maintenance. We would like to have preventive measures, especially when it comes to tubing in which there isn't any specific PM plan.

"Challenge the planned shutdown issue. To me if your air system is causing unplanned shutdowns then it has elevated itself to a very high priority." - Personally I agree with you, but we're dealing with corporate people and bureaucratic environment in which $$$ are the actual priority. E.g. "Safety comes 1st" for the tagline of most companies, but if we've worked long enough we will not that is not always the case.

"What can you check in terms of accurate pressure guages at the headers / instruments / valves. There are usually test ports/ vents/drains to tap into." - This suggestion might be do-able. Nonetheless, it would be difficult to gauge in terms of the branched out tubings though; back in old days there's no isometric drawings for tubings. A few years back I was involved in a project where even tubings are modeled in the PDMS.

"If a lot of the tubing is inaccessible then why are you bothered about inspecting it? Either program in a replacement or live with it and fix all the leaks you can." We don't have any replacement program in place. Need to bear in mind that these platforms are old and not producing much. This initiative started as there has been more frequent tubing (not fittings) leaks, and tracing tubing leaks are much more laborious.

"DO a check on where the trips have been occurring and install an air tank. I'm puzzled though how the air supply drops then comes back." - Usually what happens is there will be either PSD or ESD, then find the source of the leak, fix/repair, then bring the platform back online.

"But always think - Am I testing to find faults to fix them or am I just wasting time / money and just replace the whole section?" - I'm genuinely interested in solving problems/issues. But for this case it may be more of creating a "plan" to satisfy my boss, in which he would be able to answer to his boss.

@EdStainless
"You need to focus on areas where there have been issues and find the problems there. You need to know what you are looking for before you go hunting" - I wish this was the case, but the areas vary. Only by reaction are we able to narrow down to which system/area.

"Ultrasonic depends on background noise levels also, but I have seen it done over some distance, and by connecting sensors to lines." - I'm still looking into this possibility as well, if there are improvements in ultrasonic devices that can trace leak from a distance.

@EdStainless & @0707
"I like the ides of a small amount of tracer gas in the air, maybe SF6 since it is inert and easily detectable. + I think tracers are the comum use in these cases." - Does this mean I would need to have a gas detector as well and will it be able to trace/pinpoint the leaks from a distance away?

Thank you all for the feedback and suggestion.

Warm Regards,

Hans

 

[georgeverghese]

Leak detection would be easier if there were sectionalised isolation valves so you can identify which sections the leaks are from. Else it may all seem too difficult on a large facility, and with all the gridlock in the office. Maybe simpler to install another IA compressor and IA dryer if you have the space for the equipment and the starter cubicle in the MCC.

 

[3DDave]

It would be best to write up a requirements document. Here's a start:

Cannot touch the equipment.
Cannot cost more than $X, prefer zero cost.
Cannot require any expertise to use.
Must produce exact locations.
Must continuously monitor.
Cannot add any equipment.
There is no access or sight line to potential leak sites.

There are X meters of tubing
The tubing is in (Good) (Poor) (Is not recognizable) condition
There are X number of fittings
There are X number of solenoids

The current rate of failure is X per (day) (week) (month) (year)
Most of the leaks are determined to be (corrosion) (loose fittings) (mechanical damage)

The usual repair is (Replace line) (splice in a piece with new fittings) (use sealant)

Change or remove what doesn't apply, add what is missing.

Just helped a guy today in a similar situation. Old beater car where a wheel came off because the studs were stripped. All four tires were in rough shape, several with significant tread separation and/or enough wear to expose the belts. No wheel had all lug nuts; aside from the one wheel with zero, two had 3 of 5 and the fourth had 4 of 5. The rear window was gone. The battery was missing the hold-down and the owner mentioned it had shorted the positive terminal when it shifted during cornering. His father came to help and pounded out the damaged studs; I expect the wheel bearings got some damage from that. Did not buy any extra lug nuts for the other wheels.

Honestly, I'd do nothing at all. The most likely thing is the owners will transfer the operation to a shell company, spin it off independently, declare it insolvent and, if it's in the US, let the EPA deal with the cleanup. That seems to be how this sort of problem is dealt with.

 

[LittleInch]

Ok, I think we get the picture.

The place is falling apart and you're all rushing around applying sticking plasters to the various wounds the platform has as it reaches the end of its useful life, but with less and less throughput, there's less and less money to spend on it.

I suppose I missed the bit really about PM on the Stl Stl tubing - assume this is like the 1/2" stuff? TBH, other than a pressure test (air or N2 should be ok at that size) I would struggle to come up with any cost effective systems to apply fixes before it fails.

Of course then the issue is you find a fault / section about to fail / a hole etc - what do you do? / can you replace it?. Many tubing routes seem to be in hard to get to places. Hence why this reply "We don't have any replacement program in place." is the key issue. Rather than spend time and money trying to find the problem and then more time and money fixing it whilst the platform is shut in, if you don't identify and put in place a program to replace those locations where it is difficult / time consuming / costly to inspect or repair, then you won't get anywhere - IMHO.

To get the system back on line faster I think you really should use or install in a shutdown a whole series of isolation valves on the tubing complete with quick connect valves upstream and downstream of this isolation valve so that you can quickly isolate the leaking line, apply air / N2 pressure from a set of bottles or hoses and then either operate like that or fix the broken tubing in your own time.

This could allow you to do testing of individual circuits on line by still supplying air / N2 to the devices whilst the tubing is being tested.

But PM on stl stl tubing? Don't think anyone bothers normally.

Good luck and keep us in the loop.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[1503-44]

A circus master might help.

 

[Strong]

I have used ultrasound for compressed air leak detection surveys. This new gadget may be of use for the type of survey you are considering:

https://www.fluke.com/en-us/product/industrial-imaging/sonic-industrial-imager-ii900

Try it before you buy it!

Walt

 

[MJCronin]

3DDave says it all ....


Honestly, I'd do nothing at all. The most likely thing is the (MBA) owners will transfer the operation to a shell company, spin it off independently, declare it insolvent and, if it's in the US, let the EPA deal with the cleanup. That seems to be how this sort of problem is dealt with.o

There comes a time in the life of any machine/plant/facility when you are simply done. Things are beyond repair .... Sounds like your failing instrument air system is a final twitch of a dying patient.

Additionally, you should take some comfort in the fact that abandoning an old failing rat-trap and replacing it with a new facility will be safer

Bear in mind that the international markets for cheap investment fueled by oceans of zero interest money attracts the most aggresive sharks (ummm.. I mean MBAs)

Let us never forget the plant investment credo taught in business school and printed in gold leaf on every MBA diploma: Omne perpetuum est .... or, for those who cannot recollect their high school Latin; Everything lasts forever !!!

Sorry Hans,..... These are the morning rantings of an old engineer who has been in your place before and has not finished his coffee ....

Good luck and best regards


MJCronin
Sr. Process Engineer

 

[Hans85]

@3DDave & MJCronin,



I can't disclose which company I work for, but there is intention to sell-off the assets; it is already underway and put out in the market. Plenty of em loyal old folks here are very unhappy about this, as there are still a lot of uncertainties with regards to their status (i.e. gratuity, etc.). Most staff are most likely going to be treated like commodities or sold off to the buying company.

 

[LittleInch]

That's the way it goes.

If you're currently a "najor", then aging assets become a liability and loss making.

Moving to smaller companies brings reduced benefits, but can give adding assets a new lease of life as company costs are a lot lower so more money to spend on the asset itself and extend the life to as ĺong as possible.

But the individual usually becomes collateral damage.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

Same scenario played out here in AU when an aging FPSO in the Timor Sea was sold off to a fledgling company Northern Endeavour 4-5years ago, and now this little Co has gone belly up, leaving the taxpayer to foot the >AUD$200million decomm costs.

 

[Hans85]

@Strong (Mechanical) i.e. Walt.

We went with Fluke. Used it on offshore yesterday. Works like a charm.

Thank you again for introducing me to this new technology.