Fin coolers bend fretting under vibration. 3

[fifo62]

Hi Gents


I have in front of me n°6 fin fan cooler ( three bays / two per bay ) in sour service (high H2S contents), design 120 bars / 115°C, UNS N08825 ASME VIII div.2.

Four pass with return U bends, 1"1/2 15 BGW (1.8 mm), at the fin tube hedges. Very critical service. Bundles are already built but seems to me something is wrong since the beginning: space and process constraints led the suppliers to the decisions to do not have gaps between the bends that are staggered to match a triangular disposition of tube axis on different layers. (see attachment)

Bend sides are touching one the other. Furthermore these tubes and bends are not properly supported because the used fin tube support boxes allow side movements and are not reliable.

My concern is about the induced vibrations during service (it is supposed coming from process, fans,..) in conjunction with the possible free movements depending on supports and assembly tolerances. The bends will rub together resulting in the potential loss of containment of sour gas in the event of bends failure. There is potential risk for personnel and can result in fatalities.

Support boxes are not tight enough to reduce vibration and leave bends susceptible to rattling around even with low vibration.

Possibility to move is random depending also upon the construction tolerances different for each tube/bend/support to the other.

In fixing the support (very difficult)we may get some advantages but the touching bends to me are very bad in such a service.

Monitoring and measure the wearing it seems complicate because inaccessibility of hundreds of contact points.

The supplier has no references for such tube routing and supports and propose weak justifications (insufficient / wrong) to demonstrate these coolers are safe and suitable for the intended service.

On the basis of my experience and, I think, on a sound engineering practice I want to reject as not suitable for the purpose these device but, it would be easier to sustain position by having Code references and/or papers to address fretting and wear phenomena under vibration.

Any one can suggest on this? Any one experienced something similar? Do anyone knows what may happen? I'm searching for literature references and advice.


Thanks to all

 

[gr2vessels]

1. I suggest first to review the design and establish the compliance with the design code nominated on the drawings. If unsure, hire a competent engineer / consultant to do the review for you. The design code should be API 661 / ISO 13706, the fabricator must have ASME U2 stamp and before the fabrication commenced, you were suppose to provide the fabricator a certified User’s Design Specification. The Authorized Inspector which has been nominated by the fabricator (or by you in agreement with the fabricator) must have completed by now the review of the design and satisfied that the code requirements have been followed.
2. I assume that given the service of this cooler, a competent engineer (perhaps process engineer) familiar with HTRI or similar software has completed the check of the thermal designed provided to you by the fabricator or his process designer. Otherwise you won't be able to install and operate a such equipment.

Please note that all the above are only samples of a generic procedure for the fabrication of an air cooler in severe H2S service, stamped U2. There are many other requirements, please review the ASME VIII, Div 2, Clause 2.2, 2.3 and 2.4.
Cheers,
gr2vessels

 

[4Pipes]

Are these from Korea? I had a similar situation with regard overall system design of air fin HEs. The easy micro details like tube thickness and header boxes were fine. Associated documentation for these details was very slick. However, the overall system assembly "design" was a very different matter. You are right to be concerned.
It is very difficult to impose your will in these situations. The codes do not always give specific design control in the "difficult" areas. You will probably have to resort to the catch-all code statements to consider all loads etc. Don't give up. This is what the manufacturer will be hoping that you will do. He wants them fob.
If you do give up, your next argument will be with the operator who might not allow commissioning until you can prove everything is safe. It would not be the first time.

You must never forget that, as far as you are concerned, you are the only expert. There will always be somebody else who you might think is more knowledgable. Listen to what they say but make your OWN decision. The chances are that they can screw it up far more effectively than you.

 

[fifo62]

Very kind from you to dedicate time in replying me

Decision and responsibility is on my desck 4Pipes, I know, and I'm seriously concerned: people's life is unique. I take in my hand your suggestion, gr2vessel and 4Pipes. it's a nice program for the week end. I will return back to ASME and ISO especially.
Until now I wasn't able to find something that gives clear indication within these codes; a simple statement about wear or contact I mean. This is my problem.
I'm not confident on the possibility to fix the problem of the unreliable support boxes in a way that respect a reasonable service life time and will never became dangerous. There are different tribology phenomena associated with vibrations, I mean: chattering, tube bend-slip.... Unfortunately there is no easy literature on this.

Do you think it migth be interesting to propose this to one other forum e.g. for material I'm not familiar with Eng-Tips procedure: propose the same could be unfair?

Nice week end to all

 

[saplanti]

There is another phenomena in that application is the pulsation in the piping system if the system is connected to a reciprocating compressor.

If the system natural frequency matches with the pulsation frequency rezonance is very likely. You need to keep the natural frequency within the API 618 recommendations. If you have access to API committee web site you had better get API 688 draft documentation to see the light better.

Hope it helps.

Ibrahim Demir

 

[fifo62]

After few days here my conclusion.

Reciprocating compressor not present, fortunately.

Non compliance to ISO 13706 as follow: 7.1.1.4 2"sagging an meshing" / 7.1.1.5 "hold down member" / 7.1.1.8 "air seal" / 7.1.6.1.3 non corect hold down members lead to uncontrolled load for tube / fretting phenomena due to vibration not addressed by the design.

To me bundles are not suitable to be put in service as they are. Any remedy is very difficult. Lesson lerned: avoid touching since the beginning by a correct routing of the tube/bends lay out.

Thanks to all for the interesting advices

 

[saplanti]

Another one could be the vortex shedding frequency under the ternal air flow. To make sure you need to check the tube natural frequencies at different locations and compare with the vortex shedding frequency.

Other than that the flow may be going into the two phase. This may cause the large vibration depending on the configuration of the tubing as well. If the tubes are not supported properly as you mentioned above the chattering and large motions will occur and fatigue on the metal becomes an issue.

Of course these are all the possibilities and can not be directly accepted without any process information and a calculation.

Regards,

Ibrahim Demir