Tube leakages in 4 Pass Heat Exchanger

[processrj]

There is a 4 tube pass/1 shell pass heat exchanger experiencing failure (leaks) at some especific middle tubes of the 1st pass (6 tubes out of 125 total in the 1st pass). Localized corrosion seems to be taking place on the tubes outside, toward the floating head. Process conditions / are:
Tube Side: Diesel, in: 600 F, out 395-400 F, Total sulfur = 3 ppm ,
Shell Side: Gasoil, in: 200 F, out 300 F, total sulfur = 2.6 w %
Total number of tubes=500 (125 each of the 4 pass). Tube lenght: 6 mts. Material: CS (SA 179)
Temperature profiles gotten from process simulation are lower than 410 - 420 F near the location where the tubes have being found holed (at 5 mts from the inlet or 1 mts from the floating head end), on the contrary where the predicted tube wall temperature is higher (towards the fixed tubesheet end), no significant corrosion/metal loss is found.

I would appreciate if this comunity might provide some insights on the causes of these tubes failures. Thanks!

 

[Dhurjati Sen]

Dear processrj,

It is difficult to pin-point any particular cause of failure for these tubes apart from general corrosion. CS is used widely for such service and has given excellent result w.r.t time vs cost parameter.

The general practice is to plug the tubes at any available opportunity if the pressure drop rises across the h/e indicating leaks.

However, do check the baffles and the tie-rods are as per drawing because localized flow could be one factor that might aggravate the failures.

Regards.



DHURJATI SEN
Kolkata, India

 

[cap1a79]

Check the corrosion rate of sulfidation by API RP 571 at the outer surface of tube . I believe the tube outer surface is susceptible as combined with high temperature coming from tube inside and high sulfur content from Shell.

For example, if tube the outer skin temperature is 550 F with 2.5 wt% sulfur, corrosion rate will be 8-10 mils/year.

Considering tube does not have corrosion allowance generally, the corrosion rate as 10 mils/year will be enough reason for the failure. (Tube - No. C.A. as heat transfer, Pipe - with C.A. as mass transfer)




Lee SiHyoung,

 

[mrfailure]

You really should extract and evaluate a leaking tube if possible to identify the cause of the leak. If this cannot be done, then I would suggest running a boroscope on both the tube ID and on the shell side to look at the nature of the leak. This is too speculative without an evaluation.

 

[processrj]

Thanks Dhurjapati, capla79 & mrfailure for your feedback.

Responding to some of your comments, we did have thought of sulfidation taking place, but as mentioned, temperature at outside tube wall is lower than 450 F, at which this phenomena occurs. We are sustainably around 420 - 430 F. Using API 751, we estimate the corrosion rate being at 3 mpy, however for the mentioned tube to have leaked, the corrosion rate should have been around 7 mpy. We measure the OD and the tube thickness for the tubes that leaked finding corrosion rate of 5 to 6 mpy (some others 1st pass tubes had 2 mpy)

Past inspections on some tubes of the same pass have revealed the following estimated maximum mpy:
Inspection at 3.5 years of service: 0.8 mpy
Inspection at 9 years of service: 0.5 mpy
Inspection a 12 years of service: 5 -6 mpy at the leaking tubes

Based on above, it would look like the corrosion rate for the 1st. pass tubes have 10 folded increased in 3 years. What might have caused it?

Another question: What is the typical expected service life for CS bundles?

Thanks again!

 

[cap1a79]

Lifetime of CS tube bundle is generally 5 years. The reason behind is tube which is for heat transfer does not have corrosion allowance while pipe which is for mass transfer does have it.


Lee SiHyoung,

 

Pay for a root cause failure analysis.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."