Gasket failure heat exchanger 1

[leon_har_dt]

Hi!, i'm a new member.
First, i want to make a recognition and gratitude to this forum and the people who participate in. I always can find some guide to solve my problems and that it's thanks to the community.
I'm facing a problem wich I could not found a guide by searching it here (beside that english it's not my first language i did my best effort)
In my first year of engineer(mechanical) i had to design two heat exchangers who work with water(shell) and oil(tubes)(water heats the oil).
The design temperature and pressure it's 266F(130C) and 145psi(10bar)(for both circuits). I selected a PFTE gasket chesterton ecs-b e=1/16" for a B16.5 Slip on flange SA182 F304L.(the tubesheet was made of a blind flange, same characteristics). The HX was tested in the shop at 13bar(188psi) and didn't present leak
The problem begins when the customer was running their test and one of the HX leaked in the flange joint, we change the gasket days latter. The other one didn't present any leak and was use in the plant operation. Then the second one present leak, and analyzing the gasket(picture down below) it fail in a very similar way than the other
The customer said that the circuit consist of a recirculation circuit 1 bar pressure, and an impulsive circuit 7bar pressure (maintaining some vessel level of fluid)

Here you can see the deformation of the gasket when the second HX leaked days latter when then plant return to operation(this gasket was not the one who was replace). Red lines pictures the original form of the gasket.
After writing the context. ¿Could this effect be caused by water hammer or surge when the system change from 1bar to 7 bar or vice versa?
or ¿could be when the plant return the operation and the sudden change in pressure deform the gasket?
What else could be the reason?

I'm analyzing the change of gasket material and even the type of gasket being compatible with oil and B16.5

I would be grateful if you could expand my mind in this problem

L.

 

[r6155]

1) I guess hot insulation is considered, then it must be ensured that there is no leakage.
2) Did you witness the hydrostatic test (ASME VIII-1 UG-99)? Did you check any component during manufacturing?
3) Did you check the calculations?

Regards

 

[0707]

I finish this thread saying that a solid PFTE gasket in your case, was not the good solution, even for alimentary purposes, there are other better options.

regards

 

[desertfox]

Hi Leon

I think the gasket might have been overnighted, if the torque values you were given for the joint were meant to be for unlubricated fasteners and you stated you lubricated them, then the joints would see excessive tension with the same torque value due to the reduction in fastener friction.
On top of that you then put the joints in service and subject it to additional external loads.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[FacEngrPE]

Each gasket type has a recommended gasket clamping force.
Garlock published their recommendations for their products as a spreadsheet

https://legacy.garlock.com/sites/default/files/documents/en/Custom-k Calculator.xls

attached below.

PTFE is rather soft unless filled. In the picture, I see a gasket that looks like it slipped outward, I suspect the clamp load was too low for the pressure / temperature combination. Use of torque wrenches is recomended.

 

[LittleInch]

A few things struck me here.

1) It's note easy to see what is going on with only two small pictures. A drawing of the whole exchanger would help a lot.
2) What is that welded bar / plate across the middle of the flange? That is going to result in some high stress / forces as it tries to expand.
3) Is this just one end and the other end is the tube sheet?
4) If you're the manufacturer, how do you design these?
5) A slip on ASME B 16.5 flange looks unsuited to a HX type design. Why not use some of the TEMA flanges. B 16.5 uses a thicker flange and more widely spaced bolts than either the EN type flanges or TEMA where the flange is smaller / thinner and you have more bolts which helps spread the load around.
6) Bolt up is critical in these, both in terms of how it is bolted up in sequence and torque / axial compression on the bolts, though there is not always good correlation between torque and bolt stress.
7) That material just looks wrong - too flexible, might creep under pressure and no support for when it decides to move.
8) DO you not have a good gasket vendor to discuss this with?

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

 

[leon_har_dt]

Hi!
i didn't had time to answer until now.
Desertfox-> i would be more caution when it comes to tighten the bolts, to ensure that would not happened.

->FacEngrPE Thanks for the recomendation, i would read it.

->little Inch 1) i would upload some drawing when i had more time
2) That plate it's a partition plate of a heat exchanger, and yes, we were cautious about this plate
3)Yes, this was the one of the header of the heat exchanger
4) We mainly use compress for mechanical calculations
5) because the material available in the industry(here we cannot count on tema flanges)
6)we do torque the equipment with bolt sequence and torque parameters
7)you mean the gasket material?
8)yes, that's a good option, I would definitely discuss with them

Thanks

 

[Aconnell]

Hello Leon,

Right away I would say this is an issue of gasket selection, PTFE gaskets are prone to cold flow or creep and loose seating stress even at relatively low temperatures and often require additional torquing after initial service. Others in this forum have already given you some good advice but if you wanted a good reference on gaskets in general, I suggest the document at the link below which you might want go through before you speak to your gasket vendor. If you look at page 30 specificially it talks about some of the issues with PTFE gaksets.

https://www.fluidsealing.com/wp-content/uploads/FSA-Gasket-Handbook-June-2017.pdf

 

[SnTMan]

Aconnell, nice material, thanks

The problem with sloppy work is that the supply FAR EXCEEDS the demand