Rupture disc frequent failure

[amar122amar]

We have an ethylene system pre heater, shell side low pressure (LP) steam (4-5 kg/cm2), tube side ethylene (inlet pressure 40 bar, inlet temperature 30 deg C, outlet pressure 40 bar, outlet temperature 100 deg C). Ethylene outlet temperature is maintained by shell side LP steam. There is a temperature control valve at steam inlet. Shell side condensate goes to condensate header through a steam trap. Rupture disc is given at shell side (steam side) with set point of 11 kg/cm2 (considering tube leakage scenario). Rupture disc outlet goes to flare. In past one mont rupture disc has ruptured twice. We hace checked probable tube leakage but could not find any leak. LP steam header pressure remains in the range of 4-5 kg/cm2 always. What may be the other reasons of failure. Rupture disc is checked and there are no signs of aging.

 

[Latexman]

Is this a new problem, or has it always been this way? Any recent changes?

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[amar122amar]

Steam preheated is taken in line recently. Soon after this problem started.

 

[MortenA]

Sound like a transient phenomena similar to a water hammer (steam hammer although that is believe is geometry dependent)

 

[don1980]

A safety valve is the right type of device for this application. As you have experienced, disks have the inherent liability of premature failure risks. The most sure way to solve the problem is to remove the disk and install a PSV - that's what I would do. Avoid using disks unless there's a good reason for not using a valve.

For disks that discharge to the atm there are a number of possible explanations for disk premature failures. There are even more possible explanations for disks that discharge into a flare header, such as this one, and that's another reason to not use a disk for this application.

Here's a list of possible explanations:
[ul]
[li]Burst temperature specified too low - the specified burst temp should be maximized (should correspond to the steam temperature)[/li]
[li]Disk metallurgy - chose a material that has a relatively low sensitivity to temperature variations, such as Inconel 625[/li]
[li]Specified burst pressure - This specified value has to account for the backpressure in the flare header. To ensure that the disk doesn't burst at a value above the vessel MAWP, one has to choose the upper end of the expected pressure range in the flare header. However, this means that when the flare header pressure is lower than that value, the disk will burst at a lower pressure.[/li]
[li]Lack of vacuum support - Since the disk is routed to a flare header, it probably needs this mechanical support.[/li]
[li]Metal fatigue due to pressure fluctuations on either side of the disk[/li]
[/ul]

 

[LittleInch]

amar,

I appreciate English might not be your first language, but can you explain this sentence a bit more as it doesn't make sense to me.

"Steam preheated is taken in line recently"
What was it before?
And what exactly has changed?
Is there a steam outlet or is it a simple steam and condensate issue?

Have you checked operation of the steam trap?
Could the shell have become filled with condensate?
Do have a record of pressure in the actual shell?
Do you have a record of pressure in the flare header?
Is this a continuous service or subject to temperature transients?

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

 

[amar122amar]

Dear little inch,
This preheater is used to heat ethylene feed supply to ethylene de-oxo vessel.Earlier ethylene de-oxo bed was in bypass condition so does the preheater. Now we have take de-oxo bed and steam preheater in line. There is a steam inlet with temperature control valve on shell side. Condensate outlet is having pressure pump steam trap. As you mentioned what are the implications if shell becomes filled with condensate? We don't have pressure transmitter on shell side. Flare side we pressure record and it is well within the range. Sorry for my English. Thanks.

 

[amar122amar]

Dear don 1980,
Steam side operating temperature is 140-150 deg C. Burst temperature specified is 100 deg C. Burst temp specified in data sheet is 13 kg/cm2. Flare back pressure is considered 2 kg/cm2. Stamped bursting pressure is 11 kg/cm2. As you mentioned what is vacuum support?

 

[amar122amar]

Bursting pressure in data sheet is 13kg/cm2.

 

[LittleInch]

"Steam side operating temperature is 140-150 deg C. Burst temperature specified is 100 deg C."

Kind of points towards an issue. Have you checked with the vendor what the impact is??

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

 

[Latexman]

Can you attach the P&ID?

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.