9 replies to this topic

[reihane1]

Hello Everybody,

 

I need to redesign a safety valve on Ethylene pipe in gas phase (3 ich pipe-Length:5 meter). Scenario is thermal expansion. My questions are.

 

1-Is it realy necesary to have a safety valve on ethylene pipe for thermal expansion?( it is in gas phase P=80barg ,T=Enviromental temperature)

 

2- Temperature downstream of valve will be around -100 deg C, should the pipe material be SS or because of very low flow rate the pipe will not crack (safety valve releases into 6 inch pipe)?

 

3- does heat tracing of downstream pipe helps? can we use heat tracing for CS pipe instead of using SS pipe?

 

 

Many thanks

Rayhaneh

[Bobby Strain]

You should resolve your first question. If the answer is yes, then apply good engineering practice to the installation. And adhere to codes and standards.

 

Bobby

[Art Montemayor]

reihane1:

 

I agree with Bobby's succinct and thought provoking comments.  But first, go back to basics: your scenario word, "thermal expansion" does not apply a pipe filled with gas.  Thermal expansion is a word reserved in the safety valve world for a 100% LIQUID filled, blocked-in system - not for gas.  Make sure you have the right terminology when you specify your needs to a safety valve supplier.

 

You haven't defined your system.  What does the diameter and length of the pipe have to do with the alleged need for a safety valve.  First define all the details of your safety scenario, identifying the need for a safety valve.  You haven't even stated if your system is static or continuous flowing gas.  If the ethylene gas is flowing, what does the temperature downstream, the pipe material, or the supposed need for heat tracing have to do with the need for a safety valve?   Identify all the details of your system and the credible, overpressure scenario first - just as Bobby points out.

[reihane1]

Thank you very much for the prompt replies. yes you are right Thermal expansion is considered for liquid only (or liquified gas)  and exactly for this reason I think releif valve is not necessary for Ethylene pipe (not in a cryogenic condition ). The problem is that already there is a safety valve installed on the pipe and in the old data sheet of psv ( dated 1971) , the scenario defined as thermal expansion . I can not think of any other Scenario. The only possible  Scenario is fire. I personally never designed a psv for a pipe for fire case. Do you think fire case shall be considered for pipe as well?

 

regarding secound and third questions : we have a continious flow of the Ethylene ( we receive Ethylene via a Pipe line form our supplier). On a few meter pipe ( about 20 meters) there are 4 PSVs.There are two  PSVs which are designed for other Scenarios. These PSV are releiving into our flare system ( pressure in flare sysrem: 0.5barg, material: CS). I think at least a few meter of the flare system should be SS ( releif temperature can be as low as -80 deg C) or when it is not possible at least should be heat traced ( although I am not sure it helps). Otherwise I think there is a chance of brittle fracture in flare network if psv opens. What do you think?

 

Thank you for your Help

Rayhaneh

[Bobby Strain]

Rayhaneh,

       Where is this facility? I don't want to go anywhere near it. It's quite a dangerous situation. There are other overpressure prevention methods that don't require a PSV.

 

Bobby

[reihane1]

 Bobby,

 

can you tell why you think the situation is dangerous? Because of relief valve downstream piping material or something else?

 

Regards

Rayhaneh

[MrShorty]

I think Bobby's first question is right on the money -- first determine exactly what is in your pipeline. You have called it "ethylene gas at P=80 bar (Pc =  50 bar according to DIPPR) and T=environment (Tc=282 K or 9 C according to DIPPR)". I am reminded of a day in an early PChem class where we talked about what to call a fluid that is well above the critical pressure but near the critical temperature (note that we decided not to call it a "gas" without some careful clarifications).

 

At 80 bar, there would certainly be temperatures (maybe above 290 or 300 K) where ethylene might behave more like a gas.

At 80 bar, there would certainly be temperatures (maybe below 280 or 270 K) where ethylene might behave more like a liquid.

Even though there is no clearly identifiable phase boundary, there would be a transition region where ethylene's behavior is ambiguous.

 

You have not said what your expected environment temperatures are. 250 to 320 K would seem to cover most of the planet, though, so it is not difficult for me to hypothesize that the original engineers anticipated that the pipeline could contain ethylene in a liquid like state and designed it for "thermal expansion" with appropriate safety valves.

 

As Bobby said -- you need to figure out exactly what is in your pipeline. I would go further and suggest that you pull up your phase diagrams and/or equations of state and/or whatever else you want to use to estimate the PVT properties of ethylene and figure out exactly what is in that pipeline at your expected conditions and what happens if/when the pipeline is blocked off and allowed to warm up.

[PhilippM]

You're wondering why this situation sounds dangerous? First of all, you have a relief valve with an old data sheet from 1971, and no one really knows if it is required at all and for what overpressure scenario. This in itself is a problem, as you shouldn't have any relief valve that isn't actually necessary, but furthermore you (and by "you" I don't mean you personally but the company/organisation operating this plant) obviously don't understand the process safety of your own plant. Maybe there are other pieces of equipment that actually need overpressure protection but don't have it.

Edited by PhilippM, 14 August 2019 - 12:22 PM.

[reihane1]

The situation is not so bad. Unfortunately always there are some cases which are not so clear and exactly because of this we are reviewing all safety valves again. Anyhow we could find the overpressure Scenario. It was not thermal expansion. The reason was another PSV which could not be connected to blowdown system.

[winchester427]

1) Sounds like Ethylene will be in the dense phase and therefore thermal relief could be a credible case. You could check this by calculation by heating the fluid along an isochoric pathway and seeing what pressure you get at the higher temperature. Normally thermal relief happens slowly and API codes suggest a D orifice is sufficient but you would be best to confirm via calculation.

 

2) @ -100 degC yes you would need a stainless steel tail pipe.Although this temperature sounds like an adiabatic flash temperature to ambient. You could model the tailpipe to see what the coincident back-pressure and fluid temperature are expected to be. Exit temp should be > -100degC once you account for the back-pressure but likely still will need stainless. I think carbon steel lines can only go down to approx -50 degC. You will probably also need a stainless steel relief valve due to "cold creep"...i.e. where heat is conducted forward along the tail-pipe and the valve itself chills.

 

3) heat tracing is not regarded as reliable enough to do this. I guess it is feasible but then you would carry a large risk of cold brittle fracture if the trace heating fails. You might have to consider administrative controls if the trace heating is offline the section of line is taken out of service and de-inventorised (not an enjoyable situation for operations/management/finance guys).