7 replies to this topic

[jprocess]

Dear all,
enterance of light hydrocarbons like propane in flare network can lead to low temperatures and consequently increase the possibility of hydrate formation.
Any hint about the solutions?
One way can be the steam or electrical tracing of headers.But for what lenth of network we should do it?It does not seem to be logical to trace whole the network length.
Thanks in advance.
Cheers.

[joerd]

You have a very valid concern. It is common practice to segregate the flare systems, and make a cold collection network and a wet collection network. Preferably, these have segregated flare stacks as well, but I have seen applications where the cold flare header is sent through a heater (always on! - 100% reliable) and then tied into the wet flare system.

[jprocess]

Dear Sir,
Thanks a lot for your reply.
I should mention that your idea is valid at design stage of a flare network but my question is about an existing one.On the other hand my problem is a kind of troubleshooting and not sizing.
cheers.

[JoeWong]

Dear Sir,
Thanks a lot for your reply.
I should mention that your idea is valid at design stage of a flare network but my question is about an existing one.On the other hand my problem is a kind of troubleshooting and not sizing.
cheers.

jprocess,
joerd has advised a common, workable and practical approach for most flare system...Segregated hot & wet flare header (sometime even splitted in Sour wet & Non-Sour wet) and Cold & dry header...

If your system is existing facilities and no way you can change the arrangement...first you may check if you should consider Hydrate formation is possible in practical sense eventhough theoritically it will.
There aer several factors you may like to consider :

1) Is discharge inventory (hydrate) sufficient to plug the flare line ?
2) Device discharging fluid at very high speed...can hydrate have time to settle on the pipewall ? can fluid momentum sweep-away the formed hydrate ?
3) Flare piping is a heat sink...Hydrate form in gas phase will destroyed by itself once it contact with flare piping. Can hydrate stick on the flare piping wall and cause blockage with a limited relief time ?
4) Is your discharge is sustainable ?
5) Heat transfer between gas & flare piping wall is poor...High speed gas (PSV/BDV discharge) make it worst. Can heat from tracer has sufficient "transfer time" ?
6) If you can hydrate formation is potential and heat tracing is provided to eliminate this problem, you may have to consider a highly reliable heat tracing system e.g.emergency power supply, High SIL loop, etc
.
.
.
To answer above questions, i find it very difficult...

I would think you should consider the general approaches.
i) Segregate Hot wet fluid with cold Dry fluid as much as you can
ii) Avoid small and short discharbe pipe
iii) minimise elbow tee before it end at large subheader
iv) increase reliability of high pressure tripping system to minimise relief events
.
.
.

Some engineer has considered (infact has implemented) glycol injection at BDV outlet...It was published in Hydrocarbon Processing if i am not mistaken. Can't remember right now. I will try to find out and inform you this paper.

BTW, if you are handling cold fluid, i would remind you to check the potential of extenal moisture freezing in the PSV vent chamber and result it stuck close...



Hope this help.

JoeWong

[joerd]

Some engineer has considered (infact has implemented) glycol injection at BDV outlet...It was published in Hydrocarbon Processing if i am not mistaken. Can't remember right now. I will try to find out and inform you this paper.

Is that the article by Zeydabadi in the April 2006 issue?

[JoeWong]

joerd,
You are absolutely right.

"Prevent system hydrate formation during sudden depressurization - Using this technique can help size gas plant inhibitor packages"
B. Asadi Zeydabadi, M. Haghshenas, S. Roshani and M. Moshfeghian
Hydrocarbon Processing April 2006 page 83-91

JoeWong

[Wouter]

You have a very valid concern. It is common practice to segregate the flare systems, and make a cold collection network and a wet collection network. Preferably, these have segregated flare stacks as well, but I have seen applications where the cold flare header is sent through a heater (always on! - 100% reliable) and then tied into the wet flare system.

We have an application like that, where we have a "cold flare header" with line heaters coming back into a common flare header.
If you are using ultrasonic flare flow measurement, make sure that you keep the steam tracing away from the measurement, it would screw up the signal

[ogpprocessing]

You have a very valid concern. It is common practice to segregate the flare systems, and make a cold collection network and a wet collection network. Preferably, these have segregated flare stacks as well, but I have seen applications where the cold flare header is sent through a heater (always on! - 100% reliable) and then tied into the wet flare system.

You mean a heater is considered as part of the flare network?! If so, where this heater is exactly located and what is the heating medium? What to do at the onset of utility (heating medium) failure?