9 replies to this topic

[clockwork]

Hi everyone,

I just graduated and started working a few months back in a company that mostly deals with FPSOs. Recently, my boss asked me to perform a hydrate study. I have done some work on it but i am not sure if what i am doing is correct. I would be very grateful if someone here could tell me if i'm in the right direction.

I know that hydrate formation depends on temperature, composition, pressure and whether the gas is at or below its dewpoint. I also know that large pressure drops can cause hydrate formation. So bearing this in mind, i have identified a few possible areas where hydrate formation can occur in the process.

Let me describe the process as briefly as i can. It is mostly gas processing. From the wells, the reservoir fluids go through the flowlines and manifold and is routed to an inlet separation system which consists of a 3 phase separator (Inlet Separator). The gas exiting the separator goes through a feed gas compression system while the condensate goes to the condensate stabilization system. The separated water is routed to the produced water system for treating. From the feed gas compressor, the gas (which contains H2S and CO2) is sweetened by absorption with amine (MDEA). Then the gas is dehydrated with TEG before going through a JT valve. From the JT valve, the gas is fed through a Low Temperature Separator to remove any liquids that might form. The gas is then compressed before being exported.

From the process, I have identified some areas where hydrate can occur:
1) Hydrate can form immediately downstream of the JT valve.
2) When the downstream gas processing facility isn;t in operation, the gas from the Inlet Separator goes through a PCV which is routed to the HP flare header. I think hydrate can form downstream of the PCV.
3) Hydrate can form downstream of the Feed Gas Compressor anti-surge recycle valve.
4) Hydrate can form downstream of the Inlet Separator condensate LCV (this is what my boss has told me).
5) I also checked for hydrate formation for the gas exiting the Low Temperature Separator.

I have used HYSYS to simulate each of these scenarios. The main process was already simulated so this is what i did:
1) For (1) above, I did not simulate the stream since it was already in the main process simulation. So i just checked the possibility of hydrate formation using Hydrate Formation Utility from HYSYS.
2) For (2) above, I copied the gas stream going out of the 3-phase separator. Then i decreased its pressure to the HP flare header backpressure (about 12.7 bara) using the valve unit operation from HYSYS. I then used the HYSYS Hydrate Formation Utility to check if hydrate will form downstream of the valve. The Inlet Separator operates at ard 66 bara if i remember correctly.
3) For (3) above, I copied the gas stream of the feed gas compressor discharge and decreased its pressure to the suction pressure of the feed gas compressor using the valve unit operation. The i checked for hydrate formation downstream of the valve.
4) For (4) above, my boss told me to decrease the pressure of the condensate exiting the Inlet Separator to the critical throat pressure. This has got me confused. I looked through API 520 Part 1 and while it does mention critical flow for gas, it does not mention critical flow for liquids. So i looked through a Masoneilan control valve sizing handbook which mentions that critical/choked flow for liquids occurs when the pressure in the valve falls below the vapor pressure. This means cavitation rite? The book also had an equation to calculate the pressure drop for liquid choked flow. I used this equation.
Was I right in using this equation?
5) For (5) above, the stream was already simulated in the main process so i just checked for hydrate formation.

I'm very sorry for the lengthy post. I sincerely hope someone can tell me if i'm on the right track.
If i'm not, I would very much appreciate someone pointing me in the right direction. One of the reasons I'm so uncertain is that I used the Envelope Utility in HYSYS to plot the Hydrate curves. Then i plot the operating temperature and pressure of the relevant stream on the same graph to show where the operating point is with respect to the curve. Some of the curves look really weird. And in some cases, especially in case (4), the operating point is so far from the hydrate curve (I suppose this is a good thing really since that means no hydrate formation) that i wonder if i'm doing the right thing. I especially worry that I might not be calculating the critical throat pressure right.

I'm also worried that the 5 identified scenarios are not correct. I identified them based on the fact that there is a pressure drop being experienced by the wet gas. My boss informed me of the no. (4) scenario. Can someone pls explain to me why number (4) contains a possibility for hydrate formation. Is it pressure drop as well?

Also, there were 14 main simulation cases, each meant to simulate different years of gas production. I am wondering how do i determine which is the case to use for my hydrate studies. What i did so far is to simulate (1) to (5) for all 14 cases. Is this correct?

As for the gas composition, it is mostly hydrocarbons from methane to C30. there is also H2S, CO2, He, N2, Benzene, Cyclopentane...

Thank you very much to anyone willing to help me. I'd be really grateful. I have attached my excel spreadsheet here. I haven labelled some stuff and it is not really neat but I thot some might want to look at the hydrate curve to see the weird curves i was talking about. At least i think its weird. Thank you!!!

Attached Files

•  hydrate_study.xls   95.5KB   256 downloads

[Majid-Process]

Hi
Hydrate was my Master Thesis subject. I have some interesting files for you if you wish. If you are ok shake your hand to send them.

[clockwork]

Hi
Hydrate was my Master Thesis subject. I have some interesting files for you if you wish. If you are ok shake your hand to send them.

Hi, Majid.

Thank you very much for your reply. I am interested in the files. I have no idea though what you mean by shaking my hand. Could you please send the files? I would appreciate it very much. My email address is colleen_thor@yahoo.com.my. (in case you need it). Thank you.

Another question i have is: does gas at sonic velocity necessarily cause hydrates to form?

[Zauberberg]

Another question i have is: does gas at sonic velocity necessarily cause hydrates to form?

Actually it is the way opposite - gas velocities which approach or exceed sonic velocity suppress hydrate formation, by affecting reaction kinetics. Twister technology for gas dehydration and dewpointing is based exactly on these principles.

For more info look at: Twister BV

[clockwork]

Thank you very much, Zauderberg, for your reply. I have downloaded the twister documents but I cannot open it. I also went to the website and downloaded directly from there but the same thing happened. But i will try to read about it if i can. I haven't actually browsed properly through the website but there should be some information there. I am still doing my hydrate study.

I would like to ask another question pertaining to hydrate formation. When you have a J-T expansion, does that mean that there is a hydrate formation risk? I know a J-T expansion leads to low temperatures but is that by itself a possible cause for hydrate formation? Or we have to look at the temperature and pressure conditions first? - meaning that not every J-T expansion is a hydrate formation risk?

Thank you to anyone replying.

[JoeWong]

Majid and Clockwork...seem like both of you are from MAL...

1) Hydrate can form immediately downstream of the JT valve.

General understanding is once the gas is dehydrated to 6-7 lb/mmscf, hydrate will not form. Thus, it will not formed downstream of JT valve. No issue with it. This concept is supported by thousand practical experiences...

2) When the downstream gas processing facility isn;t in operation, the gas from the Inlet Separator goes through a PCV which is routed to the HP flare header. I think hydrate can form downstream of the PCV.

Hydrate is normally form in this kind of arrangement. Typical example is Slugcather pressure dumping valve (PCV to flare system). However, this PCV is "normally no flow" or intermittent service. Good engineering practice is to provide larger PCV outlet line and made this line as short as possible before it is connected to main flare header.

3) Hydrate can form downstream of the Feed Gas Compressor anti-surge recycle valve.

This is possible. But the compressor downstream cooler outlet temperature may need to be adjusted to avoid hydrate formation.

4) Hydrate can form downstream of the Inlet Separator condensate LCV (this is what my boss has told me).

When liquid is passing LCV, the pressure will drop to minimum pressure (at "vena contrata") and recover to LCV downstream pressure. This process may lead to very low temperature and phase separation. Hdyrate and ice (in aqueous phase) may form just downstream of LCV. Thus, always check with vendor the minimum pressure and try to avoid...

5) I also checked for hydrate formation for the gas exiting the Low Temperature Separator.

Fluid is dehydrated. Thus no risk of hydrate.

[clockwork]

Hello, Joe Wong. Are you a Malaysian too?

At any rate, thank you very much for your detailed reply. I appreciate the trouble you took. Before you replied, I didn't really have the confidence to submit the report because I wasn't sure whether what i was doing was correct. I have since submitted my report. So i'll just wait and see how it turns out.

Thanks once again!!!

[Zauberberg]

I would like to ask another question pertaining to hydrate formation. When you have a J-T expansion, does that mean that there is a hydrate formation risk? I know a J-T expansion leads to low temperatures but is that by itself a possible cause for hydrate formation? Or we have to look at the temperature and pressure conditions first? - meaning that not every J-T expansion is a hydrate formation risk?

Hydrate formation downstream of J-T valve depends on the gas water content and pressure/temperature drop across the valve - it is as simple as that. If the gas is dehydrated deep enough, hydrates cannot form. Also, even in the case of completely water-saturated gas, hydrates may not form if temperature drop is not resulting in conditions below hydrate equilibrium curve. On the other hand, if you operate at cryogenic temperatures (-80 to -120C), almost complete dehydration of gas is required in order to avoid hydrate/ice formation. Your analysis is exactly the way how these conditions can be estimated, assuming that you are using appropriate software/correlation methods.

[clockwork]

Hello, everyone. Its been a long time. Thank you to Zauderberg once again for answering my question.

I just got back my report and there are a lot of comments

One of the comments my lead engineer made was concerning the gas from the Inlet Separator going through the PCV to the HP Flare header. What i did was flash the gas from its maximum operating pressure to atmospheric pressure and then check if hydrate formed.

My lead engineer commented: if the process is shutdown for a long duration but NOT depressurised and blowdown/depressurization is requested after the pressurized system has cooled down, at what upstream temperature would there be hydrate problems?

So what i did was take the same stream as before (the gas exiting the Inlet Separator at maximum operating pressure) and cooled it down to the minimum ambient temperature (as an initial temperature). Then I flashed this cooled stream to atmospheric pressure and checked if hydrates formed in the flashed stream. I kept changing the temperature upstream of the cooler until hydrates formed in the flashed stream. The highest temperature upstream of the cooler that resulted in hydrates forming downstream of the PCV (the flashed stream) is the temperature at which hydrate problems would occur.

I would appreciate any opinions on how i got my upstream temperature. What i did was simple so i'm not sure if it is correct. However, sometimes, solution can be simple.

My lead also commented: if the system is traced, what temperature does the tracing need to be set to to avoid hydrates?

I assume its a certain margin above the temperature at which hydrates would form. Is that correct? what is the usual margin?

I would appreciate any comments or if i have done something wrong, someone pointing it out to me.

All of you have been really helpful. Thank you very much!!!!

[Zauberberg]

If unit is shutdown and the system has been cooled to the ambient temperature, there will be a change in the system pressure. If there is no condensation taking place during cooling, you can apply the ideal gas law (P1/T1 = P2/T2) to calculate the final system pressure - assuming that temperature is equal to ambient temperature. If partial condensation occurs, you can use HYSYS to calculate final vessel/system pressure (the key is to find the pressure by using "Adjust" operation, at which overall mass density of the system is equal to the mass density before cooling).

Then, after blowdown, you'll encounter lower temperatures downstream of the blowdown valve - as compared to blowdown starting from normal operating conditions; therefore you will increase the probability of hydrate formation - depending on where are you on the hydrate equilibrium curve.

Best regards,