7 replies to this topic

[Akhi99]

I am Process Engineer having 20 years of experience in the design of oil & gas, petrochemical & fertiliser process facilities.

I am doing a conceptual study for a gas processing plant for replacing an existing J-T valve by a Turbo-Expander for dew pointing of the export gas. Currently MEG is injected in the process gas upstream of JT valve to avoid hydrate formation. Existing plant doesn’t have a Glycol or molecular sieve dehydration unit. Setting up of a new dehydration unit may not be economically viable.

The expected temperature at the outlet of the expander is around -15 Deg.F. Hence the water dew point for the gas at the inlet of TE is to be around -35 Deg F (assumed 10 Deg C lower than TE outlet temperature). The pressure at the inlet of TE is around 950 psig and outlet is around 700 psig.

In this situation, I would like to know whether MEG injection is sufficient to meet the water dew point requirement and safe operation of Turbo Expander.

[paulhorth]

Akhi99,
The answer is Yes, you can use MEG injection with a turboexpander. The Statoil plant at Kollsnes in Norway, built 1996, has such a system, and I designed a similar system for a gas plant in 2002, which has been running OK as far as I know.
These plants had a temperature in the LT separator of -23 to -26 C, similar to your -15 F, at pressure of about 67 barg
You should check with the expander vendor, but as I remember, we were told that the machine could accept liquid droplets in the feed at a much higher mass flow than we needed for MEG injection.
However, you should note that Statoil found that the actual freezing point of MEG-water mixture in the presence of hydrocarbons was higher than predicted by the software available then, and they did some research on the subject.

Their paper, which is worth reading if you can get it, is:

Nordstad et al, A new approach to the use of glycol in low temperature high pressure gas processing appplications, GPA European Chapter, 1999

It's possible that modern simulation packages have now got better data, you need to use the best data that you can find for hydrate temperature prediction and MEG freezing points.

Paul

[Akhi99]

Paul,

Thank you very much for your fast response.

One of the vendors indicated that it is a common practice to use MEG injection for turbo expander application.

Is it advisable to provide a static mixer after the MEG injection point to ensure proper mixing of MEG with gas?

Do you have any idea how accurate is Hysys to predict hydrate temperature and MEG freezing point?

Thanks and regards,

Akhil

[ankur2061]

Akhi99,

Static mixers are routinely used for dispersion in the O&G field. You might want to look up the following link:

http://www.cheresour...s-need-to-know/

HYSYS is OK in predicting hydrate formation temperature at high gas pressures. But there is a school of thought that says HYSYS overpredicts the hydrate formation temperature at low gas pressures. There is another DOS based program called CSMHyd from the Colorado School of Mines which you can use to compare with HYSYS which uses the SRK equation of state for hydrate prediction and hydrate inhibition injection. It can be downloaded from the following link:

http://www.cheresour...__fromsearch__1

Hope this helps.

Regards,
Ankur.

Edited by ankur2061, 09 November 2011 - 12:54 AM.

[Akhi99]

Ankur,

Thank you very much for the detailed information on static mixer and Hydrate calculation.

Thanks and regrads,

Akhil

[Technical Bard]

The latest HYSYS release (v7.3) has improved hydrate prediction algorithms, and should be better than it used to be.

[paulhorth]

Akhi99,

I have located a scanned copy of the Statoil paper - see attached. This includes their measured MEG freezing point data.

Paul

Attached Files

•  Norstad & Gjertsen MEG freezing.pdf   7.83MB   172 downloads

[Akhi99]

Paul,

Thanks a lot for the report. It was very useful and the process conditions of the case study were very apt for my study.

Regards,

Akhi