8 replies to this topic

[CheGuy]

we have series of equipments with different volumes and pipes connecting them. we would like to purge the system with nitrogen before startup. source of nitrogen is from a main supply header.

it is also required to size a line from the supply header to the inlet of the first equipment in the series of equipments.

can someone comment on the volume of nitrogen required and the the line design approach? and the time it would take to purge the system?

thanks

[Art Montemayor]

Guy:

What you are trying to resolve is the flow rate of Nitrogen required to purge your system and the time needed to carry out the process.

This is a very difficult and nearly impossible calculation. It all depends on a variety of factors - most of which are empirical or related to the actual physical composition of your process. The basic method(s) I know of relies on number of process volume displacements - and this assumes you approach something resembling pure Nitrogen plug flow - i.e., the sweeping Nitrogen flows like a virtual gaseous piston. This, of course, isn't true in real life; consequently you have to apply several times the amount of theoretical Nitrogen - and at a "correct" continuous gas flow. What that "correct" flow really is depends on your process configuration and its idiosyncracies. That's why you have to resort to empirical experience in applying the Nitrogen velocities (or expressed as flow rates) and their duration. The physical method and scheme of injecting the Nitrogen also has a bearing on the final solution, so you really have to develop your own empirical factors.

Basically, common sense plays a very important role in how and where the Nitrogen is applied. For example, high velocities are important in using a sweeping technique that is frequently very effective. However, if you have a rather large and complex volume - like a large, packed bed reactor - you can't justify high velocities and assume that you won't get channeling. Therefore, many times you may have to resort to breaking up your process into different sectors. Piping - especially small diameter piping - can be swept very effectively with a minimum of inert gas. Large vessels may have to be treated with more than 3 or 4 pressurizations of Nitrogen with subsequent venting in order to dilute the contents down to a tolerable level. Of course, all these operations should be sampled and "tracked" as to the levels of Oxygen (or other undesirable contaminant) within the purged system.

I recommend you break your process down into sectors and attack each sector as to its size and configuration, in order to maximize the inerting and sweeping effect and also economizing on the Nitrogen consumption. This is where you sit down with other process-savy colleagues and exchange practical and common sense ideas on forming a good application plan of attack. This kind of exercise is not for the "fast-learners" or the green, young graduates. I would only employ mature and experienced process engineers who are not only knowledgeable in process design, but in the specific mechanical design and fabrication of each of the major equipments involved. Depending on the hazards of your process and what the inerting represents, this could be a very important and potentially hazardous operation. Do not forget to take into consideration the constant danger of a Nitrogen environment. It is a notorious and stealth killer!

Good luck to you and your endeavors, and always stay safe.

[CheGuy]

Jedi, Thank you for your input, I appreciate it! the system is a large system, and indeed i had to break it down equipment by equipment. as a rough estimate of amount of N2 required to purge the sytem was to calculate the volume of each equipment and the volume of the pipes, that total volume would be amount of N2 required for the whole system. based on the total volume and the flow rate, the time required to purge the system can be calculated

[Dharmesh]

Very good expaination given by Art.
I would like to write from the same that instead of continuous purging pressurisation / depressurisation is more effective and reduce nitrogen consumption also. Later on you can keep just positive pressure of N2 to avoid any Oxygen content inside it.

There is no need to write more as already Mr. Art has given detail description.

[JoeWong]

cheguy,

Some points for your consideration...

Methodology
I guess your purging activity is just the normal purging activity where to get the oxygen content down to xx% where normal N2 purge is acceptable. If you need to reduce to a very low oxygen level e.g. light water reactor, you may need to consider other method. Please review your method.

N2 quality
You may need to assess you N2 quality. Sometime plant N2 may contain 3% - 5% Oxygen, if your system need to go below 3%, obviously no way...You may need to get Liquified Nitrogen which may have >99.5% purity.

Criticallity of purging section
If you are purging reactor section, please always consider high purity N2.


Continuous versus Pressurise/depressurise
I have conducted these kind of estimation many years ago...just purely my experience, large volume system better to apply pressurise/depresurise method...You have mentioned that source of nitrogen is from a main supply header. Supply pressure probably 8-10 barg. If you have large volume, you may take very long time for purging. You may consider the Liquified Nitrogen


Minimise N2 usage
Not sure if your system can take water after purging...if so you may consider to fill the system with water and follow by N2 purging (replacing the water content). I have applied this method before...significantly reduce N2 consumption

Safety
As Mr. Montemayor advised, "the constant danger of a Nitrogen environment. It is a notorious and stealth killer!"...If possible, try to limit the volume of purging section filled with N2. The discharge section should be directed to minimum manning area...

Noise level
Normally pressurise / depressurise method will create high pitch and high SPL noise. You may need to comply to some authority requirements...

Last but not least...I will try to find my old collection on this matter but chances is slim...

Hope above you

JoeWong

[CheGuy]

JoeWong,

Thank you for the reply, our plan is to purge the system in batches i.e. Pressurize/ depressurize.

the equipments have psv's to prevent over pressuring. any suggestions on design condiserations for the line size from header to the inlet of the eqipment? estimated pipe length is about 400 ft from N2 supply header to equipment inlet

[JoeWong]

JoeWong,

Thank you for the reply, our plan is to purge the system in batches i.e. Pressurize/ depressurize.

the equipments have psv's to prevent over pressuring. any suggestions on design condiserations for the line size from header to the inlet of the eqipment? estimated pipe length is about 400 ft from N2 supply header to equipment inlet

It really subject to section volume and time for pressurise/depressurise cycle...I would guess a 2" line should be sufficient...

Generally 3-5 time cycle would be sufficient...



JoeWong

[BABU]

check this paper 'Properly purge and inert storage vessels' in Chemical Engineering Progress, Feb 2001, Vol 97 / No 2 @
http://www.aiche.org...1-02/index.aspx

I went through this paper some years back. It gives simple equations to estimate the purge gas requirements. Note that subscription is required to access the magazine.

[BABU]

Properly Purge and Inert Storage Vessels

Inert gases are used to prevent fires and explosions in the vapor spaces of equipment. Correct selection of oxygen concentration and inert gas flow rate are critical to ensuring safety.

George R. Kinsley, Jr. -

Chemical Engineering Progress
February 2001
Vol. 97, No. 2

http://www.aiche.org...1-02/index.aspx

The above paper is also a good reference for purging of tanks/vessels. It gives simple equations to calculate the quantity of inert gas required for purging.