We are having a Heptane tank of 200m³ capacity. The Nitrogen blanketing valves are located at ground level and the pipe diameter is 11/2". The ambient temperature varies from 15°C to 40°C during winter and summer.

The boiling point of Heptane is 98°C is this set up OK and can be accepted as good engineering practise.

AFD

AFD,


It is not accepted practice to place the nitrogen blanketing valve at ground level, or at any elevation other than above the liquid level.

The reason being, as most tanks are blanketed at pressures of several inches water column, it would take only a column of liquid equivalent to that to accumulate in the sensing line and prevent the valve from opening. This can lead to developing a vacuum in the tank and result in tank collapse. This is not just theoretical. I have seen it happen.

My recommendation is do not use that construction.

Regards,

The facility is working this way for the past decade and to initiate a change will be difficult. I read your expert comments in the previous posts and thought the volatality of the fluid may play a significant role in placing of the blanketing valves at grade.
AFD

AFD:

As a veteran contributor to these forums, it is good to hear from you once again. You often bring serious and high-valued issues to the forums and this happens to be just another of your indirect contributions. This is an issue that I consider as very, very important because of the serious hazard that it can cause – particularly in your application.

First, and foremost, be sure to heed and understand exactly what Paul Ostand has written in the above post. What Paul has stated is not only the engineering truth, it is an accurate description of what can go seriously wrong with a storage tank installation if the basic natural laws of physics and chemistry are not observed. Allow me to extend Paul’s explanation to further details in order to show you what the results and the impact of locating the Blanketing Valve at ground level can cause:

1. First go to: http://www.cheresources.com/invision/index...st=50#entry8246 and download the Excel file in post #58; in the Blanketing valve tab you will find details of this valve and its workings;
2. Next download the attached Excel workbook which is a compilation of data found in Paul Ostand’s website – something that I recommend every engineer reading this thread to do. With Paul’s permission, I have made this information a part of my archives and engineering database to refer to and to refer others to. Please note the physical and mechanical features of Paul’s design and how the competing designs comply with his and follow the same general design. This feature agrees with and complements what Paul is stressing in his thread.

Please note the effect that I’ve tried to illustrate in the Paul Ostand Workbook. Note the tab “Vapor Pressure Effect” and the system that I sketch out. Initially, the Storage tank contains a liquid (say Heptane) at ambient temperature of 20 oC. This corresponds to a vapor pressure of 35.4 mmHg for Heptane. If the blanketing valve is located at grade level, besides the tank, you will have created an inverted “U” tube system that is initially filled with saturated heptane vapors at 20 oC. Once the tank system is subjected to a lower temperature, the vapor space will condense the portion of heptane vapors that are in excess to the new, colder vapor pressure. The corresponding heptane that condenses drops by gravity to its lowest location, directly below where it finds itself. The portion of condensing vapors found over the main liquid level in the tank pose no problem. They drop back into the tank, where they came from originally. HOWEVER, those vapors that find themselves in the 1-1/2” pipe leading down to the grade-level blanketing valve will also condense but these will fall into the discharge port of the blanketing valve and start to fill the 1-1/2” vapor pressure sensing and valve discharge line.

As you can appreciate, the presence of a growing hydrostatic head of liquid over the blanketing valve makes it harder for the blanketing gas to enter the tank. More importantly, as Paul points out, the same hydrostatic head will impose an additional pressure to the blanketing valve and cause it to remain closed. Since the blanketing valve remains closed due to a false signal being generated, the tank can develop a sudden vacuum if liquid is pumped out and no make-up nitrogen enters through the blanketing valve. This is a very hazardous situation that must be avoided and is the reason, as Paul points out, it is not accepted practice to place the nitrogen blanketing valve at ground level.

I hope this helps out in reinforcing what Paul has stated.

Click to view attachment

Art,

As usual you explain these matters better than I, but we are in agreement.

To All:

I have been reading the posts from Art for some time and I consider him to be an extraordinary resource to the world engineering community. Thanks Art.

Thank you Art for your excellent explanation and I also recall an instance where the Hexane blanketing valves were relocated to the tank top. I could not understand the significance that time.

The blanketing nitrogen header pressure is at 3 bar and the tank height is 5 metres.
The blanketing valve is self actuated PCV and is set at 300mmWC(30mbar).

Now if the pipe is full of Heptane:

P =H* rho*g 5*770*9.81=0.3bar(300mbar). Or will the PCV sense this Hydrostatic head downstream the PCV and never open..

Due to the wide swing of temperatures from day to night the HC will vaporise and not retain its liquid phase.

The liquid draw off is only 2m³ per day.So this may be one of the reasons why the valve is still at grade.anyway I will speak to my bosses.

Finally, Art I respect your engineering knowledge,expertise and experience.As an operator the knowledge I have gained from you through this forum is immense.
Thank you for sharing your knowledge with us around the world.

Art can you shed some light on this post:
http://www.cheresources.com/invision/index...?showtopic=3200

AFD
A candle does not lose its light by lighting another candle

afd

I am not sure that I fully understand your last post, but let me try.

First, I have the impression that your blanketing valve is sensing pressure in it's outlet connection rather than by a separate sensing line connected to the tank vapor space.

This can be a problem because if the sensing is at the valve and there is a length of pipe between the valve and the tank, then the variable (friction loss) pressure drop will falsely influence the valve, causing it to close upon flowing and then reopen when the flow is reduced and the sensed pressure is closer to to the tank vapor space pressure. This will cause cyclic operation.

Second, regardless of whether a sensing line or just a discharge line is used between the valve and the tank, any accumulation of liquid having a head pressure in excess of the setpoint will prevent the valve from opening and likely result in a pressure reduction in the tank when liquid is withdrawn or when there is a temperature drop.

Dear afd,

If you can not change the position of the blanketing valve I suggest some things that you may select after critical evaluation of each:
1. you can take tapping for pressure sensing from the tank vapor space and not from the downstream pipe of the valve so that effect of condensed liquid head is zero. But still the sensing tubing will be full of liquid after condensation of heptane vapor.
2. provide heat tracing of the sensor tubing/complete vertical line downstream of the valve to vaporise any condensed liquid.
3. manually or automatically drain the vertical pipe/sensor tubing intermittantly to remove the condensate.

Regards.

I would agree with the comments by Dhirajkumar in a perfect world. However, it is too easy for this to fail, especially with relying on a manual drain. The drain interval is a variable that would be near impossible to specify.

I realize the difficulty in relocating a valve but because it hasn't failed in ten years, is meaningless when it does. It is a bad application and bandages are not the solution.

Sorry for the bluntness, but if you have ever seen a tank reduced to a height equal to twice the metal thickness, you would agree.

Good luck