2 replies to this topic

Hello All,

I'm looking for a way to estimate tank inbreathing/outbreathing rates during normal cooling/heating (day to night) of tanks. I've read API 2000 and I cannot find any guidance for estimating normal rates and it seems to conservative for my purpose.

Some history - The reason I'm looking for this information is because the design of our blanket system (N2) is comprised of two parts. One is the LP system which provides blanket N2 to our continuous users. The second part is the HP system which is basically a standby system that is used for blanketing during extreme consumption rates. My thought is that API2000 gives values that fall under the extreme case. However, I have not found any guidelines as to how to estimate the "normal" consumption rate. I thought about simply looking at the temperature changes throught the day and assuming the tank to be at the same temperature, but this does not consider the effects of solar radiation.

Your input is greatly appreaciated.

Regards,

Wally

[proinwv]

Wally, this is a good question. Most agree with you that the API2000 breathing rates are extreme. API apparently has done this because breathing vents are relatively inexpensive, but they were not considering blanketing and its higher costs.

I do know that some companies have a "rule of thumb" which revises the API numbers downward by a fixed percentage. This method makes me uncomfortable.

A more precise method would be to do a full analysis considering the rise in tank temperature taking into account the radiation, conduction and tank color. Then to look at the transfer to the liquid contents, and the vaporization rate plus the increase in vapor volume due to the temperature rise. Of course this is outbreathing, which is cheaper to control vs inbreathing.

For inbreathing, the cooling would be a similar calculation based upon radiation from the tank but these are all difficult calculations and the time for the changes to occur are important. I suspect the sun might rise before the tank stabilized.

But the most important issue is what will be the result of oversizing the hp blanket gas supply to protect the tank against damage or the contents from degradation due to ingress of the atmosphere. These alone, might justify using the API numbers.

Your question does bring up a good point in that we all need more information on this subject. Has anyone monitored a tank over a year and recorded the actual values? Even the API numbers are theoretical.

I don't believe that I have answered your question. The reason that I have written what I have is to hopefully encourage an exchange of information here on this subject.

[Art Montemayor]

Wally:

API 2000 is titled “Venting Atmospheric and Low-Pressure Storage Tanks”. It was intended as a safety guideline to protect personnel and tanks from over-pressurization as well as from partial vacuum conditions. It was not intended for optimizing Nitrogen consumption in tank blanketing applications – therefore, you find that it is conservative in its scope. Its conservatism is founded on well-based and sound engineering practices.

When you apply Nitrogen blanketing to a storage tank you will find that the consumption of Nitrogen is due to:

1. Nitrogen Leaks;
2. Frequency of stored fluid pump-ins and pump-outs;
3. Tank size;
4. The stored fluid liquid level maintained in the tank;
5. The quantity of fluid pumped out in each pump-out;
6. The pressure settings between the Nitrogen feed valve and the conservation vent;
7. The thermal changes within the tank due to outside temperature changes as well as stored fluid.

It has been my experience that the amount of Nitrogen lost (or “consumed”) during any reasonable time interval (say 2-weeks to a month) is influenced by the above 7 factors more-or-less in the order that I have listed them. In other words, I have found that leaks contribute more loss and thermal changes the less. And by leaks, I include faulty conservation vent pallet seating and adjustments, gaskets, etc.

Granted, I have had my blanketing experience in relatively mild climate changes and not in places like Canada. So one has to consider location as well. I have not had that opportunity. I anticipate other engineers on the Forum will contribute their experiences and findings and I am very interested in what they have found and can share with us.

I don’t know what you call “the ‘normal’ consumption rate”. That is dependent on the location and the application. In some instances you might have to expect higher than “normal” Nitrogen consumptions. However, the best (& probably the only practical) way to gauge or measure your expected Nitrogen consumption is to carefully monitor and measure the quantity and duration of all your pump-outs and pump-ins. This, in my opinion, is the best and most accurate empirical way to judge what you should be consuming. Granted, it’s a laborious and tedious method; but it’s the basis for a large amount of Nitrogen inventory in your tank if you are emptying an entire static tank into another, for example.

I believe that if you assume a liquid inventory of approximately 80% in a tank and then study the Nitrogen blanket expansion and contraction effect in the tank, you will find that a daily consumption of Nitrogen due to those effects alone won’t be much. However, a lot depends on where you maintain your liquid level during the expansion/contraction periods. Monitoring your vapor space temperature during those periods will give you an indication.

I hope this experience helps you somehow.