Demin water tank - nitrogen blanketing 7

[BigRedEnvEng]

I'm working on designing a nitrogen blanketing system for a 500,000 gallon demin water tank. The plant uses a large amount of demin water about 12 times a year for startup operations, but the rest of the time it's at steady state operations with a low flow. From my rough calculations, I'm thinking we're probably going to need about 80 scfm of nitrogen during the large pushes, and small amounts to maintain stead state for the rest of the time. Does anybody have any suggestions as to the pressurization required on the tank? Right now I'm working my calculations off of a 1 psig tank, but wasn't sure if that made sense. I would like to put a relief valve on the top of the tank and allow the nitrogen to release to the atmosphere when necessary as opposed to trying to include an expansion tank or bladder, but am open to suggestions. I also wanted to hear ideas of nitrogen generators versus cylinders if people have had experiences with either/both. Any thoughts would be appreciated.

 

[unclesyd]

I think you should take a look at this type valve or a conservation unit with a N2 pipe-in.

This valve can be sized to keep the N2 consumption to the very minimum. Then it becomes matter of economics versus obtainable purity between the two.

http://www.protectoseal.com/blanketvalves.html

 

[jamesbanda]

Big red...

why do you need an N2 blankett for the tank.. is it to maintain the purity.. since it is only water..

 

[BigRedEnvEng]

We're using the nitrogen blanket as a means of preventing the oxidation of the water after it's completed the cycle through the demineralizing system. We're looking for a very high purity content for the water.

 

[shahyar]

I think "oxidation of the water" is not right phrase to explain what you are looking for. The better thing is: "to prevent demin water from absorbing carbon dioxide from air and decreasing its impurity".

 

[sailoday28]

A mass energy balance of the N2 control volume in the tank is whats needed. Some simplifing assumptions
- is an adiabatic control volume (no heat transfer from the inside walls of the tank or from the N2 interface to the water.
- Since the N2 is at low pressure--a perfect gas
-A relation between the pressure of the N2 and the displacement of the water.
-The stagnaion conditions of the source of N2.

dE = ho*dm -pdV +Q Where E=me
m= mass and e specif internal energy
ho stagnation enthalpy
p pressure of N2 in the tank and
V volume of N2 in the tank
Q heat transfer
With no heat transfer
dH - V*dp = ho*dm

With ho = constant and knowledge of V as a function pressure
H = Hi + integral of V*dp + ho (m -mi)
where subscript i refers to initial conditions

H= mCpT for perfect gas, constant specific heats
For any amount of N2 injected H is therefore known

But
H= Cp(pV/R), so that pV is also known as a function of mass infected.
To obtain pV, we also knew, p as a function of volume-see differential equation.

I leave the rest of the exercise to others. Please check for minor errors-- and not for conversion of units

Regards

 

[Montemayor]

The answer is so simple it’s a shame to waste more time on it. Unclesyd hit it on the head. Just call up ProtectOseal. They will help you out and show you exactly how it’s done.

In the last 10 years I’ve probably averaged 5 tanks per year with blanketing applications. And with environmental laws getting stricter, the pace will be heavier in the future. The basic problem is that you must have a good identification of your tank’s mechanical strength/weakness – the MAWP and the MAWV (Maximum allowable working pressure and Maximum allowable working vacuum). You need these two values in order to be able to safely set your blanketing and relief points on your instrumentation. The wider the “dead band”, the better off and the smoother the operation. But you must have a dead band that allows you to work your instrumentation. Most tank owners don’t have the MAWP or MAWV – simply because they are too cheap and don’t want to pay for the rating. To install a blanketing system on a storage tank without this knowledge is next to madness. It’s an accident waiting to happen – and it often does. If you don’t know how to design the system or have little experience, I highly recommend you obtain the services of a professional engineer who has done this and has the know how.

You will be using a Nitrogen feed valve just like the one Unclesyd shows you, together with a conservation vent and possibly with an emergency vent. You should have all your credible hazard scenarios totally identified in detail in order to adequately size your conservation vent and operate the tank(s) in a safe manner. If you go to

http://www.cheresources.com/indexzz.shtml

and look around the Forums there you will find where I have furnished a lot of people with a Workbook that details out how the operation is designed and what equipment is used. Post a thread there and I’ll furnish you with a copy.

You’ll find you can’t justify a nitrogen generator just for this tank. You may have to go with a cryogenic storage tank on site and an atmospheric vaporizer. This equipment is all rented or leased from your nitrogen supplier – like Air Products, Praxair, etc.

There is no Thermodynamics involved in this application. It’s simply a displacement problem. Additionally there are no expansion tanks or bladders used. The nitrogen blanket in the tank when the same is allowed to go down in level will be vented to the atmosphere on the next subsequent fill of water. At one time I know some people tried to save nitrogen by using a giant balloon to inflate with the excess nitrogen – but the thing is so big (you can easily calculate the required size) that it needs protection and supports, eating up a lot of valuable storage space and area. The best thing to do is keep a true steady water level – i.e., keep a steady outflow matched with a steady inflow. In real life, I’ve never seen this accomplished and the result is that you have to vent the excess nitrogen you feed due to low levels in the tank.

One of the bad hazards is pulling a vacuum, and this relief case should be carefully analyzed.

I hope this helps you out.

 

[sailoday28]

Montemayer:
Your suggestion may be fine. However, I am supprised that
you state there are no thermodynamics involved.

Other than your suggestion, please tell me where the thermodynamics and model that I have presented are wrong?

Regards

 

[pleckner]

If I may add my two-cents.

"sailoday28": Although I haven't gone through your thermodynamic evaluation for correctness, I can tell you that from a practical point of view, Art is correct, a thermodyanmic analysis of this system is not required and everything is based on displacement of vessel contents. Saying this, I will also grant you that there is displacement due to thermal effects, which we call "breathing". All one needs to do it find a copy of API 2000 and there you will find the tables indicating the required breating rates for vessels for this application. These breating rate tables replace the need for your thermodynamic analysis. (Note that the tables are based on hydrocarbons and but are still used even for this system.) If you can't find API 2000, many of the vendors that sell blanketing systems, such as Protecoseal, usually publish the appropriate sections of API 2000 to help you calcualte the amount of breathing needed. Your analysis It may be a great academic exercise in seeing how the system works on a theoretical basis and may be of interest but again, it is just not needed.

So in summary, nice job but not required for the real world.

 

[sailoday28]

I don't believe anthat API standards are used in design of emergency ballast systems on submarines. Air displaces water from the submarine to give additional bouyancy.
For that case, academics are used with a backup of practical testing.

Regards

 

[pleckner]

OK, can't say I know anything about ballast systems on submarines but the original post was about a 500,000 gallon Demin Water Tank for a plant site. I feel we should be trying to help the engineer with the best practical and cost effective solution to their particular problem and this usually implies using accepted industry practices.

 

[sailoday28]

pleckner
Demin water tanks are used in Nuclear Power plants
Do you think API standards would be acceptable by the Nuclear Reg. Commission as a design basis for deminwater tanks?
This ends my comments-

Regards

 

[JimCasey]

A couple of points: 1 psi design pressure is too high. Just a few inches of water positive is plenty.
Take a look at the thickness of the steel on top of the tank. Using the density of steel, calculate the weight of the roof of the tank per square inch. This is how much pressure it will take to support the roof of the tank. Blanket at a lower pressure than that magic number because reversing stresses will ultimately fatigue the top-to-side weld by oil-canning the top and cause at least a leak if not a failure.

Protectoseal is a good reference. So is Groth.

Thermodynamics: Look at the API requirements for inbreathing. There are 2 cases: (1) replacing the liquid as it is drained from the tank. (2) The rain-shower scenario where the tank is sitting in the hot sun on a summer afternoon and suddenly gets drenched by cold rain. The inflo required to offset the thermal contraction of the internal tank atmosphere is almost always greater than the pumpout value.

Rubber bladder: May be a good idea sometimes, but I visited a large nuclear plant where they stored DI water in a tank with a rubber bladder and the bladder crinkled and somehow blocked the breather. Pressure actually inflated the tank & lifted the bottom edge of the tank several inches away from the slab. Massive paperwork ensured, although the tank did not suffer significant damage...that time.

 

[unclesyd]

If you look at the literature of the several manufacturers of tank venting equipment you will find that the tank blanketing valve will accommodate the larger draw downs while a conservation vent with an N2 pipe-in will handle the normal everyday breathing.

We have 2 high purity water storage tanks(150,000 gals ea) that use the conservation vent with N2 approach as every mole of O2 in the water is a mole of cross linked polymer. These storage tanks are in the make and take mode. No problems in 50+ years of operation.

Numerous other tanks for O2 sensitive chemical components of the process have both a tank blanketing valve and very elaborate breather pots, steam jackets, N2 feed-in, on stream cleaning. These tanks also have a dead weight relief system.

Air products now has the whole package with cryogenic storage and an "Inerting Box"

 

[BigRedEnvEng]

First, thank you all for all your responses. This is my first time using these forums and this is really fabulous.

I spoke to Protectoseal and, like some of you said, they were quite helpful.

I obtained a copy of API 2000, which did show some good stuff for the vents and such. I came up with 73.07 scfm for inbreathing and 142.86 scfm for outbreathing; does this make sense?

Montemayor, I'm going to post the same thread on the CheResource site under the tank blanketing forum; I'm curious anyway to see how that forum works.

 

[JStephen]

JimCasey is correct about the pressure. Typically pressures somewhat higher than the weight of the roof plate are no big problem (although the tank may require anchorage due to it) but 1.0 PSI on a 500,000 gallon tank can be a headache to work out. Usually on an inquiry like this, owners decide they can live with less pressure once they find out what that pressure costs.

Also check out Tyco/ Varec and Shand & Jurs/LJ Technologies on the P/V valves.

 

[DeltaCascade]

WOAH! It appears that you want to nitrogen blanket to keep dissolved oxygen out so that the demin water won't become corrosive to carbon steel tank? (it is after all, upstream of your deaerator(s))

10 ppb dissolved oxygen makes demin water surprisingly corrosive. At room temperature, about 10 ppm (10,000 ppb) dissolved oxygen will already be in the demin water entering the tank. The iron will scavenge and consume the dissolved oxygen, by rusting and pitting! Talk with your Materials Selection Group (metallurgist?).

This is why demin water tanks are stainless steel, or lined or coated, or plastic (watch temperature, and size), but not just carbon steel.

Thanks for the postings on blanketing.
Hope this may be helpful.

Good luck

 

[MedicineEng]

BigRed:
I am not going to discuss the blanketing issue as you already had plenty of insight from other fellow engineers.
Even so, I would like to comment that after 8 years in Pharma industry and working with ultra-purified water for injectable grade, I don't recall that we ever had to blanket the storage tank with afraid of contamination.

I would just like to leave an idea and see if it is feasible:
You say that mainly this 500 000gallons tank is because of your shutdowns and during normal operation you DI water need is quite reduced. The best way to avoid contamination of the water by air is to decrease its storage time.
Did you consider to use a smaller thank for your daily use and then fill up the big tank before the shutdown?
Is this feasible?

 

[BigRedEnvEng]

I guess as long as the water just keeps moving quickly from the demin system to its final application, we wouldn't have to worry about maintaining its purity - it's a good idea and I'm definitely going to keep that in my bag of tricks for future applications. I'd have to take a look at the flow rates during steady state operations and see if that would be feasible. Thanks for the idea!

 

[IFRs]

Could you build a small tank inside the large tank and when the small tank was full it would dump into the large tank? Suction nozzles on each tank and a few valves could do it all. The inside tank would not reduce your large tank capacity by much...