Should I insulate the top head of my vessel?

[curiousjon]

I am looking for a few people to double check my logic on a design.

Background
I have a few new very large API vessels getting put into a cold climate region and to avoid long term corrosion under insuation we've elected to leave the top head un-insulated and include a lip overhang to prevent rain from getting underneath the insulation on the side walls. I've seen very large/expensive API tanks disapear b/c the insulation looks good on the outside, but once liquid gets in it wreaks havoc.

The tank runs at 20-40C full of a water solution and is nitrogen inerted.

My Logic/Concerns
1. In the winter the tank will reflux as it condenses water vapor in the nitrogen on the head space and drops below. This is accetpable as the tank is innerted and it won't hurt the product or cause corrosion.
2. Condensation around the radar level trasnmitters may be an issue. We will try and select the best radar knowing there will be condensation, however if they find there are an issue we will trace and insulate the nozzles, etc. We may just do that to begin with.
3. We have to insulate most of the tank as the product will freeze/difficult to move around 0C. I'm making a key assumptions that it's acceptable to NOT insulate the top head b/c the heat loss through the top head will be significanlty lower per square foot due to vapor being on the inside (instead of liquid) and the temeprature gradiant in the vapor space of the tank keeping the delta T fairly low. I've tried to varify this with calculations, but it's difficult to know what the true vapor temperature will be in the headspace. In generally I'm hoping the heat transfer in the headspace (per square ft) will be 5-10x less than you'd get with liquid on the inside due to lower U value and lower dT. If so, my pump is plently to keep that heat loss at bay.

Seem reasonable, overly optomistic? Other suggestions?

 

[LittleInch]

The thing I would be worried about is the differential temperature and consequential stresses in the vessel, especially, as you say, one part will be liquid and kept warm (40C and then a fairly sudden change to -10? with not much heat coming from the gas. I don't think there are easy ways to calculate this.

Internal cold gas circulation might prove interesting and internally how do you know that your condensation on the top of the vessel won't actually become a growing internal ice crystal? How low is the outside temp?

For me it would seem to be much better to find a better form of insulation - there are many solid ones around - PU foam or Foam Glas spring to mind which can be bonded to the vessel on top of your painting system then covered further with Aluminium backed foil or other outer coating. A better anti corrosion coating would go a long way as well to prevent this.

Also "cold climate" places can get really quite hot in summer - another set of changeable conditions where black body steel temperatures could hit 50 - 60 C?

How big / thick / pressure rating is this "API tank". Not sure exactly what you mean by an "API" tank by the way, but assume it is some sort of pressure vessel.


My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[MortenA]

Shell prefers only to insulate (and trace) the wet par of the vessel (i think, been a while since i worked with DEPS). I dont think internal "circulation" of water will be a big issue - and corrosion shouldnt be an issue if there is no O2 or CO2. The heat transfer from the gas filled part of the vessel will be low so unless the residence time is very long i dont think you would see a huge flux in temperature.

Maybe if you choose a guided wave radar LG then condensation woth be a risk with your LG either?