HEATING MEDIUM EXPANSION TANK 2

[Buchi]

Dear members,
I am trying to design a Heating Medium System for generating Hot Oil to be used by Heat Exchangers in a Flow station configuration for an FPSO.

i looked up the PFD of a past project and i dicovered that i require an expansion tank upstream of the hot oil circulation pump. What is the "EXPANSION TANK" surpose to achieve and how is it sized. Buchi

 

[MortenA]

Most liquids expand when heated. Since the system is required to be full of liquid (for good pumping) then it must have an expansion tank in order not to explode when the liquid is heated (just as for auto mobiles). As for sizing: In principle its the total volume timed with the exoansion coefficient of the heating medium and the design temperature span. Im no expert on this subjet but i think a volume (pressurized) obove max liquid level is required for keeping a constant pressure in the system.

Best regards

Morten

 

[MJCronin]

Buchi,

I suggest that you purchase the book "Design of Fluid Thermal Systems" by Janna. Available on Amazon.com for about $80.00

Also, contact some of the vendors of "hot oil" products ( try a search on "Paratherm" and "Therminol&quot I believe that they have guidlines available for sizing expansion tanks based on usage of thier products.

Expansion tanks, as I recall, must be:

- vertical axis, ASME VII vessels,MAWP of 30-50 psig

- Have a PRV, sight glass, and manual vent on top

- Are sized at four times the expected fluid "swell"

- Are located on a branch to the suction side of pumps

- Are at the highest point of the system

- typically have a 25-50 psig nitrogen "pad"


Good Luck !!

MJC

 

[TD2K]

Expansion vessels, as pointed out, contain the fluid 'swell' as it warms up from (typically) ambient conditions to operating temperatures. I'm not sure about sizing it 4x the amount of swell though that wouldn't cause problems other than cost. Depends how close you can calculate the volume of equipment and piping.

They do not have to be vertical vessels, I've seen several as horizontal.

Typically they do have an N2 pad on them to prevent O2 ingress as O2 will often react with your heat transfer fluid and cause degradation. 25 to 50 psig is on the high side IMO, typically the fluid entering the pump's suction is well below boiling and you therefore have a fair bit of NPSHA. If you need more NPSHA then an N2 pad can provide this.

Typically, these are on a standleg on the suction side of the pump (as pointed out) thus keeping a constant suction pressure on the pump (the vertical head and pad pressure). The circulating hot oil typically does not flow through the vessel so you don't need to insulate it for energy conservation or personnel protection (your company's practices may be different). The piping however should have valving to allow you to force the fluid to flow through the vessel and back down another standleg to the pump suction as this is used during the initial startup to slowly warm up the system and boil-out any water in the piping. During the boil-out of our hot oil system, we would raise the temperature 25F or so, watch the amount of steam coming out of the expansion tank vent and only raise the temperature again once the plume had decreased significantly. Once boil-out is finished, the valving is switched to normal lineup so that the liquid does not flow into the expansion tank and it remains at ambient temperature (which also reduces the chance of degradation if you get o2 into the vapor space).

Lots of heat transfer companies have technical information on their systems and design. Check out Dow's, Monsanto's, etc for information.

 

[MJCronin]

TD2K

Respectfully, I disagree with some of your comments...

Paratherm, a vendor of hot oil systems, suggests the 4X rule for expansion tank sizing..... I could not find suggestions or "rules of thumb" on this matter anywhere else

Specifying an ASME pressure vessel is necessary in most states for this application ( for a system operating pressure over 15 psig ASME certification is the law) For a vessel that will be (relatively) small, specifying an ASME MAWP of 30-50 psig ( or higher) really does not increase the cost of the system.... you will end up with 0.25" wall..... a standard tank

In this service, a vertical vessel, in my opinon, is preferable..... more sensitive to volume changes and easier to observe or detect level.

Your comments on "boil-out" are well worth noting...

For more info on systems design and "boil-out"

see...

http://www.paratherm.com/newsletter.htm

I would like others to list websites with design info on these topics...


Good Luck


MJC

 

[TD2K]

I suspect the 4x is likely based providing a safety margin, not that I disagree with it. Let's say I go through and estimate the entire system volume and calculate the swell. Well, if I size the expansion vessel to catch the swell from 25% full to 75% full I have to double the swell volume. Now, how much do I believe my calcs? Panatherm may just have added in another 2x margin (I'll have to visit their site and ask them, it's an interesting point). Unless these tanks are huge, and they typically aren't, the additional cost isn't much.

I don't disagree with your comment about the expansion vessel being a pressure vessel or the minor cost to accomodate 50 psig operating pressure. I'm just not sure I see the need for this high of pressure as a process engineer. Typically, the N2 pad has been described in my experience to keep O2 out, you don't need 50 psig for that. Hot oil circulation pumps are also fairly low head so I wouldn't expect large enough NPSHA requirements for that much N2 pad pressure.

 

Gentlemen,
I propose Buchi should approach a vendor. They shall be happy to guide him. I worked on heating oil system while designing an LPG plant and to my rememberance it was not a big tank.
The only thing one has to keep in mind that the tank should have volume equivalent to entire lines volume. once you feed the empty line there will be enough space in the tank to accomodate the expansion.
The basic point here is that your pump should not be air locked or cavitated otherwise the coke may form in the furnace tubes.
Therefore just design for the volume of entire system, lines and tubes only.
Have a fun.
Asghar Khan

 

[Buchi]

Hi guys,

i appreciate your contributions. they have actually gone a long way in throwing more light on the expansion tank design.

Buchi Buchi