8 replies to this topic

[Fr3dd]

Dear all,

 

Once again, I have a question for you.

 

Background: consider an upstream heavy crude oil (arounf 16API) facilities. It comprises crude separation, desalting, dehydration, etc. Currently we're evaluating one shutdown scenario where the production needs to be shutted down; this means that we will have standstill heavy crude oil (not flowing because of the shutdown) in our piping.

 

The problem: The drainage facilities design temperature is 60deg. C (the crude is expected to be as high as 130deg C in some sections), so we're evaluating the possibility to let it cool before draining taking care, of course, of the viscosity increase that this may represent. We would like to know how much time do we need to wait for it to cool AND how much time do we have before reaching the pour point.

 

The question: My question is regarding how to approach this calculation, steady-state modeling (as in Hysys or pipephase) doesn't seem to be applicable here since the phenomenon is transient. I've been digging into my college heat transfer memories and literature, but it seems that the fully rigorous approach may be too complicated. What approach do you suggest to calculate this? or Which classic heat transfer case should I look if I decide to go through a more rigorous approach?.

 

Thanks in advance.

 

Fr3dd

[PingPong]

Do you know what the pour point of this crude is?

 

Do you know the viscosity of this crude at two different temperatures? If not, which crude is it?

 

Is the system containing the 130 degC crude insulated? If so, how thick is insulation?

[Fr3dd]

^{Hello PingPong}

 

^{Thanks for your interest. First I would like to do some clarification, this project is in a very preliminary phase, I'm traying to do this calculationsto estimate if it's a logical approach to let the crude cool down while the upset situation is being solved or we need to think in another strategy. This means that I basically have crude data but not a finished model (simulation), or plot of the plant.}

 

^{I still haven't found the pour point, I do have a comprehensive viscosity curve and we may expect around 200 and 300 cP at 60deg C. The system will not be isolated since it will be handling crude with sour gas in a marine environment (corrosion problems under insulation may be frequent. However, this calculation will also help to define how much necessary is to have insulation in certain lines.}

 

^{Do you have in mind which approach should I take to make this transient heat loss calculation in piping?.}

 

^Thanks, 

[ankur2061]

Fr3dd,

 

I would not worry much about the integrity of the drain system components (vessels + piping + instrumentation) at the elevated temperature of 130°C if the MOC of the system is 'Carbon Steel'. This is something which you need to tell us. 

 

Assuming, that your entire drain system is having a rating of 150# with carbon steel as the MOC of all the wetted components, the difference in the pressure rating for a design temperature of 60°C and an elevated temperature of 130°C is as follows:

 

Pressure rating @60°C:  19 barg

Pressure rating @130°C:  16.5 barg

 

Apparently the pressure rating does not go down drastically for a temperature increase to 130°C from 60°C. 

 

So, unless you are having plastic or polymer components in your drain system, I don't see any reason of not being able to drain the hot material. The only concern would be the surface temperature of the drain system components which may require some insulation for preventing injuries due to burns, i.e insulation for personnel protection where personnel are likely to come in skin contact with drain system components.

 

Hope this helps.

 

Regards,

Ankur.

[PingPong]

However, this calculation will also help to define how much necessary is to have insulation in certain lines.

I am not sure that I understand this. Or do you mean that otherwise some pipes may cool down too fast compared to the vessels?

 

In general: when an object cools down there are two theoretical situations possible:

 

(1) The object contains a low viscosity fluid and thus maintains a homogeneous temperature; and the heat transfer to the surroundings is only determined by the heat transfer of the outer surface to surroundings by (natural) convection and radiation. This is the case with a tank of water.

In this case Newton's Law of Cooling is applicable.

Note that when not insulated, the emission coefficient (emissivity) of the outer surface has a very big impact on the heat loss rate.

 

Or:

 

(2) The object is a solid and thus will have a temperature gradient from center to outer wall due to conduction. The heat transfer to the surroundings is determined by the heat transfer of the outer surface to surroundings by (natural) convection and radiation, and also by the thermal conductivity of the object.

In this case you need to work with the Biot number.

 

Assuming your system is not insulated, you will have something in between (1) and (2) due to the high viscosity of the crude.

At 130 ^oC it will be mainly situation (1), but as temperature drops and viscosity increases it becomes less (1) and more (2).

 

 

EDIT: I now see that Ankur replied with respect to the maximum temperature that the drain system can handle. Of course the steel drain piping can easily withstand 130 ^oC, but there may be a stress problem due to expansion at higer temperature than design. Moreover the drain system can contain water and that could cause a steam explosion if oil hotter than 100 ^oC is dumped into it.

Edited by PingPong, 05 March 2014 - 06:38 AM.

[Fr3dd]

Ankur, definitely I should check which are the reasons for the limitations of draining above 60 C. I cannot assure you the material of the drainage system piping now, but definitely there will not be polymer components. And of course, I will take into account the steam explosion / thermal expansion scenarios pointed out by PingPong.

 

At this point I would like to clarify that, even when I have experience as a process engineer, It's been a long time since I saw a transient heat transfer calculation and I may be a bit confused in how to approach to it. So I apologize if my questions seem dumb.

 

PingPong, I understand the problem. I guess an approach as in (1) could be more conservative. In this case I would be considering properties of the fluid (specific heat) and the convective coefficient of air when applying Newton's Law of cooling (specifically for the exponent of the equation); is that right?. How can I consider radiation effects on this (how to express them in a non-steady-state equation?)?

 

Thanks for your insights about this.

 

Fr3dd,

[PingPong]

How can I consider radiation effects on this (how to express them in a non-steady-state equation?)?

It is not possible to include all effects in a single equation, because transfer of heat from oil to metal wall, and heat transfer from wall to ambient air (and surroundings) by (natural) convection and radiation all change as the oil and metal wall temperatures change over time. Moreover there can also be heat input from solar radiation.

 

To calculate it properly you need to do it in small temperature steps, starting at 130 ^oC, and calculate all relevant factors to determine the heat transfer from oil to ambient air. Based on the weight and specific heat of the oil, water and metal you can then calculate the time it takes for the pipe or vesel to cool down each temperature step. This kind of calculations is best done in a table format in a spreadsheet.

 

If there is a desalter vessel in the system it will take a lot of time, maybe a week before its oil and water content is cooled down to 60 ^oC if there is no wind worth mentioning. Actual cool down time is affected by the weather: ambient temperature, wind, rain, solar radiation.

[Fr3dd]

Gentlemen,

 

I've got some updates regarding this post.

 

First, rergarding the reason why drainage at more than 60 C is not considered. This is an offshore installation, drainage is conducted to a cargo tank, this tank has an internal coating (probably as a corrosion counter-measure) which allows a maximum temperature of the fluid of 60 C. Drainage above this temperature would result in damage to the coating and possible consequences in the future (so it's avoided).

 

Second, regarding the calculation by steps, I'm still struggling with the "calculation of the time it takes for the pipe or vessel to cool down". I did some research and I manage to find several approaches of estimating this in the case of conduction with convection (one-dimensional heat transfer) and manage to find (graphically or by an approximated solution) the time required for the cooling. However, I still don't know how to do the same with radiation.

 

I understand the purpose of doing the calculation step by step, however I feel like I'm missing something. PingPong, can you give me some more lights on this, regarding the approach I should use to find out the time it takes to lose the heat by radiation? I just need you to point out which direction should I take (maybe a general equation, some section of a textbook, the title of a paper...).

 

I really appreciate your help so far, and thak you in advance for your upcoming assistance.

 

Regards,

 

Fr3dd

[PingPong]

However, I still don't know how to do the same with radiation.

Radiation heat loss depends on the outer wall temperature of the vessels and pipes, and on the emissivity (emission coefficient) of the paint that was used.

Surely you are familiar with the Stefan-Boltzman equation: http://hyperphysics....rmo/stefan.html

Als T drops over time, so does the heat loss due to radiation.

 

What is the biggest vessel in the system, and what is its L and D ?

Edited by PingPong, 13 March 2014 - 07:13 AM.