4 replies to this topic

[robjul]

Happy New Year to my colleagues!

I was given an assignment to calculate the pressure drop on a 10"ND pipeline where 600MT/hr of refrigerated Propane (-42deg. C) would be discharge from an LPG cargo vessel (via 2,000 KM of submarine pipeline) at the Jetty then, goes 45deg upwards diagonally then horizontally onshore for 500m long up to several "bullet type" storage tanks. I assumed the storage tank temperature would finally be in equilibrium with an ambient temperature of 25deg. C. Discharge static elevation is 25m.

I don't know what condition to use (to based on -42deg. C or 25deg. C) for my pressure drop calculation since I am aware that, with the lenght of the pipeline, Propane will pick-up heat and therefore, cannot assume an isothermal condition. I am thinking of a weighted averaging calculation where an isothermal condition can be assumed for every kilometer or every 100m of pipe lenght then, add the individual pressure drops to get the total. This is of course a tedious calculation but is this a right approach?

Please share if you have any sample calculation or spreadsheet. Actually, my boss prefers a manual calculation because he was used to it during his 1970's schooling and then, I have to countercheck it with an online spreadsheet. I actually had done some manual calculation using Crane's Technical Paper 410 but my counterchecked with an online spreadsheet yielded with an answer 10 times greater than the manual calculation! Hence, obviously as per my engineering "feel", the manual calculation is wrong and can't find the error even with further reviews and checks of the input data and calculations.

My online spreadsheet calculation yielded an acceptable pressure drop but failed on velocity. The criteria I used for a 10"ND line and above is: Allowable velocity = 3.0 - 4.6m/s whilst Allowable max. delta P = 0.45 bar/100m. Are these criteria correct which was just given to me by a colleague.

I had done so many pressure drop calculations in the past but it looks like calculations involving LPG mix, propane & butane is a bit complicated as its density varies along the lenght of a long pipeline.

Indeed, many thanks for the help/assistance.

[ankur2061]

robjul,

It would be better that you model the hydraulics of the propane pipeline using a commercial pipeline simulator such as "PIPESIM" or "PIPEPHASE" to arrive at exact pressure drops and velocities. I am assuming that your pipeline is running above ground on sleepers from the jetty up to the point where the line lifts to 25 m to be routed to the bullets. Heat loss is definitely an issue over the length of the pipeline if it is not insulated. The change in temperature profile over the pipeline length is also causing the density profile to change.

The commercial simulators also take into account the change in temperature profile over the entire length of the pipe thereby providing accurate pressure drops. Velocitie should be kept such that build up of static charges are prevented since propane electrical conductivities can be low and it can cause a flammable atmosphere.

As a summary, your pressure drop calculations really require the use of a commercial pipeline simulator to get accurate and reliable results.

Regards,
Ankur.

[kkala]

1. As I understand the case, propane density in cargo (-42 oC) is roughly 580 kg/m3, at end of pipe (25 oC, to your assumption) rougly 490 kg/m3. Average 535 kg/m3. Density variation contributes to a frictional ΔP uncertainty of 45/535=8.4 % (maximum), which could be acceptable. Of course mentioned "weighted average" calculation would give a more precise ΔP; heat transfer has to be taken into consideration to estimate the (range of) temperatures at end of pipe (now assumed 25 oC).
2. Likelihood of two events can be investigated through said heat transfer consideration, and measures be taken:
(α) Temperature of arriving propane (to bullets, or even to above ground pipe) be lower than the temperature limit for the specific steel (risk of rupture).
(β) Temperature of arriving propane to bullets be higher than 25 oC, increasing bullet pressure (cannot theoretically be excluded in hot summer days).
3. Ten times difference in results in the two ways of calculating ΔP may indicate unit conversion error in manual calculation. Check by another colleague could be fruitful, since such errors can escape our own reviews.
4. The boss may intend to check the result through manual calculation. The latter offers more insight, the software more precise (and faster) result on the condition that input data is correctly inserted. And two ways of calculation can verify each other.
5. 600000 kg/h would result in a velocity of 600000/490/3600/508.7E-4=6.7 m/s, which is too high. Comparing the presented criteria versus general line sizing criteria of http://www.cheresour...id-line-sizing/ gives following:
(α) v=3.0-4.6 m/s, versus max 3.6 m/s for 8" pipe and above (2 m/s also reported as "average liquid process")
(β) frictional ΔP=0.45 bar /100 m = 2 psi /100 ft, versus max 1 psi/100 ft for long pipes and 3 psi / 100 ft for short pipes (up to 250 m in total equiv length).
I think presented criteria could be acceptable for a line under the sea (length must be 2000 m, not km), despite their higher max values compared to general criteria. Advice from others would be welcomed on this matter.
I also remember high velocities (roughly 4 m/s, 1990) in the preliminary design of a line (dia 12"?) transfering liquid propylene from a local refinery sphere to harbor (~ 5 km).
But at any case propane line should be of diameter higher than 10" to transfer 600 MT/h.

Edited by kkala, 08 January 2012 - 10:04 AM.

[ankur2061]

robjul,

Check out the links provided below for propane pipe sizing and LPG line pressure drop. This could provid some guidelines for your sizing exercise:

http://www.engineeri...zing-d_830.html

http://www.engineeri...rop-d_1204.html

Hope this helps.

Regards,
Ankur.

[robjul]

Hi Ankur / Kkala,

My sincere thanks to your enlightening comments. I am aware of using Pipesim, actually; but apparently, my company doesn't have / can't afford expensive softwares. Hence, the manual calculation option. My deep apologies for a very long time to reply due to unbelievable / frequent changes on our client's decision, whose main concern is the erratic market demand? Good grief! I can't imagine the so many tons of engineering time and money lost!

Best regards,
robjul