Cooldown of Liquid Ammonia Pipeline Prior to Ship Loading 2

[ankur61]

Dear All,

I have a liquid ammonia pipeline from the ammonia storage tank used for ship loading with a 15 day cycle for each ship load. There are two aspects I am looking for the design of cooldown system for the nearly 10 km of pipeline.

1. The cooldown is to be completed just before the ship hooks-up for loading with a ship loading rate of 1200 T/h. Any vapors due to flash of the cold liquid ammonia (-33 deg C) during cooldown operation shall be recycled to the BOG compression and refrigeration system at the storage end and not at the ship end. Based on this do I have to provide a separate cooldown pump or can I have a small stream diverted from the main loading pump with flow control injecting a controlled quantity of ammonia in the line. At the loading end if I provide a vapor /two-phase flow line returning back to the ammonia storage tank would it work, or pressure losses in this much length would prevent the vapor / two-phase fluid to return to the tank. What would be the practical and economical scheme for the aforementioned cooldown scheme?

2. While slow cooldown to prevent thermal shocks to the pipeline is known to me, the time required to cooldown this pipeline with a given temperature decrease (degrees C per hour) is something I would like to know. The heat transfer calculations based on metal mass of the pipe cooling from ambient to final cooldown has already been performed by me. In this I have considered that the cooldown liquid quantity will be based on both the sensible heat transfer to the metal by the cold ammonia as well as the enthalpy of vaporization which will flash the liquid ammonia. In addition heat ingress from the ambient has also been considered while performing cooldown operations. The mass flow rate has then been arrived on a one-hour cooling basis which basically may not be correct. Any suggestions for such a calculation considering steady-state heat transfer. This is not really steady-state in the truest sense but I don't have the required knowledge or tools for a transient or dynamic heat transfer analysis for the pipeline cooldown calculations.

 

[georgeverghese]

What is your BOG recompression throughput in t/h, with throughput corrected for the higher feed temp during the initial stage of cooldown ? That should give you some idea of what cooldown flow for this line should be. Any more than this and it would have to go to vent or flare. Can your mainline 1200t/h pump turndown to this flow ?

 

[LittleInch]

What size pipeline?

Is this insulated?

How is the ship dealing with BOG during loading

10km is a long way.
You might well need a chiller at the loading point

One hour cool down is not feasible, neither is trying to do this without transient analysis.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ankur61]

Hi George,

The BOG recompression capacity is yet to be fixed. The calculations for tank vapor, rundown line and rundown pump heat ingress, loading pump recirculation line heat ingress and BOG generation by pump is taken into consideration for the tank area BOG calculations.

Hi Little Inch,

1. Pipeline size is tentatively 18" transporting 1200 TPH of liquid ammonia at around -32 to -33 degrees C

2. A Polyisocyanurate or Polyurethane foam pre-formed pipe insulation shall be employed. Thickness shall be approximately 5 inches.

3. During ship-loading all the BOG generated shall be taken care of by the ship BOG system. However, any BOG generation during cooldown shall be accounted for by the BOG compression system at the tank storage area.

I did my preliminary steady state calculations for the cooldown liquid flow rate as 26,000 kg/h based on a maximum ambient temperature of 45 deg C (in steady state I have assumed that the ambient and the pipeline metal temp. will be in equilibrium when the line is empty).

 

[LittleInch]

Ankur,

I've been doing something very similar for the last few years. Mine is LPG import over about 6km and it's proving to be not easy.

10km of Ammonia going to a vessel in India(?) is at least three times worse. You can't vent or flare Ammonia AFAIK, where you can for hydrocarbons, the arrival temperature after 10 km at that sort of flow rate is going to be in the region of 3 to 5 C higher than when it set off. That will deliver a HUGE amount of BOG onto the vessel which I very much doubt the vessel can cope with. It's easy for you to say the ship will cope with that, but the reality might be very different. You need to look very carefully at the max arrival temp / BOG capacity of the tankers as I doubt they will be able to cope with that sort of quantity. You could easily need a chiller package of 3-5MW to chill the liquid down again before it goes onto the tanker. I've ended up with one to avoid having to build a HUGE BOG capacity for something which only runs for 8% of the time. Ships generally don't have that capacity, especially when in port.

What you also tend to find is that the vast majority of these systems are not 10km long, more like 1km long and the plant looks after the BOG from the tanker during loading. Going 10km with a cryogenic product stored at atmospheric conditions is not normal and hence everything which is standard at other terminals isn't at yours.

The usual practice is to keep the pipeline cold by recirculating, especially at a 2 week interval between ships. Yu normally need about 5% of the loading rate as a recirc rate but using a separate much smaller pump.

Cool down will be more like a day, not an hour, mainly to reduce peak BOG rates.

But you can't do this without transient analysis.

Oh and is the pipeline above ground or buried?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ankur61]

Hi Little Inch,

Many thanks for your response. And you have definitely raised some serious concerns on the large amount of BOG at the jetty end. Maybe this needs to be discussed with the end-user. If they have made arrangements for handling large BOG at the ship end I don't foresee any problem.

The pipeline doesn't run buried, it runs above-ground on sleepers from the storage to the jetty as of the current design basis.

I would like to know about the 5% recirculation rate. Is it some kind of rule of thumb? Can't we estimate the actual cooldown flow rate known the line size, insulation characteristics and the ambient conditions?

Ammonia has a Normal Boiling Point of -33 deg C. Assuming that there is a 3 degC temperature rise at the loading end (final temp: -30 degC) if my termination point at the jetty is around 1.5 to 2 bar (gage) there will practically no vaporization. The isothermal properties of ammonia at the -30 deg C isotherm are attached for ammonia extracted from NIST webbook.

Let me know your opinion.

 

[LittleInch]

ankur61,

An A/G pipeline in India will get warmer faster than a buried one, especially with only 125mm insulation on it. But you can calculate that.

The key point is what is the tanker design pressure? Cryogenic tankers are normally "atmospheric" but normally up to about 150 to 200 mbarg max. Hence why you get loads of BOG when you flash the liquid into the tanks on the tanker.

If it is 2 barg then you don't have a massive problem probably. Seems unlikely to me based on 35000m3 cryogenic tankers.

Yes 5% is a ROT, just to get you in the right place. A longer pipeline like yours might need more, but still not more than 10% of the normal export flow rate. Yes you do the analysis to see what return temperature you get at different flow rates and hence the amount of BOG you get when the warm return line goes back into the tank.

Like I said - you're doing something well out of the ordinary here so only when you tell people exactly what your process conditions are will they understand that. They probably think it's a "standard" 1 -2km pipe from a tank next to the jetty with a vapour return line / handling facilities.

Also make it clear that there is no vapour return line form the tanker.

Good luck

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.