7 replies to this topic

[Dtan76]

Dear All

I just post the thread last month ago regarding a FEED study but not reply others member... Is ok also because I don't think the buffer tank is cost effective design.

I complete the study and propose simple control philosophy.

Here some interesting topic regarding how we can estimate amount of vapor Ehylene generate inside the Liquified Ethylene Storage Tank during unloading Ethylene, loading Ethylene from Jetty or pumping of liquid Ethylene to pipeline.

This project's scope of work is EPC the Ethylene Storage Tank and it associate Ethylene vapor compression, refrigerant, chiler and condensing system.

I had built Hysis for this study also.

The liquified Ethylene product is unloading from ship/vessel through marine pipeline to transfer to the storage tank. During the unloading process, the storage tank will generate massive vaporized Ethylene which will be treat by compressed, chilled, condensing, depressurized and temp reduction and recycle back to storage tank.

The recycle Ethylene will be normal pressure and around -102 degC. Through recycle the evaporated Ethylene wil able to control the storage tank temp & pressure hence ensure tank operate at safe 7 stable manner.

The amount of Ethylene unloading from Jetty to tank is 300,000kg/hr, inlet temp variance from -102 ro -104 degC

The storage tank working volume is 20,000m3

The storage tank operating pressure/temperature is 107KPa / -104 degC

Vapor pressure for Ehylene at -102 and -104 deC is 1.119Bar / 1.01 Bar respectively

Density of Ethylene is 566kg/m3 at 163K/-104 degC

Please refer to attached PFD.

Regards
DTan



vapor

Attached Files

•  Ethylene_StorageTank_VaporChiled_Comp_CoolingSystem.xls   23.5KB   195 downloads

[joerd]

Nice sketch. So, what is your question (if any) ?

[Dtan76]

Hi all

I'm need to estimate the amount of massive Ethylene vapor generate during unloading/loading process, furthermore, I will use the vapor amount to size the compressed, chilled & refrigerant system.

Can anyone share their ideal?
Is there any rule of thumb to determine the amount of Ethylene "evaporation rate" during ethylene liquid loading/unloading stage (unsteady-state)? The Owner told me evaporation rate not more than 0.08% of total 20,000m3 per day (This figure is for normal operation, not loading/unloading)

Thanks and regards

Cheers

[joerd]

The amount of vapor from the tank is due to displacement (because the tank becomes fillled with liquid where you had vapor before) and evaporation (flashing).
The evaporation rate is calculated by the total heat input to the ethylene:
1) pump inefficiency
2) heat input from the outside into the transfer lines
3) heat input into the tank

If the transfer lines and tank are not at storage temperature, you'll have to cool them down, which will cause additional flashing. If you know the inlet temperature of the ethylene into the tank, you have already covered items 1) and 2) above, so you can just use a flash calc. in HYSYS at the tank pressure and the inlet temperature, and add the owner's figure for "normal" evaporation to the calculated rate.
The amount of vapor to be compressed should also include the flash vapor from the economizer/letdown flash vessel, but I assume you have that in your HYSYS model.
I am not aware of any good rule of thumb for this calculation.

[Dtan76]

Dear Joerd

I had build this model in HYSIS 3.1 but with error occured...

Additional info you can refer:
a) The Storage tank is ISOTHERMAL at -104 degC, singer containment, double layer, SS316L
The incoming Liquified Ethylene feed to storage tank have temp range from -102 to -104 degC, the Ethylene incoming pipeline inlet pressure assumed is 3 to 6 Bar(a)
c) Ethylene liquid density at -104 degC = 566 kg/m3

Amount of Ethylene unloading from ship/vessel feed to tank = 300,000 kg/hr
Inlet stream of Ethylene, unloading from Jetty will have temp as lower as -102 degC to -104 degC
Amount of Ethylene pumping to Truck = 25,000 Kg/hr
Amount of Ethylene pumping to downstream, pre-heat by steam through heat exchanger (kettle Reboiler type) = 25,000 kg/hr

Vapor Pressure of Ethylene
At T = -102 degC, Pvap = 1.119 Bar
at T = -104 degC, Pvap = 1.01 Bar

Enclosed here the Hysis 3.1 model, please have a look and let me know how to rectify it.

Thanks and regards

DTan

("...the force is strong with this young guy")

Attached Files

•  EthyleneStorage_System.zip   13.52KB   78 downloads

[joerd]

Here it is.

 EthyleneStorage_System_revised.zip   26.87KB   95 downloads

Remember, one of the most important things when doing a simulation is: don't overspecify your problem!! If you specify temperature, pressure and composition, the vapor fraction follows. Normally, you don't need to specify the vapor fraction, let HYSYS figure it out. You'll see that I did specify the vapor fraction after the ethylene condenser. This is done to simulate a total condenser, which produces a bubble point stream. So the pressure of this stream is a free variable (like I tried to say before, given the composition, you can only specify 2 out of pressure, temperature, and vapor fraction).
Then, I noticed that your inlet flow rate in the simulation doesn't agree with your post. So check it.
Also, you'll need to check all input parameters, since I made them up, such as condenser outlet temperature, compressor efficiency, heat input into the tank, etc.
Good luck

[Dtan76]

Dear Joerd

Thanks for prompt reply, I know where is the mistake there.

Apart from the Compressed/Chiller/Cooling, vapor recovery system, I did modelled the downstream Ethylene pre-heating system. The Pre-heat/Evaporator will heat up the Ethylene liquid rundown from the tank to vapor Ethylene, prior discharge to pipeline connect to end user/plant, 5 KM away

Enclosed the revised Ethylene simulation file (incorporate the preheater) and simpliefied schematic here.

Please refer to the Simulation/Schematic for explanation

Steam available:
let say 150 degC at 4.67Bar(a). The amount of steam estimate is 3 te/hr, condensed at 30 degC.

My query here:
a) Assume the I choose the preheater - shell & tube BEU type, I obtain the heat duty of heat exchanger = 2,288kW. 1 series + 1 parallel. What is the advantage if I using Kettle Reboiler type? I'm asking so becoz most of the people told me using Kettle Reboiler for steam at tube side and product at Shell side. I seen the KB type using for application - steam condensate at Isothermal and release latent heat to evaporate the product. What if I'm using BEU type? Maybe you can share some.

If I'm using double shell - 2 shell in series, I expect I'm getting smallest shell ID compare if I only specified 1 shell, 1 parallel. (Correct me if i'm wrong)

c) Anyway, can we modelled the Tank as "bath" instead of continous flow process? Notice that the Tank eventually is a batch - means that the tank is filled up with 20,000m3 working capacity and start to unloading to discharge to pipeline (downstream), truck or "backwash " the incoming pipeline (during ship/vessel loading, to purge the line content).

Thanks for help.

Regards
DTan

Attached Files

•  Ethylene_Downstream_Preheating.xls   33.5KB   84 downloads
•  Ethylene_VaporRecoverySystem.zip   39.67KB   65 downloads

[joerd]

@Dtan

I have a few observations re. your preheater.
1) If you model water / steam, you should include a separate Basis, based on Fluid Package NBS Steam or ASME steam, with one component (water). Hysys will allow you to select a different fluid package for each side of the exchanger.
2) We would normally specify the steam pressure at the exchanger, and either the temperature (if it's superheated) or the vapor fraction 1.0 (if it is saturated steam). Then, specify the pressure drop and the outlet vapor fraction 0.0 for the condensate. Rarely are exchangers designed to subcool the condensate.
3) If you want to design an exchanger, Hysys is not very good. It doesn't know kettle type, for example. So, consult a heat transfer engineer, and use HTFS or HTRI software to design the exchanger. There is a lot of information out on the web (the Wolverine site jumps to mind) if you want to read more about heat exchanger design.
So, to design a heat exchanger, figure out the process conditions from Hysys. Set the Heat Exchanger Model, on the Design tab, Parameters page, to Exchanger Design (Weighted), specify (as you have done) the outlet conditions, allowable pressure drop, and print the datasheet, and take it to an expert.
Only if you specify as Heat Exchanger Model: "Steady State Rating", then Hysys will look at the provided configuration information. I have only used this method in revamps, to see how a particular exchanger will perform in a new situation. However, that means that you can verify that the predicted performance in the current situation can be verified, either with plant data or backed up by an HTRI run. You'll find that you have to tweak the Hysys "configuration" to match actual performance, which means that the prediction is not very good.
A kettle should work OK for this situation, and will ensure that you're getting only vapor out of your exchanger. HTRI will do entrainment calculations.
This pretty much answers your question a)

Re. your question , multiple shells gives you smaller shells, but a higher cost since you'll have more equipment items. So, up to a practical diameter (mechanical constraint), a single shell is the way to go.

Finally, c) if you want to model a batch type process, you'll need a dynamic simulation, if you really want to see the tank fill and empty like in real life. What you could do in this situation without purchasing the dynamic license, is model the system as 2 separate steady state models: one for filling, and one for emptying the tank. The composition of the ethylene would be the same in both models, but the flow rate is differrent. So, copy your stream toload_1 from the tank to a new stream, say toload_2. Put the pumpout flow rate in toload_2, and connect it to the TEE-100 instead of toload_1.

Hope this works for you!