10 replies to this topic

[PE123]

Dear All,

BACKGROUND
I am in the midst of relieving a constraint in the steam condensate system in a butadiene plant.
As attached. exchanger, Steam condensate system in the aromatics plant and butadiene plant are integrated.
E-940 is heated up steam codensate from D-982. During normal operation, steam condensate from the aromatics plant and butadiene plant is sent to D-982. Steam condensate undergoes flashing at 5.6 kg/cm2G and steam condensate ( ~ 160 degC) is used to heat up E-940.
Under normal operation of D-982 level control, LCV2 is closed and D-982 level is controlled by LCV1.

CONSTRAINT
During the shutdown of the aromatics plant, the amount of steam condensate sent to D-982 decreased significantly. Subsequently, LCV2 (Recycle valve) is opened in order to maintain D-982 level, introducing "cold condensate" to the drum. As a result of the lower steam condensate temperature, E-940 heat duty is affected and the tower tray temperature cannot be maintained.

PROPOSED SOLUTION
I have a proposal to heat up the steam condensate by injecting medium pressure steam ( 14.5 kg/cm2G, 280 degC) into condensate drum (with the help of some distributor or spray nozzle) to heat up the steam condensate. The heat transfer principle i imagine is similar to the stripping steam in a deaerator where the steam heats up the boiler feed water (Although the stripping of O2 in the boiler feed water is not similar).

Can anyone advice on the feasibility of such an operation and what are considerations to take note ?
thank you.

fsk

Attached Files

•  system condensate system.pdf   30.27KB   131 downloads

[TS1979]

Not a very good arrangement. The addition of steam to the drum will potentially reduce the NPSHa for the pump and cause problems for the pump operation. A baffle to separate the pump suction and steam injection may mitigate the problem.

The another problem you should consider is the vessel design conditions: design pressure and temperature. The pressure of the medium steam is quite high compared to your current operating condition. Will the medium steam overpressure the vessel?

Can you directly introduce steam to the heat exchanger? This option may be easier to implemented.

[Art Montemayor]

This proposal may have possibilities of working out, but first you have to generate a detailed heat and mass balance regarding the adiabatic mixing of the condensate and the sparged steam. What you essentially are proposing is a low-pressure, direct-contact, steam generator - a de-superheater, working in reverse. You have to calculate the steam flow rate required to generate the hot, saturated condensate you require. Knowing the steam flow that you must sparge into the condensate drum, you can then determine what size safety relief valve you need to protect the condensate drum in case the steam inlet control valve fails open or gets stuck. You may – or may not – have enough PSV nozzle capacity to install the appropriate PSV. In any event, you should install a restriction orifice downstream of the sparging steam control valve in order to ensure you control the maximum flow rate of injected steam.

You fail to state or show how you intend to control the flow rate of sparge steam into the condensate drum. I presume you propose to do it by controlling the condensate vapor pressure created in the drum.

You will obviously experience violent steam condensing noise – similar to cavitation – and the condensate level will be constantly moving about due to the agitation caused by the sparging. This can’t be avoided – depending on the size and configuration of the drum and the internals. To sparge successfully, you require an effective height of condensate in the drum that will be dedicated to direct-contact heat transfer – something you may (or may not) have available in the drum height. This is probably the most difficult obstacle to overcome.

I don’t foresee any NPSH problems for your condensate recirculation pump. The vapor pressure conditions of the condensate won’t change, and neither should the positive height of condensate over the suction of the pump.

I do not believe you have the option of injecting steam into E-940 heat exchanger. First of all, you would convert a liquid coefficient side to a condensing film side – and you would have to undergo a phase change with condensate removal (steam trap?) capability. This is asking a lot out of what appears to be a liquid-to-liquid heat exchanger. Besides, you have to re-rate it to ensure you have enough heat transfer surface and that the shell side has a sufficient MAWP for the steam.

[breizh]

As you will have time to shut down, you may consider a permanent solution - like having a second heat exchanger to heat up the stream from E940 with, let's say, steam . This will not jeopardize your existing set up.

Breizh

[acer_asd]

You can also explore the possibility of feeding the steam condensate in flash drum from bottom. If you feed from bottom steam generated due to flashing will come in direct contact with cold condensate and it will heat it up again. This will reduce your LP steam generation but in this case, you do not need to put additional MP steam.

[PE123]

Dear all,

Thanks for the reply.

I have done some heat balances based on enthalpy. Basically, to heat up ~ 9,225 kg of steam condensate from 135 degC to 162 degC. I need ~ 444 kg/hr of medium pressure steam (230 degC, 14.5 kg/cm2G).

I have assumed that all the steam injected will be condensed, and thus heat up the steam condensate. Attached is my Process Flow Diagram, where it is seen that steam condensate controls the amount of steam to be injected. However, if the vessel pressure becomes too high, the pressure controller takes control. Safety features such as the restriction orifice provide another protection against over injecting the steam.

I will check the safety valve capacity, based on the maximum steam injection.
Kindly ignore the steam injection at the inlet of the exchanger, as this steam injection is inadequate for the heat duty because the injection rate is limited to 0.5 t/hr due to the high line pressure (Suspected to be 2 phase flow and steam is not able to condense).

Regarding the detailed design of a steam sparger, how can I be confident that steam injection will be sufficient to heat up the condensate to its desired level? I am not so certain that merely steam injection alone is able to alleviate my plant constraint. As such, I have decided to construct 3" line from the ethylene plant to transfer medium pressure condensate to my condensate drum. With this modification, the drum level can be maintained and the recycle valve need not be opened. My plan is to install the nozzles first at the next opportunity for the steam condensate transfer. The lines will be constructed on stream if direct steam injection does not suffice.

Can anyone advise how can I ascertain the probability of success of the steam injection into the steam condensate drum to raise its temperature? What other things can i check or do?

thank you

fsk

Attached Files

•  Steam Condensate PFD for Forum.pdf   75.55KB   101 downloads

Edited by Art Montemayor, 26 May 2012 - 10:57 AM.

[Art Montemayor]

fsk:

You have not submitted your calculations that show how you arrived at the amount of steam required for sparging into the cold condensate, so I have to assume they are correct. I estimate the required steam line size for the sparger to be approximately 1” nominal size. Please refer to the attached workbook where you will see that I cannot complete the heat and mass balance because of a lack of basic data: I don’t know the thermal condition of the “cold” condensate that you are sparging.

The first thing I would recommend you do is to make sure that your heat and mass balance calculations are sound and correct.

The next thing to be concerned about is the amount of internal condensate volume that you can dedicate to the sparging and mixing of the superheated steam and cold condensate. As I indicated in my prior post I identify this item as the most important item in this revamp. However, bear in mind that you can only achieve the capacity and success that the existing condensate drum dimensions allow you. That is the main thrust of the scope of work before you: You must subject yourself to the existing condensate drum and work with its design limitations. You should try to maximize the amount of cold condensate subjected to the steam sparging in order to ensure that the mixing will be complete – and with a minimum of internal agitation. This means that you will have to sacrifice some of the vapor space that exists in the drum. Additionally, it is important to pay particular attention to the type and size of sparger that you employ. I recommend that you employ a sintered metal (stainless steel or brass) sparger. I have used these before that the effect they give is a gentle flow of very fine bubbles that yields a minimum of agitation and movement. Usually you can obtain these spargers as pipe sections that are simply flanged internally. The sintered metal works around the entire periphery of the sintered pipe. You should be able to find suppliers on the internet.

I apologize for not formatting all of the steam thermo data into a nice, complete table. I am having problems with my Excel program that has suddenly failed to copy and paste the NIST HTML tables directly into Excel as it has done in the past. I had to spend a lot of time fitting the data that I think are important to you into the correct columns and I didn't want to hold back this information from you and hold up the thread.

Good Luck.

Attached Files

•  Steam Injection into Condensate Drum.xls   914.5KB   106 downloads

[TS1979]

Heat balance and mass balance calculation should ensure you that by injecting enough steam you reach the temperature required. However, for this configuration, it seems very likely that steam will be trapped in the condensate to the suction of the pump depending on the location of the steam sparger. The trapped steam will potentially create water hammer, cavitation for the transfer pump even if the NPSHa does not change.

[Zubair Exclaim]

First of all do a small cost analysis to see if installing a second smaller heat exchanger in series which runs on steam is cheaper than this arrangement ,... if so then do that.

secondly have you confirmed that the overall flow rate of the condensate (under turn down condition) to exchanger doesnot drops too much. i.e, just the droping of condensate temperature in drum is the cause or is the over all flow to condensate even at desired temperature (after steam injection) too low for heat duty.

Thirdly yes steam injection can be sufficient to raise the temperatue of the condensate provided that you do your heat and material balance calculation right. You just have to make up for the enthalpy lost in the recirculating condensate and make up that enthalpy through steam injection. But should see following :

• Condensate in drum is not brought above equilibirium temperature
• there is ample depth of water above sparger to ensure no steam escapes
• Check pump NPSH calculations
• you may consider using a baffle between sprager and pump intake nozzle

Edited by Exclamation, 22 July 2012 - 04:52 AM.

[PE123]

Hi Exclamation,

1) There are site constraint which limits the installation of an new exchanger in series. Assuming the new exchanger (Increasing the Area) utilizes steam condensate as a heating medium as well, there are doubts if heat duty is sufficient, considering that the heat source ( Steam condensate is limiting).

2) The low heat duty of the is caused insufficient steam condensate. As such, the recycle valve opens to maintain the steam condensate drum level, introducing cold condensate into the drum. As such, temperature of steam condensate in the drums drops, causing the heat duty of the exchanger to drop even further -- > Requiring even more flow to the heat exchanger -- > Recycle valve opens even bigger to maintain drum level -- > Steam condensate temperature in drum drops even further.

3) I am interested in your considerations listed in your point 3. However, i could not fully appreciate.

• Could you explain the significance of "Condensate in drum is not brought above equilibirium temperature"
• How do i check the effect on NPSHA with the introduction of the steam sparger ? I am familar with NPSH calculations. But how do i take into account the effects of the sparger?
• Do you have any examples of such baffle plate installation ?

thanks for your help
PE123

[tomyaf]

Dear PE123,
Sorry if I ask such a silly question, but I don't see any necessity to add another facility to solve your problem.
Here's what I think about your problem.
1. During aromatic plant shutdown, you need to maintain level of D-982. Why don't you just stop the pump or do full recycle (close LCV1 manually)?
2. During aromatic plant shutdown, you need to maintain pressure of SL Header. You already have SM header connected directly to SL header as backup in case the SL supply from D-982 is reduced. Why do you propose to build another SM header connected to D-982?
Please CMIIW.