16 replies to this topic

[aanita]

Debutanizer column temperature controller (Master controller) TIC, is controlled by steam flow measurement inlet of reboiler, FIC cascade with reboiler outlet control valve, FV. Control valve, FV is getting signal from lower output between FIC and Reboiler level troll, LIC. Most of the time reboiler level is not holding. Steam inlet properties: 40 barg saturated, reboiler outlet steam condensate (SC) is collected in a pot where flash steam en route to low pressure steam header and steam condensate to the header.  Level in reboiler, LIC maintains to avoid steam blow out from reboiler.  My question is why level not maintain all the time. Anyone face the same problem?

Attached is the sketch of the reboiler with column.

Thanks in advance.

Attached Files

•  Reboiler level.jpg   59.74KB   12 downloads
•  Reboiler-1.docx   211.03KB   48 downloads

Edited by aanita, 11 October 2013 - 08:10 AM.

[Bobby Strain]

You didn't supply enough information. All we can do is speculate. You have all the information, so you should carefully analyze it. Share it and maybe we can contribute.

 

Bobby

[aanita]

I am not sure what sorts of information required to analyze, however my concern is why in the level troll, level is not holding all the time. Sometimes level is holding and I don't see any specific reason. Inlet steam pressure/temperature is steady, reboiler/ column liquid level also maintain. For mechanical consideration, I am not aware about it as a plant operation staff. However reboiler bundle is always submerged by process liquid.

[Bobby Strain]

Add when,how, etc. Knowing elevations of the reboiler and pot would help, too. Duty, and pot dimensions.

 

Bobby

Edited by Bobby Strain, 11 October 2013 - 03:58 PM.

[paulhorth]

Aanita

It seems to me that with the control system you have shown, the reboiler heat duty is controlled by the condensate level rising or falling, to cover or uncover the surface in the tube bundle. So, the condensate level is not meant to be constant but would change in response to column reboil heat demand, such as change in feed flow or temperature.

In some steam reboilers,there is a valve on the steam supply, and with this valve, the steam pressure can vary to control the heat duty. Your system does not have a steam valve so the steam pressure is fixed. The only way the heat duty can vary is for the surface area to be partly flooded.

Paul

[K J]

I think your your valve operation may be vary slugish due to improper tuning. let say as per your TIC demand your FIC out put will increase and so condensate level. now your FV opening will start increasing and attains higher opening then due to increased temperature in column reverse action happens but before FV opening come down condensate level drops below LIC troll.

[Bobby Strain]

So now the guessing begins.

 

Bobby

[aanita]

To K.J.: Control lop is working fine and I didn't find any tuning problem. Column control temperature is being well maintained.

To Paul: Since tube bundle is submerged all the time, it is not partly flooded. As per elevation/ Column configuration, since process liquid level in the column maintain, reboiler tube bundle is submerged. Steam condensate level also constant.

 

[curious_cat]

To Paul: Since tube bundle is submerged all the time, it is not partly flooded. 

 

If your bundle is fully flooded as you say it is, can you please explain your reasoning as to how your control scheme works? What is the manipulated variable. 

 

Say, your TIC indicates a small drop in T from set point, how will it be corrected by your control scheme. 

[Babu Prasad]

Debutanizer column temperature controller (Master controller) TIC, is controlled by steam flow measurement inlet of reboiler, FIC cascade with reboiler outlet control valve, FV. Control valve, FV is getting signal from lower output between FIC and Reboiler level troll, LIC. Most of the time reboiler level is not holding. Steam inlet properties: 40 barg saturated, reboiler outlet steam condensate (SC) is collected in a pot where flash steam en route to low pressure steam header and steam condensate to the header.  Level in reboiler, LIC maintains to avoid steam blow out from reboiler.  My question is why level not maintain all the time. Anyone face the same problem?

Attached is the sketch of the reboiler with column.

Thanks in advance.

 

 

 

 

To K.J.: Control lop is working fine and I didn't find any tuning problem. Column control temperature is being well maintained.

To Paul: Since tube bundle is submerged all the time, it is not partly flooded. As per elevation/ Column configuration, since process liquid level in the column maintain, reboiler tube bundle is submerged. Steam condensate level also constant.

 

Your question is is why level not maintain all the time but last reply you mentioned Steam condensate level also constant..

As  KJ & paulhorth  pointed out correctly

Generally steam inlet will be designed to have separate FCV to control stream flow based on TIC requirement and condensate will be controlled independently depends upon condensate level either from reboiler or condensate pot. FIC & LIC controller action to be tuned accordingly .

Edited by Babu Prasad, 15 October 2013 - 09:02 PM.

[aanita]

To Babu Prasad: I attached a sketch of reboiler loop for better understanding. To avoid steam blow-out from reboiler, level is maintained and confirm from LIC. Steam condensate out from reboiler loop is also maintained from different condensate pot which is independent of reboiler loop.

[Babu Prasad]

To Babu Prasad: I attached a sketch of reboiler loop for better understanding. To avoid steam blow-out from reboiler, level is maintained and confirm from LIC. Steam condensate out from reboiler loop is also maintained from different condensate pot which is independent of reboiler loop.

 

It is little bit confusing because, In topic #1 , you asked why level not maintain all the time which is  contradict  to topic # 8 reply to KJ,  you mentioned  Steam condensate level also constant.

I considered that, you are facing level fluctuation in reboiler condensate level.

As Mr. Paul correctly said, reboiler heat duty is controlled by the condensate level rising or falling, to cover or uncover the surface in the tube bundle Hence it is difficult to have stable level with present control setup.

As per your drawing, FV on condensate line receive command either from FIC or LIC thro a low selector which decide the FV valve opening. It is quite difficult to synchronize two controllers in parallel to control one control valve particularly with your current steam flow control  

Normally, condensate FV opening will be controlled by LIC and sometime FIC will take control and gives close command based on input from TIC requirement, during that time your level will be high and LIC will be generating higher output up to 100%. After sometime FIC generates higher output based on input from TIC requirement during that time your level will drop fast and LIC will come into action and takes the control over FV. So it is cyclic action.

 

My recommendations,

 

1st option: If Possible use LIC as slave (instead of FIC) to TIC on temporary basis to study out. It is not industrial practice to have LIC cascade with TIC but on trial basis it can be tried. Try to limit TIC output from 20% to 80% range so that LIC will be operating between 20% to 80% as your reboiler heat duty is controlled by the condensate level rising or falling. TIC has to be configured as reverse controller & has to generate lower set point to LIC when TIC needs higher heat duty. TIC needs to be tuned slower and LIC needs to be tuned medium faster.

2nd option: with existing setup if you are operating plant at higher load, keep LIC setpoint lower around 25% and tune LIC medium faster so that LIC will try to minimize the offset created by FIC. Other way if your plant load is less keep LIC setpoint higher  around 75%  and level will be controlled between 60to 75% . This operation will minimize level variation.

Edited by Babu Prasad, 16 October 2013 - 04:44 PM.

[Bobby Strain]

Believe me, option 1 above will NOT work. If you would supply the information I suggested, we can be more helpful.

 

Bobby

[Babu Prasad]

Mr Bobby, I agree with you, first option is not standard one but it was suggested as an alternative to existing configuration and there is no need of field modification required, only software modification in the DCS is enough to perform this change even a selector to choose the option between LIC & FIC will allow the user to switch over. Generally steam flow will be control at inlet with FCV and condensate by LCV. As control valve situated on outlet, the reboiler is always fully charged with steam and reboiler duty can be controlled only by varying level to limit the surface area for heat transfer.

If existing control setup does not disturb tray temperature widely then existing control system is optimum one for this type of design and you cannot expect steady control over the level with Low selector option which provided as safe guard to maintain the minimum level in reboiler to avoid steam blowout. My second option may limit level variation if controllers tuned as advised.

Attached Files

•  LIC.JPG   65.36KB   5 downloads

Edited by Babu Prasad, 17 October 2013 - 06:10 PM.

[GSaikrishn]

Aanita: as per ur drawing, pls correct me if im wrong, the 40 barg steam is on the tube side. so the column liquid is essentially on the kettle side. Why are u trying to maintain the level at the steam side.

 

Shell side: 

as per ur sketch im assuming, the steam tubes will be submerged in the column bottoms liquid. The tubes will always be submerged because of surface boiling and the level on the kettle side will depend on the column level and column operating pressure.

 

Tube side: I can think of the following reasons for the level not holding.

a)The saturated steam is not always say pure in nature, it has fouling tendencies i.e. the condensate results in fouling on account of TDS (total dissolved solids).

Change in the bottom composition, see if there is any change in the column bottom temperature. It may so happen that the approach between the column bottoms and steam has reduced, so there may be possibility of the steam not condensing. (i.e. latent heat not fully utilized). 

c) Now if ur steam condensate is routed to a flash pot, the control valve FV is floating with the condensate pot. Please check if there is any pressure fluctuation in condensate pot vent which is being routed to a low pressure header. Increase or decrease in the vent header pressure will result in fluctuation of the FV opening and hence the level.

d) Also on the steam side, please confirm if it is indeed saturated steam at all times, in fact it is normally desirable to put slightly superheated steam. there may be possibility of slugs entering the reboiler.

 

I can think of these reasons:

 

I have following recommendations:

a) If indeed there are any changes on the column bottom end, i.e. increase or decrease in the column bottom temperature profile, try normalizing. (see if there are any pressure fluctuations (day/night phenomenon) in the column, change in pressure will change the column bottom temperature and hence the steam requirement should change. Even if there is a TIC in the bottom, check if the console guy is making any changes to its set point. Also check if the column bottom level is changed.

Disconnect the LIC signal. It should not play any role. Cascade the bottom temperature TIC with the FIC and control the FV, because the tubes are anyways going to see two phase flow as the steam is condensing, so level should not matter.

 

Hope this helps

[GSaikrishn]

Also how are u controlling the condensate pot level. 

[aanita]

As industrial practice there is little bit fluctuation always which should be handled in the control loop. However LIC kept OFF(manually LIC output keep 100% so that only TIC will take action for FIC) for operational convenience to avoid any malfunction, which we don't want. Column bottom level is under control.