Column Packing fall down 3

[SHARQ]

We have a column used for CO2 Stripping. it contains two packing layesr inside. Each packing is 9 mtr height. The packing is Rashig Rings IMTP # 40. The event happened are as follows

1- Plant was under shutdown
2- As per logic, CO2 unit circulation was kept running, Ejector steam isolated and the reboiler stopped after one hour and the column level was around 45%
3- Initial CO2 level was 48% and it started gradually increasing and reach 94%
4- The level was varying from 50% to 90%
5- Plant startup after 36 Hrs and the level was at 90%
6- Operation lined up both the reboiler and the direct steam but level not decrease.
7- Operation confirmed the level check in field LG and found the level is high
8- after 9 hours from startup the level reduced to 50% and one ejector (out of two) stopped.
after one hour Delta pressure increased from .007 to .035. And the reboiler flow started fluctuation. (some times showing zero and some times shwoing a vlaue)
9- after 3 hours islated ejector take in line but condition not improved. level not increased but Delta P still remain higher side.
10 - Operation asked maintenance DP TX for the column and TX of the reboiler and found OK.
11- Opening of the Reboiler control valve (thermosyphone valve) from 30% to 50% and found no change of flow
12- Operation checked in site the column and reboiler for any abnormality (sound / hammering) and found OK.
13 - Column other parameters (Temerature, circulation flow, gas flow) all remain steady and normal
14 - Operation decided to check the pump circulation and tried to isolate the pump but not succeeded and the suction valve was stucked up at 85%
15 - Operation asked maintenance to check the AOV of the pump suction and found OK, but stil lthe valve could not be closed.
16 - Delat P remained high and the reboiler flow remained fluctuating.
17 - after 7 dats of operation , operation decided to isolate one ejector to see the effect on Column Delta P. the flow of the reboiler became normal and delta P decreased and reached 0.012 and all other column remained normal
18 - after 9 hours of operation with isolatee ejector, operation faced problem to maintain the lequid level of clumn. So they reduced the load in the CO2 unit increas carbonate flow but not succeeded.
19- After 1 hour from this event, operation decided again to take back the ejector in line and the reboiler flow started to decrease and level not getting controlled
20- after one day of operation with these conditions. operation observed the pump circulation flow started to fluctuate and slowly going down. They changed over the pump from A to B to clean the pump suction strainer. However, the valve could not be closed for isolation
21- maintenance tried to isolate any of the pump and found both AOV suction can not be closed. Moreover, cavitation sound apperaed from the pumps
22- Pump sound started to increase and operation vibration were high. So, operation decided to plant shutdown and to inspect both pump suctions strainers
23- Both pump suctions were opened and found both fillwed with Rashig Rings and the whole suction line from the colum found filled with RAshig.
24- Operation decided to open the manhole of the columns and unfortunately found in MH # 1 Huge amount of Norton IMTP Rings were found accumulated on the bottom head of column. Almost all packing rings were found as crushed.
25- Found on MH# 2

A-Crushed packing rings accumulated on the baffle cover.
B-Liquid collector fallen down & found damaged.
C-Bottom packing support was found dislocated & some gaps were noticed between segments.
D-Some damages/bends were noticed on the schopentoter. It may occurred due to falling down of liquid collector.

26- Found in MH # 4
A-Bottom bed level has been gone down (approximately six meters).
B-Redistributor was found dislocated & damaged.
C-No abnormality noticed on the top bed support.


Please help me to figure out the root cause of the problem

Regards

Samir

 

[unclesyd]

I saw damage to a packed column similar to what you are describing where operations managed to "bump" the column internals with a high gas flow and/or steam. By bumping I mean that a sudden increase in the gas flow tried to lift the packing and liquid in the column. This occurred several times in short peroid and the results essentially the same as yours, failure of the internals.

 

[ash9144]

As I read through the scenario I had the same thought as unclesyd. High level combined with a volatile compound on start up can result in a "bump". A large amount of volatile flashing with some circulation is just like hitting the bottom hold up tray with a sledge hammer straight up.

 

[SHARQ]

We are suspecting about the line up of the ejectors were not as per procedures which lead to hammering inside the column.

Our concerns are the followingh

1- What caused the delta P to raise ? and what does it mean
2- Why the rashig rings found crushed and deformed.
3- Why is the reboiler got fluctuation


You may help me on finding more root causes

SHARQ

 

[RogerH]

Samir
Agree with what the previous replies indictae, The level was high and direct steam (as per p.6) was started and I assume the ejectors were in service. This caused a "bumP that lifted the holdup tray. Then the collapse came and blocked the outlet, reboiler etc which you could see on the starvation of the reboiler, resulting in "jumping" flow.

High level, if it were 90% or higher might be debatable. Particularly how this is checked. If one doesnt drain the glass and refills during checking the level in the LG might not be true. Conclusively it could be a lot higher than expected. LI's can freeze on a high level fooling one to believe this still is in range.
The crucial step, I guess, is the introduction of direct steam under (partial) vacuum conditions, this can create a large lift of liquid and to the hold up tray. Then the rings gets crushed, intially during the bump(s) or during a later stage when falling and the irregular packed semi-crushed rings results in the pressure drop you did see.
Without detaild knowledge of the column and trends I guess thats what can be concluded.
BTW: a very good step-by-step description of the event, it was very useful
Roger

 

[siretb]

9 mtr (I understand meters), is a lot.
do you confirm you have two beds, each 9m high? If yes this is too much.

 

[SHARQ]

sirteb

the colimn is 31 mtr height, 4.4 dia and contains inside it two sectiond of packings (upper and lower). each section is 9 mtr height. If you want, i can send u a drawing. the problem of this site, u can not insert drawings to explain more


Regards

 

[SHARQ]

unclesyd and ash9144

can you elaborate more about the bump phenomena and how do you think the scenario went on ?


We have collected 21 causes but frankly we did not get into this bump at all. I am now re drawing the scenario and found your saying is a strong possible cause. I will go and see the control valve opening what was and how fast was it.

for your information : the column faced the same trouble 1 year ago, and the process enginers resulted thier report due to hammering.


Regards

 

[SHARQ]

Dears

in order to view the column pic. I have flickered it and upload it in the following link

http://www.flickr.com/photos/13946217@N02/1417956239/

I hope we can now start analysis in with a visual ideas along with ideas.

 

[ash9144]

Hammering could just be a different term for bump. Though I would not use it that way.

By bump we mean that on start up you have too much liquid in tower. Maybe even above the hold up tray. When you start heating (or pulling vacuum) you get a large amount of flashing since the start up liquid would be all feed liquid. This amount of flashing is much more than would normally boil at the reboiler. The sudden flash of volatiles back into the liquid pushes a slug of the liquid upward rapidly. This moving liquid has enough mass and velocity to lift the hold up grid which allows your packing to fall into the bottom of the column. And as explained above the falling of the packing is what creates the other symptoms you saw.

Most people take steps to prevent this such as starting up with a liquid other than normal feed if possible ideally you could put the bottom liquid in the bottom and heat up then start feed or adding a small line for steam.