Naphtha stabilizer operation

[Jan1182]

Hello,
I have a question about a naphtha stabilizer which processes wild naphtha coming from a gasoil hydrotreater to remove H2S and control naphtha RVP.
Our problem is that since the beginning we do not have any reflux and so the naphtha is not stabilized. It's probably due to a lack of light compounds in the wild naphtha.
We tried to push up the reboiler temperature, to decrease a little bit the column pressure but always when the reflux pump are started, the drum level fall down immediately.
Could you help me?

 

[Milutin]

Jan,

Is your RVP and H2S content from column bottom OK?

Regards,

 

[0707]

Go to thread
Reboiler in the column
thread124-100423

Maybe it gives some light to your question.

By the other side

"The stabilizer feed contains traces of water and ammonia, which, along with the trace levels of HCL, produce undesired side reactions, and the bulk of the operating and maintenance problems associated with stabilizer operation.
Depending on the amount of impurities present in the feed the net result is costly corrosion, and plugging and downtime of the fractionator and perhaps the entire unit"

If you tried to push up the reboiler I suppose you have a plugging problem on trays. If possible you can try on line dispersants to unplug the trays or scan the column with a TRUTEC scanning or equivalent to see in what column section you have the plugging.

For column scanning go to

http://www.tracerco.com/products-processdiagnostics-towerscan.htm

Good luck

Luis marques

 

[mbeychok]

Jan:

What is the composition of your "wild" naphtha? Does it contain any butane or propane? What is the pressure and temperature in your reflux drum?



Milton Beychok
(Visit me at www.air-dispersion.com)
.

 

[Jan1182]

To mbeychok:
Pressure and temperature of the reflux drum are 7.5 barg and 60°C. We think that the amount of light compounds as C4-C5-C6 in the naphtha is not sufficient as we are running at low temperatures in the reaction section. And so, the level build up in the reflux drum is very slow.

To 0707:
Feed Temperature 127.5°C
Top Temperature 82.5°C
Top Pressure 7.7kg/cm2
Reflux Flow Nil
Bottom Temperature 188.0°C
Reboiler Outlet Temperature 196.7°C
Sensitive Tray Temperature 130.5°C
Maybe you are right, there is something with the reboiler as even with a high temperature at the reboiler outlet we do not arrive to obtain a high enough temperature on the "sensitive tray".

To Mulitin:
That is the problem, we have a lot of H2S in the naphtha from column bottom...

 

[Milutin]

Jan

You can try decrease operating pressure.
I suppose that column simulation, in this case, can help you to decide optimum operating parameters for naphtha stripper.


Regards,

 

[sshep]

Jan1182,

The temp profile is nice data, but the real question is: can you close an energy balance on your column? For example I wonder how you can increase the reboiler temperature (at lower pressure no less) if you can't take any distillate (an assumption I make from no reflux, although some vapor distillate might still be possible).

Closing the material and energy balances are the first steps to diagnosing this type problem. The overall material balance is probably easy, but I don't see how the energy balance could possibly close from the data you give unless you just aren't giving the tower enough heat. First see if the energy input at the reboiler matches the heat removal at the condenser (utility side balance)- this will determine if you should actually have reflux. Only after the energy balance closes (or fails to close) on the utilities will you be in a position to draw conclusions on the reflux or boilup.

A few weeks ago I was troubleshooting a column with far lower than expected reflux and problems with high pressure. As hot vapor was carrying over to the tower vacuum system, I suspected a condenser problem. This theory was destroyed when I measured the cooling water flow (ultrasonic) and found the heat removal exactly matched the reboiler input. From this clue I suspected free water and had them check- sure enough the reflux was 50% water (from leaking reboiler). This explained everything: 1) higher heat of vaporization of water --> low reflux flow; 2) lower condensing temperature --> condenser now undersized --> higher operating pressure and vacuum system carry over.

Tell us about the energy balance!

best wishes,
sshep

 

[Jan1182]

sshep:
It is a total reflux column, there are no liquid distillate out of the reflux drum, only the light vapour compounds (mainly H2S, C3,C4) go after to a LP amine absorber. All the liquid condensed (if there were some or enough) goes back to the column.

 

[sshep]

Good, this explains how you can change the bottom temperature.

You are either not putting in enough heat to get real reflux, you are pushing uncondensed hydrocarbons over to your amine system (you would know it), or you have "hidden" reflux (such as my example). Which of these will be determined by calculating your reboiler and condenser duty on the utility side. Is your condenser using air cooler, water, or process stream?

I am sure you can solve this. Do not go straight for a tower scan because it is probably unneeded and will likely give little new insight to your troubleshooting.

best wishes,
Sean

 

[Milutin]

Jan/Sean

If water present because of reboiler leaking it can be found by naphtha hazy appearance or you can make naphtha analysis, water content should be up to 150ppm, but if leaking present up to 1000ppm.

Have you ever have periods with good H2S content in naphtha or it is bad from plant startup?

Regards

 

[Jan1182]

We got an air cooler to condense the gas at the top of the column.
Here an analysis of the reflux dum gas:
H2 29.6
H2S 54
C1 1.86
C2 6
C3 5.2
i-C4 0.74
n-C4 1.7
i-C5 0.5
n-C5 0.4

 

[Jan1182]

Mulitin,

we never got a good H2S content in the naphtha since plant start-up. And never arrived to have a reflux flow.

 

[0707]

Jan1182

If your h2s is not being retrieved from the top of the stabilizer column that is a sign that something is wrong with your reboiler or with your trays. My feeling is that your trays are mechanically damaged or plugged obliging your h2s going to the bottom. Do you also have in your system a feed bottoms exchanger? Maybe your feed is contaminating your bottoms column decreasing h2s release and leaving it in the bottom.

luismarques

 

[Jan1182]

I don't think that it's because a leakage on feed/bottom exchangers. It looks more like to my mind, that there is a problem with liquid/vapour exchanges on the trays. Too much liquid because the feed is heavier (less light compounds than design)than expected. I mean that the there is almost no light compounds (except some H2, H2S, C1, C2,C3) going to the top of the column and so no condensation at the top of the column and so no reflux. All the naphtha is going down without any separation because oversized column.

 

[epoisses]

Pump up the reboiler and see what happens.
I suppose you won't be able to drop the pressure much if your gas is supposed to go to an Amine unit.
Are the upstream units now at normal conditions? Try to match pre-turnaround tower operation as closely as possible. I know how difficult this can be, especially if they recalibrated all flow meters from volume to mass during the stop, as once happened to me (that gave almost a week of fun during startup!).
Intuitively I don't see how the tower operation can be completely messed up if the feed is somewhat heavier than usual, if it all seems very strange than it could well be that there is tray damage (any corresponding pressure spike observed during startup?) and you should line up a scan as soon as possible. It costs less than a few hours worth of lousy operation.

 

[CJKruger]

Jan1182,

1) Using reboiler outlet temperature as measure of reboiler duty is notoriously bad for naphtha strippers. You have to use steam (or fuel gas if fired heater) flow rate.

2) Simulate the system (hysys, proii, etc). Make sure that you can achieve a V/L ratio of >0.3-0.4 in the bottom. Else you will not get the H2S out.

3) If this all works out, scan the tower to look for damaged/plugged trays.

 

[Milutin]

Jan

You never have on spec naphtha, for how long you running hydrotreater, or in other words did you have until now stripper trays inspection.
It might be try damage, but I have case in diesel hydrotreater tray damage because water incident, flash point was worse then usually but on spec.
Tray plugging due corrosion take some time, but you have problems from startup.

I have case with gas oil hydrotreater originally designed for visbreaker gas oil hydrotreating. Original feed was wide cut something like 175-350degC cut. When we wanted to hydrotreat straight run gas oil from ADU, 220-320deg C cut we had same problem like you. Problem is solved by adding some quantity of kero to SR gas oil, to get light fractions.
You can try to increase severity for some time on reactor to get more light fractions, if in that case you don’t have good stripping you should thing about column inspection.

Regards,

 

[STI666]

Jan,

If you are now thinking that the trays may be fouled or damaged, a Gamma Scan of the tower may help to resolve this issue.

Scanning Technologies works in several countries around the world. Visit their website at

http://www.scanningtech.com

for more information on this service.

 

[sshep]

There is no data above to indicate that you are putting enough heat into the tower to actually produce reflux. Failed tower internals theories will NEVER explain the absence of reflux- should someone like to try to explain the symptoms by some mechanism of failed internals then I will be very amused to hear it.

The similar advice of CJKruger to actually calculate the reboiler duty (rather than outlet temp) should be followed- now do it! If a temperature pinch at the bottom is limiting the heat input, then lower the tower pressure and/or add a recycle of some lighter condensible that will reduce bottoms temperature as suggested by Milutin. You don't speak much about the reboiler operation or control tray temp increases which is suspicious. NOTE: If you are bypassing live steam around the reboiler trap to get the apparent flow up to design (or some other nonsense I often see), then stop fooling yourself and line the trap up to measure what is actually condensing.

If you put heat in, then you must take it out. The vapor distillate cannot be removing the reflux equivalent of design reboiler heat input because the distillate is too light (and top temp too low) by the analysis you have given. If you check the air temperature leaving the condenser it can't be very hot (no condensing->no heat removal->delta-T on airside will be low). Estimate the condenser duty using an average face velocity, face area, and airside delta-T.

I troubleshoot these type of tower problems often and can tell you authoritatively: this is a mystery which can be easily solved in the field without calling for a scanning consultant. A tower scan will only show no liquid levels on the trays above the feed- something you already know. A general tower scan will not help you solve the energy balance mystery.

Please- how much duty is being put in at the reboiler and does it match what is removed at the condenser? If you can answer this questions then you are on the path to solving the problem.

best wishes,
sshep