6 replies to this topic

[dcs]

Hi ! We have NHT Unit which is now it can’t operate to 100% design capacity due to it’s naphtha feed lighter than design (the IBP is quite lower than design : 39 vs 54 oC) . This unit prepare/treat and split naphtha to be light naphtha which will goes to Penex Unit and the heavy will goes to Platforming Unit. In NHT Unit, Naphtha feed from storage tank goes to Oxygen Stripper first to strip off dissolved oxygen. Feed goes to Oxygen Stripper Receiver which has temperature 50 oC (the design is 37oC). Naphtha from storage is feeded to O/H Stripper Receiver. FG also is purged to the receiver to dilute the oxygen in vent gas which goes to flare. Liquid from O/H receiver is pumped back to Oxygen Stripper Top Column after passing thru Feed Bottom Exchanger. Top column Vapor will be condensed in Oxygen Stripper Condenser before going to the O/H Receiver.
Typical operating condition now is :
Top Column Temperature : 104 oC vs design 122 oC
Bottom Column Temperature : 148 vs design 165 oC
Top Column Pressure : 7.1 kg/cm2g
The problem with the lighter feed we are facing is we can’t increase the bottom temperature because the top column pressure will increase and often time the charge pump (it is the pump downstream of feed bottom HE or charge pump to transfer Oxygen Stripper bottom to NHT Reactor) will start cavitation even the temperature just downstream of the pump is lower than design. I believe, it is because of the light fraction is carried over with the Stripper Bottoms. And the biggest problem with that is we can’t operate the unit to the design capacity and also flaring will happen at Naphtha Stripper and Splitter at downstream of NHT Reactor.
We have plan to add 1 new pre-fractionation column to strip off the light fraction to achieve the design feed specification before coming to NHT Unit. Is it correct ? But, before going thru that plan, I want to evaluate the existing Oxygen Stripper to maximize stripping in the column and perhaps we just need to add a reboiler or condenser only. I made a simulation of the Oxygen Stripper but I’m not convinced. Does anybody has experience like that and how to solve it. Hopefully, my explanation doesn’t make you confused. Thank you.

Attached Files

•  PDF.doc   156KB   207 downloads

[Zauberberg]

DCS,

I do not know if this is conventional naphtha oxygen stripper design, but PFD looks very strange to me. Is there any pressure controller in the system (feed drum)?
The overhead product from your stripper has to be checked for composition. If there are any C4- in quantities of significance, you certainly must account for light component accumulation in the feed drum and subsequent tower flooding at design feed rates.

[Milutin]

What is your present O2 content in feed and product from stripper? Or in the other words what is reason for increasing bottom temperature, high O2 content in product?

Regards,

Milutin

[dcs]

Hi Zauberberg,

Yes, there is a pressure controller at O2 Stripper Receiver (31-V-101) by controlling the Stripped O2 and condensable gas to flare. I thought also there is light components accumulation in the tower when running with light naphtha feed but not in the receiver (or you called it the feed drum) because every time we tried to increase the bottom temp, by adding more steam to the reboiler, to strip more light fraction from bottom product, the top column pressure immediately increase, but the PCV to the flare doesn’t fully open. Primarily, I thought that the condenser reach its limit duty but it seems not like that by checking our simulation (even I don’t know the overhead vapor from top column to the condenser. There is no flow meter). It’s mean that probably the condenser is not a bottleneck. Is it possible that the piping system of naphtha feed going to O2 Stripper Receiver as a bottleneck because of its position tied in to the piping outlet the condenser by 90o. My follow up question : How to make sure the o/H vapor flow rate to the condenser without field measurement and also the bottom fluid going to the reboiler (there is no flow meter there). I checked flooding at the actual condition (lighter feed than design and rate below design because of the problem), flooding is still ok (by using Original Csb Sieve Tray Flooding Method, and almost flooding – 98% by using Minimum Csb Sieve Tray Flooding Method). Would you advise what is the different between this two Csb ? I submit again the PFD with control system and piping size in the feed and condenser system (PFD-1.doc).

Hi Milutin,

Present dissolved O2 in feed is 53 ~ 125 ppm and in stripper product is 0.25 ppm (I’m not sure about this lab analisys). Reason for increasing bottom temp : 1. is to make sure that only small amount of dissolved O2 in stripper product to avoid fouling of equipments, particularly the feed effluent exchanger. It’s suspicioused that there is a slip off O2 compounds to reactor because of reactor P is increasing quite fast. 2. is to remove light fraction because it will make flaring in Naphtha Stripper and Naphtha Splitter and this light fraction will load the Penex Unit (downstream of NHT Unit) to be running more than 100% capacity with light compounds that can’t be converted.

All comments and advice from all of you is apreciated.

Regards, dcs

Attached Files

•  PFD_1.doc   166.5KB   120 downloads

[Zauberberg]

Hello again DCS,

1. To start from the first point, you should never mix liquid and vapor streams in a T-pipe manner, without VLE nozzle at least. I am attaching one interesting article by Kister & Litchfield, page no.3, where it has been discussed in details about this issue (here, the hot vapor bypass was the problem). Pressure fluctuations are frequently caused by flashing of liquid stream, especially when there are non-condensables (O2) or saturated vapor in the system.

2. Is there a DP cell across the stripper, or some section of it? When raising reboiler duty, flood point can be identified by sudden increase of DP, in most cases (depending on specific service, a reasonable DP across one tray is in the range of 0.005-0.01bar). You said that column top pressure increases with reboiler duty, but PC valve to flare does not open. What do you mean exactly by this?

3. What happens with the column overhead and condenser outlet temperatures when you raise the reboiler duty?

4. The oxygen content analysys: is the feed sample taken from the "fresh" naphtha line, or from the receiver? Have you compared those two values?

5. Compare C5- content of fresh naphtha and receiver naphtha also.

[Milutin]

Design of unit is unusual to me, after you strip out O2 and some quantities of lights ends, overhead (reach in O2 and lights) is mixed with cold naphtha. I suppose that helps in dissolving some quantities of O2 and lights in feed again.
It seams that way how feed is introduced - actually extra cooling of overhead with feed - that unit is designed not to allow lights to escape from unit. Is it valve which introduce purge flow always open, what is pressure of purge gas?

It is good idea to find, from licensor, what is allowed level of O2 in feed to confirm yourself how you operate stripper from that point.

I suppose that you import feed (naphtha) because crude units usually have stabilizer column employed to remove C4 and lighter.
The simplest way is to ask naphtha producers to make proper stabilization of naphtha. Do you have composition of lights by GC in feed and product from that column. What is IBP of naphtha on unit exit?

To convert stripper to naphtha stabilizer you need about 40 trays, strippers usually have up to 10, what is your case? If you cannot make good separation you will waste C5/C6 fraction in overhead.

You should check NHT charge pump, is there enough NPSH when you running lighter feed than design.

Are your feed tanks for NHT blanketed with protective gas to prevent contact with oxygen?

Regards,

Milutin

[dcs]

Hi again Zauberberg,
1. Thanks a lot for the good article. In my case, the cool feed liquid mixed with fluid outlet condenser which is almost liquid with different T 34oC (it should be 16oC by design) and there is no pressure fluctuations but only pressure raising.
2. What I mean is : when top column pressure increase because of the condenser duty exeeded its design capacity, I think flaring should happens because receiver pressure will increase also and PCV will be fully open to maintain pressure at set pont.
The O2 content analysis is taken from fresh Naphtha line. We don’t take lab analysis for sample of naphtha from the receiver. I will find out for the rest and try to check it out.

Hi Milutin,
I think so as well. Valve to introduce purge flow is always open (small) with pressure around 8 kg/cm2g. We import feed naphtha and it’s quite difficult to ask naphtha as we require that’s why we want to make a flexibility in our plant. We blanket our naphtha tanks with N2. The lowest IBP of naphtha feed is 38 oC (ASTM D-86) and unfortunately we have no lights compositions by GC for both feed and product of the column.

Regards, DC