3 replies to this topic

[georgiatech90]

I am in the process of starting up a small recycled lube oil plant. The process is as so:

 

1. Recycled lube oil feedstock has water removed in a oily water separator and preheated to 575F.

2. Feedstock enters an atmospheric column with steam injection, yielding Naphtha, diesel, and a bottoms stream.

3. The bottoms stream is preheated to 585F and enters a vacuum column under -26"Hg of vacuum, steam is also injected into the column, yielding LVGO (top draw) , HVGO (middle draw) , and asphalt residue (bottom).

 

It is a relatively standard atmospheric/vacuum oil distillation unit, and the atmospheric column is working well, yielding the diesel and naphtha as expected. The problem we are facing is that the LVGO level simply will not establish itself. The only time any significant LVGO level was observed in the unit was when we accidentally overflowed the HVGO draw tray slightly to about 200% of level (nowhere near bringing liquid to the LVGO tray). After shutting the vac column down and checking out the internals it was observed that there was damage and displacement to the HVGO tray. We had been running the column during the start-up phase with the bottom of the column flooded so level could be established at the HVGO draw. 

 

My initial thoughts are:

- Feed preheat temperature may be too low, causing the distilled fluid in the tower to condense before reaching the LVGO tray. We need to avoid cracking so are limited in how high a temperature we can raise the feed to. In your experiences what is the highest acceptable temperature you have seen lube oil preheated to in similar columns?

- LVGO tray does not have proper gaskets to hold liquid level. We poured water into the LVGO tray while the tower was shut down and saw that water slowly makes it's way through - but with rising vapor below it in operation in theory this should still hold some level. Does this need to be a water tight tray?

 

Any tips would be greatly appreciated.

 

 

[rdcrags]

Total trap-out trays should be liquid tight except for weep holes for shutdown and gas free procedures. The startup procedure should include steps for establishing levels on these trays and establishing circulation (pump-arounds) before generating vapor. Such levels are established with startup lines. Allowing a high level to bump the bottom trap-out tray is a no-no as doing so will always cause tray damage. Is LVGO boiling range material showing up in the HVGO product?

 

If LVGO material is passing un-refluxed through the LVGO section, it would greatly overload the entire educator system and vastly reduce the vacuum. If the vacuum is normal, the LVGO boiling range material must be in the HVGO product. That would mean the operating temperature is too low.

[georgiatech90]

Rdcrags,

 

Thank you for your input. We have repaired the damaged trap-out trays and they should now be liquid tight. 

 

This plant was designed by a junior level engineers years ago and frankly has a lot of flaws. There are no start-up lines on the plant to support level establishment. If I'm not mistaken, a start up line would be a line going from the bottoms of the column to feed liquid to the HVGO/LVGO trays? Would liquid level not eventually establish (slowly) by running the bottoms at standard level, where the evaporating HVGO/LVGO would slowly start to evaporate upwards and condense higher up in the column, eventually causing level to form?

 

It sounds like it would not be a viable solution to shut off the bottom and side draws, and then allow the column to fill completely with liquid, and then drain it down to standard level, thus establishing level on the LVGO/HVGO trays. Why would the trays be damaged by this? I calculate only about 1psi of pressure that would be exerted on the bottom side of the trays during filling.

 

Thanks again for your input.

[rdcrags]

Yes. After all, that's how the column was hydrotested in place after the trays were installed and manways secure. However, some vacuum column foundations are not designed to support a liquid full column. So, they are air tested, or nitrogen tested if the expense is not a concern.