Crude Overhead Reflux (cold) Exchangers

[jayhawkrider]

Has anyone here ever installed crude overhead reflux exchangers with the crude on the shell side and vapor on the tube side? (These are the exchangers downstream of the reflux drum and before the fin fan cooling). If so, what were the benefits and problems of configuring the services this way??

We have had severe corrosion problems with existing twisted tube HEXs with vapor on the shell side. We have a proposed design to swap the services.

 

[EmmanuelTop]

Changing fluid allocation will not solve the problem, since it is a corrosion issue and not a fouling one (dirty fluid/low velocity, for example). Furthermore, if you keep the existing heat exchanger geometry, you will encounter much higher pressure drop on the tubeside, causing lower operating pressure in the O/H receiver (if you want to keep flash zone pressore unchanged, that also means higher heat load of the downstream fin-fans = more flaring), or if you want to keep the O/H receiver pressure constant, it will result in loosing distillates into atmospheric residue, as a consequence of higher flash zone operating pressure. Maximum allowable pressure drop of the overhead vapors will define appropriate heat exchanger design - it is quite expectable to face inadequate heat exchange area when shifting fluid allocation and staying within allowable pressure drop limits - so this proposal will also have its economic drawbacks (i.e., a completely new heat exchangers).

Proper crude oil desalting process is 90% of battling corrosion problems in the overhead system. Make sure that your desalter runs with minimum 80% efficiency (if single stage) or 97-99% efficiency (two-stage). Ensure the absolute control of chemical dozing (neutralizer, corrosion inhibitor) and check overhead pH more frequently, 2-4 times per shift at least. Your problem in the heat exchangers is "first droplet condensation issue", and is originating from high HCL content of the column overhead vapors. Changing heat exchanger fluid allocation and/or design will simply not solve the problem.

 

[jayhawkrider]

We have an opportunity to remove the twisted tube HEXs and go with another design all together. Wash system is on the horizon, but not this year. Continuing neutralizer injection, operating above dew point, rerouting some streams from the overhead, and desalters perform well.

Don't believe swapping services will correct the corrosion, but we are hoping to improve our corrosion rates and not buy bundles annually and plug tubes semi-annually. U-tubes are much less expensive to replace than the twisted tube design, as well. Trying to gather info about whether to expect better or worse corrosion issues by swapping services and what the drawbacks could be. New swapped service design addresses pressure drop (maintains the same as twisted tube). The alternative with straight tube was far too large at 72"D!

 

[25362]

Have you considered changing metallurgy ?

 

[jayhawkrider]

Yes, but test tubes of Duplex didn't show the promise we had hoped. Trying to avoid the expense of titanium with the prospect of water wash on the horizon.

We talked to PCS this morning, and they said they know of 4 crude units with the services swapped. 3 of the 4 have water wash systems, and the 4th is in S. America. Your "first droplet condensation" comment was also mentioned as an issue with the horizontal design and eventual tube corrosion (companies with this design metallurgy up). Sounds like a no-go for us until we install a wash system.

 

[EmmanuelTop]

Is this conventional two-stage overhead system?
Using inappropriate neutralizing chemical makes the situation worse, because of amine salt accumulation problem in the reflux loop. Have you analized O/H receiver condensate for NH3, Fe and Cl? What is the most frequent pH value? What is the average salt content in desalted crude and what test methos do you use? Any caustic addition downstream of desalter?
Also, be aware of LMTD reduction after installing waterwash system, which means you must make a compensation with additional heat exchange area.

 

[jayhawkrider]

Yes, 2 drums. Yes, caustic injection - currently 10 be with project to inject 5 be caustic soon. Have neutralizer and filmer injection on hot and cold bundles - the rates vary with test results and corrosion rates seen on the probes. Not sure what testing we do, I'm the mechanical on the team. Aware of the LMTD issues - biggest reason for not installing the water wash system thus far. Running some crudes with higher salts of late - working with tank farm to blend and reduce content.

 

[EmmanuelTop]

Do not inject neutralizing amine and corrosion inhibitor upstream of the 1st stage condenser. It results in amine salts accumulation in the reflux loop (column/HEXs). Are there any evidences of under-deposit corrosion?

Please check other parameters I asked about in my previous post.

 

[jayhawkrider]

Yes, and we've had that debate with our chemical contractor. We've talked to 3 different contractors, and 2 of 3 included neutralizer in their treatment program. No one can seem to distinguish between corrosion caused by under deposit corrosion from the process itself and amine salt deposition from too much amine. Theories abound. We have a large operating and testing matrix in progress to figure out what affects the corrosion rates.

 

[hanon]

Maybe you want to have a look to this paper "Four steps solve crude-tower overhead corrosion problems" by Lieberman

It quite useful for your case

Regards,
Hanon

 

[EdStainless]

What is your existing metallurgy? I suppose that your test was on 2205?
What are the temps on this unit? Have you thought about a superferritic tube material?

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm