Refining industry uncertainties

2009 has kicked off with the market in panic over tremendous uncertainty in the future.  Oil prices have slumped, we are seeing some short term increases in pricing today, but market ‘experts’ are still split on what is going to happen in the future.

In terms of the FCC world, many refiners are running lower throughputs, almost all are rigorously investigating dramatically altered yields from their norm if they have not already changed, and many high cost (many with previously expected high rates of return) projects are on the chopping block as uncertainty and pessimism dominates thinking.

I am presenting at the upcoming Shaw FCC Forum in Abu Dhabi in early April, and I think it will be interesting to see if the market uncertainty and current financial performance are impacting attendance.  I will also be at the National Petrochemical & Refiners Association Annual Meeting in San Antonio in ~ 10 days.

As I watch this as a partially informed observer, I cannot help but think about the role the media and the numerous financial experts have in the overarching economic situation. 

I have to wonder about the credibility of ‘experts’.  I am in an industry where it is notoriously easy to be an expert – I have repeatedly said that “Often, the only requirement to be considered an expert is that you flew to the meeting.”

I have stepped back from the direct feed of ‘facts’ from the marketplace on a daily basis, and have instead spent this time doing some reading.  I highly recommend Outliers by Malcolm Gladwell.  This is Gladwell’s third book (the other two are very worthwhile reading as well, Blink and The Tipping Point), and it explores what is behind personal statistical anomalies, both on the positive and negative sides of typical distributions (think Bill Gates, the Beatles, how a co-pilot’s culture can lead to plane crashes) .  It explores the impact of culture, family support, and what it takes to be a real expert. 

It has some interesting ideas for all of us in this refining industry.  A key idea in Outliers is the amount of time someone has to spend in their field to really be an expert – the number is 10,000 hours, but understand what this really is…it’s 10,000 hours practicing the actual skill set needed to be an expert.  The hours spent sitting in meetings talking about budget or doing employee reviews do not count towards being a process expert.

I also highly recommend The Drunkard’s Walk:  How Randomness Rules Our Lives by Leonard Mlodinow.   This book explores some of the science of statistics, their impact on our lives, how poorly we as human beings perform in recognizing true odds, and in one section, lays out how many financial ‘experts’ are the product of randomness.  The book reads very quickly, and the final chapter is extremely interesting.  In the final chapter, Mlodinow charts the relative performance of 800 fund managers  from 1 to 800 based on return minus median rate for a 5 year time period 1991 – 1995.  A positive number therefore indicates above average market performance.  Mlodinow then charts how the same managers did 1996-2000 and shows them in their prior ranking spot.  The best performer for this second 5-year period was # 360 in the first 5-year period.  The #2 performer in the second 5-year period was a tie between #s 700, 10, and 50. My conclusions do not entirely agree with Mlodinow’s in this section, but it’s clear that at least 1/3rd of the top 50 performers did worse than the market average.  This 1/3rd were in the top 10 percentile for the first 5 year look, and they tumbled to be no better than the 26-50th percentile.  

So, as we march through this year, I am cautiously optimistic but know enough to not be absolutely sure what the refining industry is going to see.  At this point, my reading tells me that only time will tell what is going to actually transpire.  My advice is that every time you see an ‘expert’ opinion on the marketplace, and they back their expertise up with what they predicted last time, there is no guarantee of them being correct now.

At this point in time, it’s your turn, the reader, to respond – what do you think will happen this year for your industry?

FCC shutdown survey results and more…

It’s been an interesting year so far, one that I misdiagnosed at the start.  I have been far busier than I expected, which is a good thing, but it also brings along an entire series of unintended consequences.  The response to the survey about FCC downtime has been great, and I have been waiting for responses to slow down, but they are still coming in.
In terms of upcoming events, Astron will be hosting an FCC Health Check workshop at the NPRA FCC Symposium being held during August 2008 in Houston.  The exact dates of the NPRA FCC program are August 19th and 20th.

If you have never attended the NPRA FCC program, it is held every other year and is very much worth attending – the focus is on both reliability/maintenance and process topics.  Attendance is typically in the 500 – 700 person range, and is generally split evenly between mechanical oriented versus process oriented people.  Feedback about the meeting has been overwhelmingly positive.  As of today, it is accepted as being the best industry wide FCC event, and serves as a great place to meet other refiners and talk to many experts in one place.   There will be a Q&A panel on the first day, followed by workshops on the second day.

I encourage everyone that is free to attend this meeting make the trip to Houston.  In recent NPRA programs, there have been numerous international companies represented, some from as close as Mexico, others from as far away as Korea, India, Australia, and Saudi Arabia. 

The results from the first survey are available.  They largely confirmed most of the expectations most people in the industry have, but were still interesting.  Information from many FCCU outages came in and have been tabulated.  Total results are shown in figure 1; approximately 70 % of outages are due to the combined reasons of major rotating equipment (air blower, wet gas compressor, and expander), cyclone failures, catalyst circulation upset, or utility failures.  While not surprising, the survey results indicate that only about 10 % of FCCU outages take place on a scheduled basis.

Figure 1

Major rotating equipment made up almost 40 % of the total and was the single most common reason for FCCUs shutting down early.  Figure 2 shows the breakdown of major rotating equipment outages.  Air blower outages make up about half of the total for major rotating equipment.

Figure 2

The sample size was respectable, representing hundreds of FCCU outages going back over the last 6 years.  Over the next couple of weeks, some geographic sorting will be done.

Also, if readers have not sent in survey results, go ahead and the survey regarding FCCU outages will be updated periodically as additional information comes in, probably 3 more times this year.

One of the things I think I have noticed in the refining industry in the last several years is that the people accountable for performing routine FCC technical service at refining locations are being stressed more and more by their employers to do new activities in addition to the historical focus on FCC optimization/troubleshooting. 

The refining industry has changed in many ways over the last several years, mostly for the better, particularly in the areas of being a good neighbor in local communities.  In most cases, I think it’s been the front line operations, engineering, and maintenance staff that have shouldered the majority of this worthwhile work.

Fortunately, the industry has been able to lean on the increased computing power of the IT world to free up some of our time so we can devote ourselves more fully to these other initiatives, but the downside is that we have to keep finding the best ways to turn our reams of data into useful information. 

Please take a moment and send your responses in to the below second survey – we will let this survey run for 40 days and post the compiled responses.  And, again, if you have not responded to the first survey, go ahead and send that response in as well.

Survey #2: Breakdown of your FCC job duties


Causes of your last 3 FCC shutdowns?

I have been working in the refining area almost 18 years.  It is an industry that has had several ups and downs in this time period, and I’m hopeful that I am into the second half of my career.  In January, 2008, I joined Astron International, Inc, as Vice President of the newly created Refining Consulting Division.

I have willingly headed down the path of being, in the words of a friend of mine, Johnny One-Trick, an FCC specialist first and foremost, with a smattering of general knowledge in enough other areas to get me by with solid understanding.  I started my career in the Marathon organization, working several FCC units, with heavy tech service and operations experience.  I then went to work an MW Kellogg (now KBR) in the FCC design and service area.  In 2001, I went to Engelhard Corporation and eventually became the Marketing & Technical Service Manager for their FCC team.

I will be showing up frequently on the Refining Online website, with one of my regular areas being this blog.  I’ve decided that initially, my blog entries will be running a poll question for 3 – 4 weeks, then sharing the compiled results with the refining online community.  I’m confident that dialogue on this site will help the entire industry and provide some clarity on issues facing us today.

At some point this year, I will also have several blogs dealing with some hard core technical nuts and bolts, but those are currently in development and probably will not be showing up until May or June 2008.

The first question for 2008 is a tangential repeat from a presentation that UOP did in the 2006 NPRA Principles & Practices FCC workshop.  UOP did a survey of several customers asking about FCC shutdowns.  Here is the follow-up question:

What were the causes of your last 3 FCC shutdowns?

– Scheduled turnaround
– Wet gas compressor failure
– Air blower failure
– Expander failure
– Other rotating equipment failure
– Utility failure (steam, power, instrument air, cooling water)
– Catalyst circulation failure
– Cyclone failure
– Main column/gas plant upset
– Human error
– Logistics (out of feed, catalyst, product storage)
– Other (please describe)

Please fill a quick survey by clicking the below link and let me know your answer.

Survey: Causes of your last 3 FCC shutdowns


How FCC Feed Hydrotreating Affects FCC Yields And Economics

How FCC Feed Hydrotreating Affects FCC Yields And Economics

The following paragraphs show an example of the effects of hydrotreating ANS VGO on a typical FCC operation.  The focus of the discussion is to compare FCC feed hydrotreating with FCC product treating as competing options for making low sulfur gasoline and diesel fuels.

Table 1 presents the properties of virgin ANS VGO and compares them with the properties of the VGO after severe hydrotreating (95% desulfurization).  Hydrotreating increases the API gravity of the VGO by about 5.5 numbers, increases the aniline point by 10°F, lowers the nitrogen level by about 50% and lowers the VABP by 24°F.  The VABP reduction is the result of a small amount of hydrocracking that occurs in the hydrotreater.

Table 1: Impact Of Hydrotreating on ANS VGO Properties

Table 1

Table 2 summarizes the effects of hydrotreating the feed on the FCC operation.  The FCC unit was air rate limited.  The results show that FCC feed hydrotreating increases conversion by over 10 volume percent while decreasing the yield of dry gas and substantially lowering delta coke. Because of the improvement in delta coke, the riser temperature could be increased and the feed temperature could be decreased slightly without increasing the coke yield or the air rate.  These adjustments helped to increase the gasoline octane (R+M)/2 by 0.65 numbers versus the base case.  The FCC gasoline sulfur was reduced to 20 ppm so that it could be successfully blended directly into a gasoline pool that requires a maximum of 30 ppm sulfur.

TABLE 2: Hydrotreating Effects on FCC Performance with ANS VGO 

Table 2

Since the LCO sulfur from the hydrotreated feed operation is substantially lower than the base case LCO sulfur, the cost of making ultra low sulfur diesel will be lower.  Also, FCC feed hydrotreating greatly reduces the level of SOx from the FCC regenerator, eliminating the need for the use of SOx reduction agents or the operation of a flue gas scrubber.

With all of the advantages for FCC feed hydrotreating, it might appear obvious that this would be the best choice to make low sulfur products.  However, due to the lower capital cost of product treating, many refiners have chosen this option.  Nevertheless, current economics have made FCC feed hydrotreating more attractive relative to product treating. This is shown in Table 3, which compares a product treating case with FCC feed hydrotreating for ANS VGO.  Using the yields from Table 2, the feed treating case shows a $5.44/bbl advantage in FCC product value. 

TABLE 3: Economics of Hydrotreating ANS VGO

Table 3

BPD moderate pressure FCC feed hydrotreater, would be about $270 MM, leading to a payback period of about 1.97 years.  The conclusion is that, in the long run, refiners will profit from installing FCC feed hydrotreaters.  For refiners who have not already been required to install product treaters and flue gas scrubbers (for SOx reduction), the relative attractiveness of feed hydrotreating is even greater.   For more discussion on these and other related topics, the reader may be interested in the following upcoming technical seminars from Refining Process Services:

Fluid Catalytic Cracking Process Technology   Sept. 17, 18, 19
FCC Unit Troubleshooting     Sept. 20, 21
Hydrotreating & Hydrocracking Process Technology  Oct. 3, 4, 5

These programs (along with 11 others on various refining topics) will be offered at the Crowne Plaza Houston North Greenspoint Hotel. More information is available at

Posted By: Robert J. Campagna, Refining Process Services, Inc.

SWS Pumparound Startup

The issue is starting up a Sour Water Stripper (SWS) with a pumparound system.  One concern was the pumparound pump suction going dry before sufficient level was maintained.
Summary responses:

On a recent SWS start up, condensate was injected into the pumparound pump suction.  A jumpover also permitted recycling bottoms to feed for warmup prior to startup.

The pumparound water will become the worst quality water in the system (high NH3 and H2S). Condensate, boiler feed water, stripped water or SWS feed could be used to the suction of the pumparound pump to fill the pumparound section up.  Water with any hardness (service, fire or hard process water) should not be used.

We filled our 500-gpm SWS via a 2” service water line to the pumparound pump suction. Once there was a high level in the bottom of the tower, we recirculated bottoms-to-feed (with the tower vented to atmosphere) and gradually ramped up the steam to the reboiler.  Once the tower pressure was up to ~ 5 psig we started the pumparound pump, increasing steam as necessary to maintain positive pressure.

Establishing pumparound circulation before the tower is venting steam could be a problem. Usually the chimney tray has weep holes, in which case the pumparound pump may lose suction without supplemental water makeup. A welded chimney tray should be considered as  gaskets tend to wash out in spots, requiring supplemental water injection to the pumparound section.

Consider locating the overhead TI in the top of the tower, rather than in the offgas line, where heat from the steam tracing will give a false high indication in the temporary absence of offgas flow during startup.

A TC bypass around the pumparound cooler allows the operator to maximize the pumparound rate and water temperature and minimize H2S concentration and corrosivity. 

On a recent SWS start up, a jumper from the feed line to the pumparound return line was used to inventory the pumparound system.

Duke Tunnell
for the ABPG

TGT SS Quench Column

The issue is the value of a 316 SS quench column and 316 cladding of the quench water coolers with caustic addition on automatic pH control.

Summary comments:

Recent new TGUs have CS quench and absorber shells.  Quench coolers are CS shell and 316 SS tubes, with quench water on the shell side.  Corrosion allowance in the quench tower is 0.25”.

HIC plate CS with a 0.25″ corrosion allowance for the quench towers, with 304L SS tubes in the quench water cooler.

Trend seems to be more requests for SS cladding of the quench column (and alloying up in general).  One factor might be reduced cost of cladding due to improved automated fabrication techniques. 

Experience with a 20-year-old unit is that CS is sufficient.  Corrosion may be due to chronic SO2 breakthroughs.  The key to avoiding breakthroughs is reliable H2 measurement. 

Quench column design with the tail gas entering downward at a 45° angle should achieve some initial cooling by impingement on the liquid level, particularly if water recirculation is lost. It should also avoid impingement on the far vessel wall and reduce localized corrosion at the opposite side.  The problem might be greater with packing, where droplets raining down are more prone to being slammed against the wall.
With a standard SCOT design (i.e., a properly sized reactor, H2 make-up available if indirect preheat, etc), CS should be adequate.  Failure to control SO2 breakthrough into the quench column can result in sulfur deposits on the walls and under-deposit corrosion, including H2 blistering.

With caustic addition for pH control, low chloride caustic should be ordered – usually a more costly grade – to avoid chloride cracking of the stainless.

Duke Tunnell
for the ABPG

Hello Refining Online Users!

Another first from Refining Online since it started over 12 years ago! A blog for the refining industry called ROL Blog.

Please review the articles on ROL Blog and feel free to post comments on the subject matter. Comments will be reviewed and posted upon acceptance.