02.understanding Process Equipment For Operators & Engineers-53

294 UNDERSTANDING PROCESS EQUIPMENT FOR OPERATORS AND ENGINEERS
Glycol
pots
Off gas
flow
Liquid Water+
naphtha
Seal
legs
Orifice
plate
Off
gas
Water
(no naphtha)
FIG. 36.3 Poor interface level control degraded the off-gas flow meter performance.
Jose said, “Muy malo, señor Norm. It’s always like that. It makes no sense.”
“But Jose,” I asked, “What is the naphtha content of the water from the seal drum?”
“Nada!” (None).
“But Jose,” I asked, “What is the rate of naphtha production?”
“Nada. We have no draw-off for the naphtha.”
“But Jose,” I asked, “where then is the naphtha going that drains into the seal drum with
the water from the seal legs?”
“Ah Norman. It must be flowing out with the off-gas. This increases the density and DP
of the gas passing through the flow meter. That’s why our gas meter reads so erratic and so
high all the time. It’s not a flow instrument problem at all. It’s that we have no indication of
the interface level between the naphtha and the water phases in the seal drum. And, no
way to drain the naphtha liquid phase out of the drum. It’s a process design error. Do
things like this happen often in America as well?”
“Sure. How do you think I made enough money to buy a house with seven bathrooms
and a swimming pool?”
No orifice flow meter can tolerate a variable vapor-liquid flow and produce a reason-
able output. It’s not the velocity that varies, but the density that is too variable.
Chapter 36 • Problems in Measuring Process Variables 295
Orifices (contrary to Fig. 36.3) are 99% of the time located in a horizontal run of piping.
The orifice taps should be located above the orifice flanges, to retard fouling and water
accumulation in the glycol filled lines. Not underneath the orifice flanges—like on many
process units.
Samples
My friend Steve has finally earned his law degree. He’s now practicing law in Beaumont
and gotten out of engineering. Here are three of Steve’s mistakes made when obtaining
samples:
• Mistake #1—Steve needed a crude unit naphtha sample to design a new crude naphtha
debutanizer. The crude tower reflux drum operated at 10 psig and 130°F. He obtained
the sample himself in a quart glass bottle. By the time he delivered the sample to the
lab, most of the ethane and propane in the naphtha had evaporated. His computer
simulation indicated he required a debutanizer pressure of 80 psig. In reality, he needed
about 130 psig to condense the light ends. Steve should have gotten a bomb sample,
with the bomb completely filled with liquid.
• Mistake #2—Steve needed a wet gas sample from the same drum to size a new
centrifugal compressor. He obtained the sample in a bomb. By the time he delivered
the sample to the lab, most of the pentanes and hexanes had condensed. He computed
the design molecular weight for his new compressor at 38 MW. Had he included the
condensed naphtha components, the molecular weight would have been 52. As a
result, the motor driver on the new compressor was undersized and tripped out on high
amps. Steve should have had the sample heated in the lab prior to passing it through
the gas chromatograph for analysis.
In both cases, my friend should have checked that his samples were at their
calculated bubble point (naphtha) or calculated dew point (wet gas).
• Mistake #3—Steve asked an operator to obtain a sample of vacuum tower residue at
450°F in a sample can. The operator dropped the can and the resid auto-ignited. The
auto-ignition temperature of hot tar is 320°F, or even lower. We’re not allowed to get
samples above their auto-ignition temperature in open containers. Incidentally, the
auto-ignition temperature of gasoline is typically 450°F.
Steve will make a wonderful lawyer. He will probably get rich, especially with his process
engineering background. I’m sure many of my clients could profit from his legal services.
He’s already billing at a higher hourly rate than I do. He will likely purchase a home with
eight bathrooms and buy his wife a Mercedes.
37
Cleaning Process Equipment
There are five types of process equipment that are cleaned on a routine basis:
1. Distillation towers: both packed and trayed.
2. Heat exchangers: shell and tube sides.
3. Fired heaters: tube side—coke removal.
4. Air coolers: exterior finned tubes (this also applies to the convective tube bank of fired
heaters).
5. Centrifugal compressor rotor wheels.
I plan to discuss both on-line and off-line cleaning techniques based on my experience,
without reference to the large amount of published claims from a myriad of vendors, who
offer chemical cleaning services.
Packed Distillation Columns

Towers that are packed with beds of dumped type packing (rings or saddles) are difficult to
clean, based on my observations after restreaming. For example, I had an 80 –600 ID scrub-
ber, packed with 300 aluminum Pall rings. During a turnaround, I decided to wash off the
water soluble sulfate salts, as the scrubber had developed a tendency to flood in the past
year. I had the 2400 top manway removed and the top of the 25 ft high bed washed down
with a 300 fire hose.
When we started back-up, scrubbing efficiency was intolerably poor. We could not absorb
the SO3 vapors with the weak sulfuric acid circulating solution. A terrible plume of sulfuric
acid escaped into the humid air over Texas City. Even my wife called from home to complain
about the dense white acid plume. A shutdown was the only possible response.
This time, after water washing from “The Top-Down,” I entered the tower to inspect
the rings. The packing was quite clean on the top few feet of the bed. But, as we excavated
the packing, removing the rings in 5-gal buckets from the tower, I could see the problem.
Channeling! That became progressively worse as we dug down into the 25 ft deep bed.
Obviously, the vapor flow had followed the path of least resistance and flowed through the
areas of clean rings and compromised the vapor-liquid contacting efficiency.
I had each bucket of rings dumped onto the ground and carefully water washed. After
we had a pile of very clean rings, they were reloaded back in the tower via buckets. But,
all this handling had deformed many of the thin gauge aluminum rings. This reduced the
open area of the rings. So, when we started back-up, the scrubber flooded, because
the liquid phase could not flow easily through the crushed aluminum packing.
Understanding Process Equipment for Operators and Engineers. https://doi.org/10.1016/B978-0-12-816161-6.00037-0 297

© 2019 Elsevier Inc. All rights reserved.
298 UNDERSTANDING PROCESS EQUIPMENT FOR OPERATORS AND ENGINEERS
Washing Packed Towers From the Bottom-Up

In 1985 I was working in the natural gas production fields south of Laredo, Texas. My prob-
lem one day was a 30 –000 ID glycol dehydration absorber tower was flooding. Natural gas is
saturated with moisture. It has to be chemically dried with glycol, a desiccant, to a few
ppm of water, before being transported via pipeline. The absorber was packed with
200 raschig rings. From experience, I had learned that such towers flood due to acid-soluble
corrosion products such as iron sulfide. So, I decided to order a truck, equipped with
a circulation pump, to meet me out in the desert at 9:00 p.m. It was July, and daytime tem-
peratures at the well site had been 110–115°F. Best to do the job in the cool of the evening.
The acid I ordered was 10% inhibited hydrochloric acid (HCl). Currently (2019), I would
use either Citric Acid followed by a neutralizer or a “ZYME” solution, as HCl is very
corrosive.
The truck of HCl was waiting for me at the well site when I drove up. The driver was
already hooking up his acid circulation lines to circulate top to bottom. Remembering
my fiasco in Texas City a decade ago, I stopped him.
“Amigo. Hook up the return hose to the top 300 nozzle. Connect the fresh acid at the
400 bottom nozzle. We’ve already washed the tower out with clean water. Comprendo?
We’re going to circulate from the bottom to the top until midnight.”
“But Señor, we’ll have to fill the entire 30 ft. of the tower with our acid. Maybe my truck
not have enough acid? Maybe it go empty?”
“Maybe you’re right. Okay. Let’s order another truck of acid, pronto.”
That’s the problem with circulating from the bottom-up, rather than from the top-
down. The entire packed bed must be filled with the cleaning solution. It’s expensive,
but it sure worked that night in Laredo. Except, I was frightened by a three foot long
rattlesnake curled around the base of the absorber.
To summarize, the smaller the diameter of the tower, the more practical it is to circulate
from the bottom-up. For a 46 ft ID 110 ft T-T vacuum tower in Saudi Arabia, this is not all
that practical.
Trayed Towers
For towers equipped with ancient “Bubble Cap” trays, chemical cleaning is not possible.
Each cap has to be unbolted, and the space between the chimney and the cap cleaned
manually. I have tried to clean such trays by skipping this step with very negative results.
That is, the tower flooded on start-up.
For perforated tray decks such as:
• Sieve
• Valve
• Grid
• Duplex
• Tabs
Chapter 37 • Cleaning Process Equipment 299
Circulating the cleaning solution from the top-down, may have mixed results, depending
on the tower diameter, on the circulation rate, and the number, size, and distribution
pattern of the holes. Other important variables are the number of trays and the length
of the circulation time.
My experience with cleaning trays by circulation from the top-down is mixed. Usually,
but not always, the tower internal inspection shows reasonably clean trays. But, on other
occasions, the results are quite disappointing.
Cleaning Heat Exchangers Tube Side

To clean the tube side of a heat exchanger, assuming the heating fluid is on the shell side,
I proceed as follows (see Fig. 37.1):
• Step 1—Close tube side inlet valve.
• Step 2—Open the tube side bypass valve.
• Step 3—Open the small flushing fluid valve.
• Step 4—Wait 10–30 min.
• Step 5—Reverse steps 1–3.
The tube side fouling deposits will likely spall off due to the thermal expansion of the
tubes. Or, dissolve due to the flushing fluid. Or, melt off. Or, be forced out of the tubes
due to the sudden reintroduction of the tube side flow.
I used this technique which I called “On-line Spalling,” in 1969 on the world’s largest
crude unit in Whiting, Indiana, with excellent results.
For off-line cleaning of the tube side of an exchanger, tubes are simply hydroblasted.
This is simple for straight tubes, but can be tricky for a “U” tube bundle.
Flush
FIG. 37.1 Shell and tube heat exchanger with “U” tube bundle.

02.understanding Process Equipment For Operators & Engineers-53

Uploaded by

Copyright:

Available Formats

02.understanding Process Equipment For Operators & Engineers-53

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

02.understanding Process Equipment For Operators & Engineers-53

Uploaded by

Copyright:

Available Formats

294 UNDERSTANDING PROCESS EQUIPMENT FOR OPERATORS AND ENGINEERS

Packed Distillation Columns

Understanding Process Equipment for Operators and Engineers. https://doi.org/10.1016/B978-0-12-816161-6.00037-0 297

Washing Packed Towers From the Bottom-Up

Cleaning Heat Exchangers Tube Side

You might also like