Detecting Bunker C Oil Mass Flow

Aug. 1, 2007
“Ask the Experts,” on ControlGlobal.com, is moderated by noted process control authority Béla Lipták. Béla and his cadre of leading experts in process automation, recruited from among the co-authors of the Instrument Engineer’s Handbook 4th Edition, are “in the box” all month answering process control questions from all comers. The best will be published here. In order to eliminate misunderstandings, please include P&ID sketches using ISA symbols with your questions.

Question: I have problem of selecting a suitable flowmeter for a heavy liquid, Bunker “C” Oil. Bunker “C” fuel oil is a sticky, black liquid similar in appearance and smell to asphalt sealing compounds. At 10 °C it has a consistency of liquid honey or corn syrup. At 0 °C it barely flows. It has the following specifications:

  • Specific gravity: 0.94 - 0.95
  • Asphalt content: 10%
  • Sulfur content: 3%-5%
  • Viscosity at 50 °C: 450 mm2/s
  • Viscosity at 100 °C: 100 mm2/s
  • Flow-rate: 80 Ton/Hour maximum
  • Pressure: 15 Bar
  • Operating temperature: 100 °C
  • Pipe Diameter: 4 in. (100mm)

What is the suitable flowmeter for this application (electromagnetic, ultrasonic, mechanical, etc.)?

Maged Attalla
Control & Instrumentation Engineer
Micom Co.
[email protected]

Answer: If volumetric measurement is good enough, use an oval gear or viscous helix PD meter with steam or electric tracing, in SS or Hastelloy C construction. If mass flow measurement is needed, use a 4-in., self-draining, heated, Hastelloy C Coriolis flowmeter. The main advantages of these meters include their high accuracy (0.1% of the actual flow rate) and excellent rangeability (up to 100:1). Their main disadvantages are high cost and that they can not be used on liquid services with significant gas content. Also be careful not to install them at high points or in a downward flows, because they must alwats be flooded. For costs, suppliers, capacities and other details see Chapters 2.11 and 2.19 in the 4th edition of the first volume of The Instrument Engineer’s Handbook.

Béla Lipták

Answer: The four types of flowmeter used to measure bunker oil are differential pressure, transit time ultrasonic, coanda effect, coriolis effect. It is relatively straightforward to get mass flow from any of these.

Differential pressure requires a flow element, either an orifice plate, or a venturi or a v-cone. The abrasive and corrosive nature of bunker oil makes this measurement (with temperature sensor for density compensation and thus mass flow) a maintenance issue. This type of meter is generally the least costly.

Coriolis meters are becoming less vibration-, abrasion- and corrosion-sensitive all the time. Sizes up to 12-in. diameter are becoming more common, with at least four vendors now supplying larger sized meters than 4-in. diameter. This is still the most expensive, but probably the most accurate option.

Coanda meters are quite accurate and can be compensated for density. There is, however, only one vendor at this time, a small company in Phoenix, Ariz,, Fluidic Flowmeters LLC.

Transit time ultrasonic flowmeters of the multipath variety are becoming more common for this application because they have fewer issues with maintenance arising from abrasion and corrosion resistance.

Walt Boyes
Editor in Chief
Control magazine
[email protected]

Answer: A screw-type positive displacement flowmeter is a good choice under these conditions; one which I have used for similar duty is found at www.maxmachinery.com/specsfm/243.pdf.

Your application is, however close to the maximum design condition for their 3-in. meter, and they don’t make a larger one.  KRAL AG (www.kral.at/flowmeters_omg_techdata+M5d637b1e38d.html) has a larger one of similar design.

Ian H. Gibson,
Principal Technical Specialist
Fluor, Australia

Answer: A Brooks oval gear meter BM type can tackle high viscosities. Otherwise, if the line can be kept warm, a Coriliolis meter with a heated sleeve would be best.

Stephan Gaertner, P.E.

Answer: I would suggest using a Coriolis because the high viscosity eliminates vortex and the non-conductive nature eliminates the magnetic flowmeter.

Due to the sulfur content, it is advisable to consider stainless steel, Hastelloy C or titanium materials of construction.

Attention must be paid to the pressure loss. We checked the application with a specific Endress+Hauser model, and the pressure drop was slightly below 1 bar with a 3-in. coriolis, and 0.3 bar with a 4-in. coriolis. The size should be decided as a function of pumping costs vs. first cost.

Finally, considering that the fluid has high viscosity at ambient temperature, the installation should be self draining to empty the meter at shut off, before the fluid becomes cold.

Carlos Behrends
Managing Director
Endress+Hauser Controle e Automação
São Paulo, Brazil

Answer: I think that a PD meter could be OK, well-heat traced.

Dr. Alberto Rohr
Consultant, Italy
[email protected]

Question: I am an instrument and control engineer, active in the field of oil & gas projects. I need the answers/clarification to the following question: Can we use the unit control panel of a turbo compressor as the ESD of the whole station? Does this meet any rules and regulations?


If Volumetric Flow is Good Enough

A heated oval gear meter can tackle high viscosities.

I would proceed as follows, asking these questions:Answer:

  1. What is the maximum required SIL of the ESD functions?
  2. What is the assured SIL of the unit control panel as a component?
  3. What is the architecture of the balance of the safety loop and what components are used?
  4. What SIL they do have?
  5. What is the resulting SIL of the whole safety loop when using the unit control panel as a logic solver? To answer this, you have to calculate the probability of failure on demand of the whole assembly. For SIL 1, the PFD should  be better than 1 on 10; for SIL 2, it should be better than 1 on 100; for SIL 3, it should be better than 1 on 1000. All these values should be maintained in the time; i.e., there shall be no downgrading of the PFD and test periods shall be defined. 

If the calculated SIL fits with the required SIL of the ESD functions, it is possible to use the unit control panel as the logic solver.
Also note the following:

A) It is necessary that the control system is fully independent from the ESD system, in order to have a different line of defense in case the ESD fails completely, for instance due to power feed or other common cause of failure.

B)According to the severity of required SIL, the components of the safety loops need to be selected properly and certain precautions taken, for instance, using self checking transmitters (double sensors with comparator), partial stroke test for the valves, etc.

Dr. Alberto Rohr
[email protected]

Answer: NO. The control system may fail in such a way as to cause a demand on the safety system. A safety instrumented function within the control system may likely be failed due to the same cause.  This represents very high risk.

Dr. William M. Goble C.F.S.E
Sellersville, PA 18960