May 23, 2013

Flue Gas Analysis as a Diagnostic Tool in Large Furnaces

Hi everyone. This is Doug Simmers talking again about improving your combustion processes.  The primary goal of combustion flue gas analysis is optimizing the fuel/air ratio, which minimizes fuel cost, lowers NOx emissions, and also minimizes the amount of carbon dioxide greenhouse gases emitted into the atmosphere. One overlooked use of flue gas analyzers is for balancing the combustion across large multi-burner furnaces.

Historically, a single flue gas analyzer placed downstream or in the smokestack was considered satisfactory for setting fuel/air ratios. Many furnace operators have moved the measurement location upstream, closer to the furnace, in order to get a faster speed of response, and added more analyzers for redundancy. This has proven to be a mixed blessing, since the analyzers almost never read the same O2 levels, which lowers the operators’ confidence in the measurement, and subsequently their willingness to operate at the lowest possible O2 setpoint. After many calibrations, it becomes clear that all the analyzers are all reading correctly, and the problem is that the combustion inside the furnace is unbalanced. When you think about it, this makes sense.

Four O2 analyzers placed zone by zone at the top of  the radiant section of a process heater furnace.

Four O2 analyzers placed zone by zone at the top of the radiant section of a process heater furnace.

The combustion process is the burner. This is where the fuel and air are mixed and combusted. The furnace is mostly just a heat exchange envelope. If you have 50 burners in a large furnace, you have 50 discrete combustion processes, and some variability can be expected. This is the case for all large multi-burner furnaces, be they a 500 MW coal-fired boiler, a large crude process heater furnace in a refinery, or a steel reheat furnace.   Strategically placing an array of analyzers in the flue gas ductwork can help operators keep the furnace balanced, and prevent many problems.

Flue Gas Stratification Tells a Story
Forward-thinking operators will use these varying O2 indications as a diagnostic tool to look for problems in the furnace, such as:

  • Fuel Gas BTU variation – will be indicated by both O2 and CO variation.
  • Fouled burners – depending on the consistency and quality of fuel burned in a furnace, a burner will change over time.
  • Flame carryover – combustion sometimes progresses well into and beyond the convective zone, and is indicated by lower O2 readings, higher CO readings, or both, along with rising outlet temperatures.
  • Tube leaks – an unbalanced furnace can cause flame impingement and associated tube wastage.
  • Furnace casing leaks – will result in elevated O2 readings, since 21% of the air leaking in is oxygen.
  • Sticking dampers – stack outlet dampers are sometimes poorly maintained, and sluggish control of O2, COe, and furnace draft are indicative of problems.

Combustion flue gas analyzers have become true analytical tools for determination of all kinds of furnace problems. We’ll talk about others in future blogs.

For more information on this subject, please visit the following links:

May 9, 2013

High Performance pH Measurement Saves Time and Money for Kyokuto Petroleum Industries

Measuring pH in liquids with many dissolved solids is a constant challenge for all types of industrial plants. For Kyokuto Petroleum Industries (KPI) Japan, continuous failure of their pH measurement in their desulfurization scrubber was costing significant time and money.

Image1The KPI scrubber is a magnesium hydroxide system. The pH measurement is installed after neutralization and the wastewater includes many magnesium sulfate solids. The flow chamber was sometimes clogged with solids.

KPI had been using a pH sensor with a water jet cleaning system. The cleaning system, however, was ineffective against solids that coated and plugged the glass and liquid junction of the sensor. After a few days of operation KPI could no longer trust the pH values being received as the readings would fluctuate significantly. Even when the sensor was cleaned manually, a time-consuming and expensive process, the pH sensor had to be replaced every month.

As a result, KPI often had to control the dosing of magnesium hydroxide manually based upon the experience and intuition of the operators. In addition to the cost, personnel time and extreme inconvenience, the plant was faced with significant local regulations regarding emission of SOx. If the manual control led to low magnesium oxide dosing, a SOx emission could occur with subsequent fines. If the dosing was too high, the costs of the expensive magnesium hydroxide soared and increased costs.

On top of all this, the pH sensor KPI had been using was connected to an analog device which made it impossible to get sensor status from the instrument automatically. The company could not predict sensor failure or receive other data without manual intervention.

With good reason, KPI was looking for a method of applying automatic control to its set points based on reliable pH measurement.

KPI switched their pH analysis to the Rosemount Analytical Model 1056 using the PERpH-X 3300HT-10-30 with SMART pre-amp as a sensor. The PERpH-X sensors are designed with an enhanced double junction reference that is specific to extreme applications. Reference flow into the process stream is controlled using a porous Teflon® junction that can be replaced in the event of fouling or plugging. The specially designed junction is chemically resistant and has a large surface area to maintain a steady reference signal in dirty or oily applications. This reference also resists poisoning which can occur with other sensors as a result of the diluted hydrochloric acid used to manually clean the sensor during preventive maintenance.

The 3300 has been able to operate accurately for two to three weeks before manual cleaning and the sensors are able to be used more than nine months in the process as KPI also takes advantage of the rebuildable reference feature of the 3300. The 3300 has lasted nine times longer than the previously used sensor – a dramatic improvement saving time and money.

The HART-enabled pH measurement by the 3300 allows KPI to monitor other valuable information such as reference impedance, glass impedance, temperature, etc. By monitoring these sensor parameters, KPI could detect a sudden large deposit of the reference impedance. Figure 1 shows the increasing reference impedance trend as well as the result of a sudden deposit. This information is used by KPI to improve the timing of maintenance. Image2They are also able to use temperature data to determine clogging of the flow chamber which saves both maintenance costs and prevents shutdowns. Using the Emerson 1056P-HT with a 3300HT sensor with the SMART pre-amp and jet sprayer, KPI has been able to control the dosing volume of magnesium hydroxide automatically saving time and improving efficiency. While savings in actual product costs measures in the thousands of dollars, the value of the accurate measurement is incalculable.