Developments in spray-nozzle technology

What are the principal factors which affect nozzle performance? How can a nozzle be designed to provide optimum performance? What's new in nozzle design? Beesh Zytynski has the answers

Anyone looking for an unsung hero could do worse than the industrial spray nozzle, writes Beesh Zytynski, Managing Director of Bete Ltd.

Because, although they may not be a particularly hot topic of conversation at fashionable dinner parties, they are used in a vast array of applications, from fire protection on oil rigs to spraying relish into mixing vats in the food industry.

Indeed, in many such applications, their role is critical.

Nowhere is this more true than in the pollution control industry, where they often play a key part in processes varying from flue-gas desulphurisation to the control of fly ash discharge.

And here, as elsewhere, not any old nozzle will do.

On the contrary, it is increasingly necessary to design a device specifically for the application concerned - a process which can demand techniques beyond the capabilities of most nozzle manufacturers.

Why? What are the principal factors which affect nozzle performance? How can a nozzle be designed to provide optimum performance? What's new in nozzle design? Here we look at these questions, with emphasis on areas where there have been significant developments.

Typically, choosing a nozzle can require the specification of factors such as flow-rate versus pressure characteristics, spray angle, material of construction and the piping which feeds the nozzle.

For many applications this is a straightforward matter, resulting in a perfectly adequate off-the-shelf device.

In others, though this is far from the case.

In fact, it is frequently the case that there exists no ideal standard product, and a custom design is necessary.

It is here that problems can begin.

Three of the most important characteristics of any spray are the amount of liquid it contains, how that liquid volume is distributed within the spray envelope, and the sizes of the droplets that make up the spray.

Measuring and interpreting the flow rate and pressure characteristics of most nozzles is a relatively simple process, but this is frequently not true for pattern and droplet size data.

And, in many applications, droplet size is critical.

For example, many processes - gas scrubbing, for example - depend on exposing the maximum possible liquid surface area to a gas stream.

Others require that the droplets be as large as possible, such as when a spray must project into a fast moving gas stream.

For such applications, good nozzle design will depend on an accurate technique for measuring droplet size.

And this is not easy.

In fact, droplet size analysis can be downright confusing.

Years ago, the accepted technique was to collect droplets on glass slides, but the process was slow and problems such as droplet splatter raised doubts about its accuracy.

Today, more advanced techniques are available, such as those based on laser light signals.

While this approach works adequately for some applications, it also has significant limitations - principally the fact that it does not cope well with the large or non-spherical droplets often found in industrial environments.

This is because laser-based instruments use a signal processing algorithm which presupposes that the signal to be analysed comes from a spherical particle.

Any signal which deviates from this form is rejected.

However, a more recent alternative - a direct video imaging system pioneered by Bete - does not have this drawback, and offers many significant advantages.

Offering very high accuracy for all droplets between 2?? and 32,000?? moving at high speed, the technique uses advanced CCD (charge coupled device) technology and high-speed xenon strobes, together with dedicated image processing hardware.

The basic principle of the process is to illuminate the droplets with a strobe light and capture the image on a CCD camera so that it can be easily digitised.

The droplets are lit from behind and appear black on the screen.

Image processing software counts the number of pixels in each image and converts this area to a physical size using a calibration ratio that is established using an image of an object of known size such as a ball bearing.

Today's computers can analyze more than 105 drops in just a few minutes and compute the various statistics describing the droplet size spectrum in real time.

The Bete imaging system is particularly good at analysing dense sprays that contain non-spherical droplets which are typical of many large nozzles used in pollution control systems.

In many sprays the droplet size is not the same everywhere, and imaging systems can also report information about the droplet morphology.

This can be a difficult and inaccurate process with other techniques, as most instruments use various options, like interchangeable lenses and filters, in order to fine-tune it for droplets in various size ranges.

If, for example, an instrument is set up to detect only droplets smaller than 2000um, and larger droplets are present in the spray, the reported data will be incorrect.

Often the operator needs to try several setups and compare the data to be sure that the instrument is reporting accurate data.

It is also important to select sampling locations corresponding to the regions of interest.

Bete selects locations based on a pattern test of the nozzle, usually choosing locations where the spray density is at its peak and locations where the density is one-half the peak.

This gives six locations per nozzle axis for a hollow cone nozzle.

Finally, it is important to consider the most useful definition of "droplet size".

For many applications the Sauter mean diameter (SMD) is of greatest interest.

This figure is defined as the diameter of the sphere whose surface area to volume ratio is the same as that of the entire spray, and is a very useful parameter for characterising sprays used in gas-liquid contact systems, being the preferred method for process calculations.

However, in some processes other measures of droplet size might be more useful.

In high-temperature gas cooling, for example, we might be more concerned about the maximum droplet size because, if complete evaporation is necessary, this will influence the size of the ductwork more than the Sauter mean.

In many applications, spray pattern distribution is also important.

This describes the location and spray density of the liquid emitted from a nozzle.

Again, this is not easily achieved and has, historically, relied on a direct collection method.

In fact, this technique is still widely used.

Usually this consists of a row of tubes placed under the nozzle, and water is allowed to collect in the tubes for a known length of time.

Measurement of the elapsed time and amount of water collected allows spray pattern calculation.

It is an approach which has a number of limitations.

Now, though, Bete has introduced a unique digital video system for accurately determining the volumetric distribution of liquid emitted from a nozzle.

This uses digitised information to calculate spray density and spray angle, and is ideally suited to current nozzle development and assessment programmes.

Consistently and accurately selecting appropriate sampling positions is extremely important when performing dropsize analysis.

The challenge lies in sampling the spray in such a way that the number and location of the individual tests chosen present a reasonable representation of the entire spray.

Recognising this, Bete has developed a number of unique sampling protocols for droplet size analysis which ensure that the reported drop size distributions most accurately reflect the overall spray performance.

This allows a high degree of repeatability and confidence.

Also available from the Bete system are measures of the uniformity of the pattern.

This figure, called the flux deviation is half the sum of the difference in the height of the water in each tube and the average height.

A nozzle with a large flux deviation would not be a good choice for an packing spray nozzle, for example, where even distribution of the scrubbing liquor over the surface of the packing is important.

Of interest in interpreting pattern data are the conditions of the test including the diameter of the tubes or other collection devices, the height of the nozzle above the apparatus and the nozzle operating conditions during the test.

Wider tubes, for example tend to mask variations in the spray uniformity, and the test height can greatly influence the reported angle.

All these techniques have already been proved in the field.

For example, despite many attempted solutions, one customer consistently failed to meet the specified SO2 removal rate in the absorption tower of a wet limestone FGD system.

However, sophisticated testing in Bete's laboratory showed that a spiral nozzle, made of Cobalt alloy 6, would decrease the Sauter Mean Diameter by 40%.

Designed and delivered in 8 weeks, the product quickly had the desired effect.

In another case, a utility operating a power plant was faced with low ESP efficiency, requiring the boiler output to be restricted in warm weather.

The solution was to increase the flue gas humidity by spraying finely atomised water into the flue gas duct - though no unevaporated water could be allowed to enter the ESP.

This meant that the largest droplet from the nozzle was 100? in diameter.

Bete's test facility was the only one capable of evaluating the competing designs within the time and accuracy limits required.

In fact, in most cases, duplicating the operating environment in the spray lab is impossible.

When, for example, the nozzle is to be used in a high-temperature or pressure environment or sprayed in a high-velocity gas stream.

In such instances, Bete has developed advanced computer modeling techniques and simulation software.

This has been used predict spray behavior in HF mitigation systems and to specify nozzles and layouts on off-shore drilling platforms.

Other applications include predicting spray drift from cooling ponds and dust-suppression systems and estimating evaporation rates from disposal ponds.

These, then, are some of the most recent advances in nozzle design.

Together, they offer customised spray-nozzle solutions to some of the most vexing problems facing industry today.

Back to top