Ditching the Dirt

Choosing the right tank washer

1 of 3 < 1 | 2 | 3 View on one page




Once you decide to change from manual tank washing to an automated process, you will find a mind-boggling array of tank-washing nozzles and machines on the market. They are available from more than a dozen manufacturers, and in countless shapes, sizes and configurations. Some work better than others, and some cost more than others. However, price is not always an indicator of value.

Engineers often find that very little information is available to help make the selection. Each supplier has its own sales literature, but only direct experience will tell you that a model from manufacturer X is better than one from manufacturer Y. Even experiences shared among colleagues are suspect if operating conditions are not similar.

Testing performance

Given the scarcity of objective information, one company decided to conduct a testing program to make direct controlled tests of tank washers under similar conditions and generate directly comparable information.* Although it would have been possible to analyze performance based on nozzle impact measurements or other lab approaches, actual residue or "soil" removal efforts were deemed more realistic and consistent with actual applications.

The testing began with identification of a representative tank and selection of a test soil that could be applied in a consistent manner. The soil selection was critical because the material needed to be difficult to remove, with characteristics that would make the evaluation quantifiable. All aspects of the test were designed to ensure repeatable results.

Although tests of competitors' units were considered, the company ultimately made the selection from its own product line. The intent of the experiment was primarily to test various design configurations instead of to measure one manufacturer against another. All the units tested are standard catalog designs taken right off the shelf. See Fig. 1.

Figure 1. Test Tank Sample


The testing program used a poly tank measuring 95 in. in diameter by 89 in. in height.

For the tests, coupons were designed and mounted in the walls of a 2,500-gallon (gal.) tank. The soil was the dried Elmer's Glue-All product, which is a water-soluble polyvinyl acetate (PVA) resin. The company formulated a method that called for running the nozzle at a specific pressure for 10 minutes using room temperature water, and then evaluating the material removal from the coupons. After running 30 tests on 15 different selected tank-washing nozzles and reading 180 coupons, the company felt it had enough data to draw some basic conclusions about the nozzles ," and about tank washing in general.

The test results did not point out a specific type of tank washer as a "universal best" for all applications. However, the results did provide more information about cleaning process and practices.

Understanding the cleaning task

Tank-washing nozzle selection must begin with an understanding of the specific cleaning process. Look at the task in front of you and ask:

What am I washing out? What is the soil composition?

How does the soil actually come off the walls of the tank?

What is the interaction of the soil and cleaning solution?

What do I expect the liquid to do when it hits the walls?

How far is the target surface from the nozzle, vertically and horizontally?

These are key questions that will help you analyze a cleaning task. Once you understand what you are dealing with, you can choose a nozzle with appropriate performance attributes. You need to keep in mind three major variables: impact, flow rate and distance.


How much impact is required to loosen and remove the soil? If the soil comes off easily and rinses away, your nozzle choice will not be difficult. Just about any tank washer will work.

On the other hand, more aggressive soils (such as dried glue) come off faster with the help of the spray impingement. Impact helps erode the soil and can lift it off the surface.

In the previously mentioned test situation, the best-performing nozzles were those that generated spray impact, even limited amounts, and delivered it frequently. When tested head to head, a small rotating tank washer outperformed a static spray ball, even though the rotating nozzle used less than one-third the liquid volume of the spray ball. The constant sweeping action combined with direct spray impact vastly improved the process.

So, how do you evaluate a given design and estimate its impact characteristics? When comparing two units, how do you determine which is the better choice?

The answers typically are a function of flow for a given orifice. In other words, you should be looking for a spray head with the fewest number of nozzles for the highest flow rate you can tolerate. However, this scenario brings with it some interesting trade-offs, including:

Time vs. liquid consumption. Which is more important, liquid use or time? A large-capacity tank washer typically will clean faster than a similar unit with a smaller flow. However, the efficiency (amount of cleaning per water volume) often decreases with the larger unit. A smaller tank washer running for a longer time often can do the same cleaning job using less liquid than a larger one running for a shorter time.

Pressure vs. impact. Increased pressure can mean decreased impact. With tank washers, especially small free-spinning units (<50 gallons per minute [gpm]), the optimal pressure can be as low as 20 pounds per square inch (psi) to 30 psi. A higher pressure will not increase impact. It might even decrease impact while using more liquid.

Rotation vs. impact. Speed hurts. Free spinning reaction-driven units can rotate too fast. This does not help the droplet propagation. Controlled rotation designs help minimize this effect, but the difference depends on the nature of the process.

1 of 3 < 1 | 2 | 3 View on one page
Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments