A Database for Tracking Information about Ultrasonic Transducer

Selecting a suitable ultrasonic transducer for a given application is by itself a complex issue. Additionally, selection criterion also depends on performance data and availability from a rather large collection of transducers from a company's inventory. In this month's "NDT Solution," the author has described a practical solution to manage a company's transducer collection. The principles involved in developing a database for managing transducers are presented and readers should be able to develop a similar database according to their needs. G.P. Singh Associate Technical Editor

 

Introduction
As a company's ultrasonic nondestructive testing (NDT) needs change over time, ultrasonic transducers are specified, acquired, used for a while, and ultimately retired from service. At a given time, a company may have a rather large ultrasonic transducer collection to manage.

This management task is difficult for the unaided employee. Every transducer has several important features associated with it, which pose a challenge for anyone's memory. Selecting a transducer for an application frequently also involves calculating certain variables associated with it, which poses a mathematical challenge.

With today's technology, it is fairly straightforward to put together a computer database to help manage a company's transducer collection. Such a database can help manage lots of transducers, and track many important features for each transducer. The database application can do any calculations necessary to help select a transducer.

This paper gives an overview of how a reader might build a transducer database. Some details of the hardware and software used are also mentioned, but the general principles presented should guide the reader toward implementing a similar database using, perhaps, quite different hardware and software.

---

The database should track information that identifies any transducer in a collection.

---

 

Objectives
The primary objective of a database should be to help specify ultrasonic transducers for applications, and to help search the transducer collection for transducers satisfying these specifications.

The secondary objective should be to "manage" the collection of ultrasonic transducers. For instance, one could track how long transducers typically last, and then order replacements for instruments that are predicted to fail soon.

 

Tracking Identity Information
The database should track information that identifies any transducer in a collection. If a collection has transducers of only a few types, this may be straightforward; having a broad range of transducers, however, makes it more difficult to implement a universal identification scheme. As much as possible, the identification scheme should be based on information explicitly stated on the transducer itself or on its permanently attached tag. This identity information should be organized hierarchically, from most general to most specific.

Immersion or contact should be the first (broadest) category. Even a completely unmarked transducer can be categorized this way using visual clues, such as whether the transducer's connector is watertight, and whether there are threads or other attachment mechanisms for shoes. (As a point of interest, the transducer collection for which the author built a database did contain a large contact transducer with no markings on either its case or permanently attached tag.) Both immersion and contact transducers can include many brands of transducer.

Brand (usually the manufacturer, but not always) should be the second broadest category. Brand information is almost always available. Place whatever brand is explicitly visible on the transducer (or its tag) into the database. Each brand can include many series/types of transducers.

(Note for the next two categories: The words series, type, model, and style are used inconsistently by different manufacturers, sometimes even within a single manufacturer's line.)

Series/type should be the third broadest category. Many manufacturers use one of these two words for transducers sharing general attributes. On some transducers, the series/type is explicitly marked, such as "Alpha." On other transducers, the series/type can only be implicitly inferred, from the first few characters of the model/style. One firm, for instance, has a line of "Videoscan" transducers, but they are not marked with this word; for the database, these should be entered as the "V*" series/type, where the asterisk denotes the trailing characters in an explicitly marked model/style such as "V324." Each series/type can include many models/styles of transducers.

Model/style should be the fourth broadest category. Many manufacturers use one of these two words for transducers sharing very specific attributes. Each model/style can include transducers with many serial numbers.

Serial number should be the fifth broadest category. This information is almost always available.

 

Tracking Application Information
The database should track information used to specify transducers for inspection applications. This includes the number of piezoelectric elements, the nominal element geometry information, and the nominal performance information. Application information includes the number of piezoelectric elements. Most applications require transducers with only one element; less common are applications that require transducers with two or more elements. Each of the multiple elements may differ in size and shape, so the database should hold separate information for each element.

Application information also includes nominal element geometry information. Shapes and sizes of piezoelectric elements help to determine ultrasonic beam characteristics. Three common element shapes are circular, semi circular, and rectangular; (the semi circular case arises in dual element transducers, where the two semi circular elements are placed in a way that has an overall circular shape). The nominal element dimensions to track are diameter for circular shapes, and both length and width for rectangular and semicircular shapes. Transducer dimensions may be in US Customary System (English System) units such as inches, or in the International System (SI) units, such as centimeters. Dimensions should be stored using one or both of these systems.

Application information also includes nominal performance information. ("Nominal" performance information may vary from actual performance information, as "characterized" using some particular ultrasonic setup.) This should include some characteristics typically marked on the transducer or its tag by the manufacturer: the frequency in megahertz (MHz), and the focus (foci) in water if any element is focused. The database should explicitly track whether each element is focused or unfocused. For a circular element, there is only one focus; otherwise, there are two foci: one for the element's length, and one for its width. These comments on units of element dimensions apply also to focus dimensions.

Some nominal performance information useful for application purposes is not typically found on the transducer or its tag, but can be calculated; (Krautkrämer and Krautkrämer, 1990). These include nominal nearfield(s), nominal -6 dB spot size(s), and nominal -6 dB focal zone(s). (To calculate spot size and zone for an unfocused transducer, use its nearfield as the focus.) For a circular element, there is only one of each of these information types; otherwise, there are two: one for the element's length, and one for its width. (The above comments on the units of measurement (English/SI) for element dimensions apply also to the dimensions used for this type of information.)

Database queries should allow searching the transducer collection by application criteria. These criteria should include: immersion or contact; frequency range; focused or unfocused; focus/nearfield range; and spot size range.

 

Tracking Dates
The database should track transducer acquisition date and retirement date. Tracking retirement dates avoids using the database to specify a transducer for an application, then looking for it in the collection and finding it has been retired from active service. Tracking acquisition dates improves the ability to manage a collection of transducers; tracking how long transducers typically last allows anticipatory ordering of transducers that will be needed in the near future.

A distinction should be made between "deleting" a transducer record, and "retiring" a transducer record. Deleting a transducer record means erasing that record from the database, which should be done only when a record contains a lot of mistakes and needs to be re-entered from scratch. Retiring a transducer record simply means marking one field in the record to show that the transducer to which it refers has been retired from active service. Distinguishing between deleting and retiring transducer records avoids recalling retired transducers as active candidates for needed inspections, while keeping information about retired transducers for reference purposes.

 

A Solution
Using the principles described in this article, the author created a database application. It was written in a database oriented, fourth generation programming language that has specialized commands for manipulating data, fields, records, and files.

From the database software's menu, the user can: recall previously stored transducer records; store a new transducer data record; delete an incorrect transducer record; and retire an old record once its transducer is retired from active service. Each menu selection prompts the user for the information that the procedure needs. Then, the procedure makes the appropriate manipulation of the database for the user.

As much as possible, every prompt issued by a procedure is accompanied by a detailed "reminder," consisting of examples which indicate to the user exactly what type of information is needed. For example, the user is not just prompted to "enter brand:" instead, the user is reminded of all the different brands in the database, then prompted to enter the brand for the transducer in question.

To accommodate the large amount of information about transducers with rectangular and semi circular element shapes, the database software makes use of a 132 column wide display mode, in addition to the commoner 80 column wide display mode. For the same reason, the database software uses landscape mode printing.

The author's database has been used many times. Existing transducers suitable for new applications have been found in the collection. New transducers have been acquired for the collection and old transducers have been retired.

 

Discussion
Once a transducer's record has been recalled by querying the database, the transducer must then be found in the physical collection. Each transducer should be stored in a small box labeled with its database identification information. The boxed transducers in turn should be stored in commercially available plastic drawer systems (or something similar) that can hold the physical collection in an organized fashion. One possibility is to organize the collection in the same hierarchical fashion as the database's identity information. (As a practical matter, model/style, while useful for identification, can be dispensed with for organization. So, the physical collection can be hierarchically organized by immersion or contact, brand, series/type, and
serial number.) Another possibility is labeling new transducers added to the data-base with sequential numbers (transducers 1, 2, 3,...), and putting them in the next available drawer positions (drawer positions 1, 2, 3,...); if this approach is used, the sequential number for each transducer should be tracked as a field in the database.

Only nominal, not characterized, performance information is recommended for tracking. A transducer's characterized performance information is only valid for some specific ultrasonic setup, that is, a certain set of instruments with certain settings, together with environmental conditions. Characterizing a transducer using different equipment or with different settings, or using a different environment (for example, tap water instead of distilled), will probably yield different characterized performance information. For characterized performance information to be meaningful, the database would have to store information about the equipment and settings used, as well as environmental data. This could be done, but it would complicate the database tremendously; instead of a "transducer database," it would be a "complete ultrasonic setup database." Such complexity is beyond the scope of this article.

The database described in this article does not capture all transducer information of interest. For instance, the previous paragraph implies that a transducer selected for an application on the basis of "nominal performance" may have a different "actual performance;" whether its actual performance is suitable for the intended application must therefore then be determined by characterizing it, or by simply trying it (this has nothing to do with it having been selected from a collection by a database; this same caution would apply to a newly purchased transducer selected from a catalog). As another example, consider inspecting the inside of a tube; a transducer selected for this application using only database information may turn out to be too large to fit inside the tube. So, the database described here is an excellent starting point for specifying transducers for applications, but it is not exhaustive.

 

Conclusions
In this paper we have presented the use of database technology to solve the problem of managing information about ultrasonic transducers. The recommended types of transducer information to be tracked are: identity information; application information; and acquisition date and retirement date information. If a company has special transducer tracking needs, the reader should be able to customize a database for those needs by simply adding a few fields, or a few canned queries.

 

Reference
Krautkrämer, J., and H. Krautkrämer, Ultrasonic Testing of Materials, pp. 59; 73-79; 81-83; Springer-Verlag, Berlin/Heidelberg, 1990.

 

* The Timken Company, RES-09, PO Box 6930, Canton, OH 44706-0930, (330) 471-2309; fax (330) 471-2282; e-mail: hartzogh@timken.com.

Copyright © 1998  by the American Society for Nondestructive Testing, Inc. All rights reserved.

[ Materials Evaluation ]