Printer Friendly

Sound advice for nodularity testing.

Inside This Story:

* Ductile iron gets its strength and ductility from carbon, which forms graphite nodules rather than flakes in the metal.

* Ultrasonic velocity testing, which measures velocity by sending and receiving sound waves as they travel through a section of a casting, is a proven method for checking the nodularity of ductile irons.

When carbon is added to steel to form cast iron, it can precipitate in several different forms. In ductile iron, it forms as spheroidal graphite, called nodules, which gives ductile iron increased strength and ductility. The percentage of graphite in the form of nodules in the matrix and their size and distribution are two important factors that affect the material properties of ductile iron, so checking the nodularity in ductile iron castings is an important step in ensuring quality parts to a customer.

When ductile iron is produced, a small coupon is poured at the end of each heat to verify that the graphite has precipitated in nodular form rather than in vermicular or flake form. This normally is done by cutting and polishing the coupon and looking under a microscope in unetched condition and comparing that to a similar structure in a wall chart. When that structure meets the minimum nodularity specification, the heat is released for further processing and the castings produced from that heat are deemed good.

If a metalcasting facility is producing parts for safety-critical applications, such as suspension components, then the customer requires the facility to verify that each of the castings meets the minimum nodularity standard. This type of inspection must be done non-destructively. To achieve this requirement, most facilities use ultrasonic techniques. In this method, nodularity is checked using an ultrasonic velocity measuring instrument, which measures velocity by sending and receiving sound waves as they travel through a section of a casting. The shape, size and amount of graphite present affect the velocity of sound passing through ductile iron. When the time taken for the sound waves to travel through the section, which is measured electronically, and the thickness of the casting section are known, velocity can be calculated and nodularity determined based on previously established relationships. Usually a minimum velocity number is set for a minimum nodularity. Castings with higher velocities than the minimum are accepted as good and stamped to show proof of testing.

In production metalcasting facilities, two kinds of ultrasonic velocity measurement-the contact method and the immersion method--are widely used but not widely understood. The contact method is used for low-volume and in-the-field testing, and the immersion method is used for high-volume testing of production castings.

Pulse-echo

In the contact method (also known as the pulse-echo method), an ultrasonic probe (transducer) is pressed against the casting surface with a couplant to maintain a good contact between the casting and the transducer, which sends and receives sound signals. Sound waves travel from the surface of the casting to the opposite side of the casting wall and are reflected back to the transducer. The time it takes for the transducer to send the pulse signal into the casting and have the pulse signal reflected back is measured by the instrument. This time corresponds to travel through twice the thickness of the casting.

The velocity values derived from the contact method for measurement depend on the thickness measurement, which is measured separately from velocity and inputted into the measuring device for calculation. In the as-cast condition, the thickness measurement is based on the peaks on the surface, which will be greater than the average thickness the sound waves will record. Hence, the velocity measured on a rough casting will be lower than that on a machined surface when all other conditions are the same.

Referring to Fig. 1, the time taken for the sound waves to travel back and forth in the thickness of the casting ('t') is shown as 'T'. Velocity then is calculated as two times the thickness of the casting divided by time, or 2t/T.

[FIGURE 1 OMITTED]

If the thickness 't' is 1 in. (2.54 cm) and the time taken 'T' is 8.9 microsec., then the velocity is calculated to be 2(1)/8.9 = 0.224719 in./microsec. (0.570786 cm/microsec.).

The transducer transmits the sound wave pulse recordings to a monitor, which shows the peaks of the pulses. The strength of the reflected pulse, depicted by the height of the peak on the monitor, depends on several factors, including couplant viscosity, pressure applied on the transducer, roughness of the surface of the casting, the parallelism of the casting surfaces, porosity and material properties. The time measured, and thus the velocity calculated, will be affected by the amplitude level at which time is measured.

The enlarged version of the original pulse and the reflected pulse is shown in Fig. 2, which illustrates the importance of choosing an amplitude level for measuring the time. When the reflected wave amplitude changes, and if the reflected pulse crosses the measuring line from one step to the other, time measurement will change. For this reason, a reference standard of similar material and surface may be necessary to calibrate the measuring instrument.

[FIGURE 2 OMITTED]

The reflection of sound waves can be represented, as shown in Fig. 2, for machined versus as-cast surfaces. The pair on the right represents machined surfaces and the pair on the left represents as-cast surfaces with shorter peaks because reflected energy with as-cast parts is lower. In this case, if the instrument is set up to measure at level 1, two different velocities for the same material would be recorded, but if the instrument is set up to measure at level 2, then velocities would be the same in both cases. To avoid problems with this type of measurement it is better to set the measuring level lower, as shown as level 2 in Fig. 2.

Pitch and Catch

The principles used in evaluating the velocity using the immersion method (also known as pitch and catch) are the same as the contact method. However, in the immersion method, two transducers are used: one for sending the signal and the other to receive the signal. These two are mounted facing each other in a rigid fixture with their surfaces parallel. The fixture is immersed in a liquid couplant, generally water, mixed with a rust inhibitor. This arrangement is shown in Fig. 3.

[FIGURE 3 OMITTED]

Inside the tank is an enclosed measuring device with holes that allow the fluid couplant to enter the chamber between the transducer and a reflecting surface. The distance between the transducer surface and the reflecting plate is fixed, generally at 2 in. (5.08 cm). This device constantly measures the velocity of the couplant. The velocity of sound in the fluid couplant can change as temperature and density changes, so checking the velocity of the fluid every time the velocity of a casting is tested is important. To keep the density of the fluid couplant consistent, filters are used to remove suspended particles in the fluid. Controlling the temperature of the fluid to minimize variations in testing also is important.

As shown in Fig. 4, several variables are involved in the immersion method: the thickness of the section where velocity is measured (t); the distance between the sender and receiver (d); the velocity of the fluid as measured by the device; the time for the sound waves to travel through the fluid between the transducers ([T.sub.1]); the time for the waves to travel through the fluid and once through the casting section ([T.sub.2]); and the time for the waves to travel through the fluid and three times through the casting section ([T.sub.3]). Using the known variables, the computer built into the measuring device solves for the thickness of the casting section and the velocity of sound waves through the casting section to check the nodularity.

[FIGURE 4 OMITTED]

Examining Nodularity

Ductile iron is identified by differing nodularity percentages that tell how much of the graphite in the matrix is in the form of nodules. In ductile iron castings, a range of velocities can be found for the same nodularity percent. This is due to metallurgical reasons as well as measurement variations in both nodularity and velocity. Optical methods of evaluating nodularity by comparison charts or the point count method are very subjective and depend on the skill of the operator in polishing the sample and interpretation of the microstructure. Also used to rate the nodularity is the image analyzer technique, which uses different criteria to define a nodule and non-nodule.

There is no accepted standard method for determining nodularity. By plotting the nodularity measured by any one of the methods mentioned above versus velocity measured, there is a good correlation. From this plot, a cutoff point in velocity can be established at the point where the minimum nodularity accepted by the customer intersects with the right side of the range of velocity (Fig. 5). With this method, some acceptable nodularity castings may be rejected by velocity testing.

[FIGURE 5 OMITTED]

Castings are usually tested in the as-cast condition. If the castings are heat treated, velocity decreases for the same nodularity. The decrease will depend on the temperature of heat treatment. During sub-critical annealing, velocity may decrease by about 1,000 in./sec. (2,540 cm/sec.). If the castings are heated above transformation temperature (1,450F [787.77C]), the velocity decreases further for the same nodularity. One reason for the decrease is the formation of secondary graphite in the grain boundary areas during high-temperature heat treatment. Repeated heat treatment decreases the velocity even more.

Factors of Variability

Many metallurgical factors can affect the velocity of the sound waves during ultrasonic testing. The shape of the graphite nodules can result in lower velocities. Other factors that can decrease velocity are surface flake graphite, large nodule size and a higher volume percentage of graphite. Testers also should keep in mind that matrix variations (ferrite, pearlite, martensite and carbide) can affect the velocity of the sound waves, as can extreme variations in alloy content (C, Si etc.).

It should be noted that a combination of metallurgical factors may result in no change in velocity. Looking at Fig. 7, one would expect the velocity to be lower due to some vermicular graphite and some large nodules. But Fig. 8, which is etched, reveals carbides in the microstructure. Combinations of vermicular graphite and large nodules lowering the velocity on one side and carbides increasing the velocity on the other side can result in velocity readings that are comparable to typical good nodularity castings, thus causing what otherwise would be a reject casting to be accepted.

[FIGURES 7-8 OMITTED]

Because carbides in ductile iron castings tend to increase velocity, castings with excessive carbides can be discerned by using an upper limit for acceptable carbide levels. To screen castings for carbides, immersion units could be set up with two limits in production: a lower limit to assure a minimum nodularity and an upper limit for ensuring the castings do not contain carbides above the maximum allowable limit. Documenting the distribution of velocities from day to day and analyzing the distribution for possible shifts in the process variables could lead to improving the consistency of ductile iron castings.

For More Information

"Nondestructive Evaluation of Structure and Properties of Ductile Irons," Y.S. Lemer and A.P. Vorobiev, AFS Transactions (98-12).

Efficient Immersion Velocity Testing

Following are tips and guidelines to help make your immersion velocity testing efficient and correct:

* Test location on the casting should have two parallel surfaces.

* Section at the test location should be at least 0.5 in. (1.27 cm) thick (preferred).

* Surface of the casting should be completely wetted by couplant. There should be no air pockets.

* Castings should be cool enough to touch. If they are hot, water vapor may form on the surface, which will affect the velocity and will result in a faulty test. Fluid in the tank and the casting should be of the same temperature to avoid problems.

* Higher temperatures also reduce velocity.

* Fixtures to hold the casting during testing should be rigid and the location of the casting in the fixture repeatable.

* Tested and accepted parts should be stamped with a mark to verify the testing.

* Never place a casting from the floor into a basket of accepted castings without testing again.

* Make sure testing is done on the critical area of the castings. Verify the casting print for location of testing.

* Castings with "no test" should be tested again.

* Castings with low velocity after a second test should raise a warning flag, and it should be investigated ASAP.

* Velocity testing is not a substitute for upstream control.

Al Alagarsamy is corporate technical director at Citation Corp., Birmingham, Ala.
COPYRIGHT 2005 American Foundry Society, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Alagarsamy, Al
Publication:Modern Casting
Geographic Code:1USA
Date:Jul 1, 2005
Words:2133
Previous Article:Practical tips on gating iron castings.
Next Article:Moving away from silica sand.
Topics:


Related Articles
Ductile iron nodule count affected by molten mental processing.
Microstructure anomalies affect ductile iron properties.
Iron foundry benefits from new nodularizing process.
'How we began pouring ductile iron ... and did it right the first time.' (Blackhawk Foundry and Machine Co.)
High Si-Mo ductile iron: views from users and producers.
Advancing ductile iron production.
'Running Out' the Ground Balls.
Mechanical Properties in Thin-Wall Ductile Iron Castings.
Using Eddy current testing to identify casting defects.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters