The Art of Failure Analysis

Failure Analysis is not an art form, but rather a systematic evaluation of the results of a process designed to provide a specific product. This process may involve many methods, materials, products, parts, temperatures, heat rates, hold times and various types of equipment. Failure analysis becomes important when the desired result is not achieved and is the tool used to determine who, what, when, where, why and how. It is irrelevant who caused the problem but more importantly the important questions are:

    * What is the problem?
    * Where or when did the problem occur?
    * Why did the problem occur?
    * How do we correct the problem and not just the symptom?

Whenever a failure occurs in the dental laboratory we must step back and take an objective look at the complete process to determine the root source of the problem. Many solutions may solve superficial problems and not solve the root cause.

There are four main steps in failure analysis

   1. Observation
   2. Evaluation
   3. Investigation
   4. Solution

In the observation phase its essential that you evaluate what failed or physical evidence of the failure. You must observe the failure from every angle, in varied light sources, and under magnification if necessary. Subsequently, you should make a list of everything that is perceived as a problem. After completing this list of observed problems you can move to the evaluation phase.

The evaluation phase is where we consider the observed problems and expand the list to include the potential root sources of each problem. Because many problems have more than one solution, it’s imperative to determine which solution will solve the root problem.

The third phase is the investigation phase. Once all the problems and potential sources have been identified, you may proceed to investigate the complete process to determine where the discrepancy occurred. It’s critical to investigate the entire process taking into account potential problems throughout the process, not just the obvious.

Lastly, the final step of failure analysis is the solution phase. This is where you determine the most optimal means of fixing the root cause. You have already determined what the problems were, where, and why the problems occurred. Now it’s time to fix the problem and to accomplish this you’ll very likely have to change something! If you have determined that there are many potential solutions then proceed and change one at a time and re-evaluate the results.

If you have corrected the root cause of the problem it should not recur. If the problem persists, you must change something else until you have solved the root cause of the problem.

Now we’ll put this procedure into practice by evaluating three failures of all-ceramic castings.

Figure 1: Miscast of two premolars

Observations:

    * Porcelain is barely touching on the occlusal surfaces.
    * Two sprues attached, the diameter of the sprues measured 10 ga. or 2.7mm dia.
    * Fluted connections of the sprues to the pattern.
    * Rounded edges of the flow of the porcelain.
    * Small gas bubbles in the porcelain.
    * Two pellets used to press two crowns.

Evaluation of potential problems:

    * Burn out/mold temperature too low.
    * Mold not burned out long enough.
    * Pellets not preheated long enough to adequately soften.
    * Pressing high temperature too low.
    * H1 preheat time too short.
    * H2 time under pressure too short.
    * Air pressure too low.

    * The small gas bubbles are a result of the incomplete casting. The porcelain was never put under enough pressure in the mold to collapse the bubbles.

Investigation: (Physically go into laboratory and check.)

    * Verify temperature of burnout and pressing furnaces.
    * Verify burn out temperature.
    * Check burn out timer.
    * Check pressing high temperature.
    * Check H1 time.
    * Check H2 time.
    * Check the air pressure on the furnace regulator.

Solution:

    * Calibrate furnaces if necessary.
    * Change air pressure if needed.
    * Change H1 time if needed.
    * Change H2 time if needed.
    * Change pressing temperature if needed.
    * Increase burnout temperature if needed.
    * Increase burn out time if needed.

Remember to change one item at a time to maintain control of the results.

Figure 2: Miscast of a single molar.

The failure analysis for Figure 2 is very similar to Figure 1 with the following additions.

Additional Observations:

    * The sprue attachment has been necked or reduced in diameter at the attachment point.

Additional Evaluation of Potential Problems:

    * The necking of the sprue can result in reduced flow of the ceramic into the mold.

Additional Investigation:

    * Check waxing techniques: The waxers were heating and sealing the sprue attachment without adding additional wax.

Additional Solution:

    * Instruct all waxers to attach the sprue with a drop of sticky wax, next flute the sprue attachment using a lower melting red utility wax. This will assure rigidity yet will not melt the inlay wax.

Figure 3: Miscast molar with 4 sprues attached.

Figure 3 is also a miscast and will have a similar failure analysis as Figure 1. The following points would need to be added to the Figure 1 failure analysis.

Additional Observations:

    * This was the second attempt at casting this crown.
    * There are four sprues attached and all four sprue attachments are necked.
    * The space between the sprues was filled in with porcelain with a smooth round configuration.
    * There was a thin wedge of porcelain extending from the sprue toward the outside wall of the casting.  This would be called a fin.
    * There was adequate porcelain to make the casting.
    * On the mesial and lingual surfaces there were thin layers of ceramic with the internal voids.
    * The casting was approx. 90 percent filled.
    * The pattern was placed in the center of the mold.

Evaluation of additional potential problems:

    * The necking of the sprues can result in reduced flow of the ceramic into the mold.
    * Improper investing techniques causing entrapment of air pockets during investing that filled with porcelain
    * Fins can be caused by too rapid a burn out, by excess pressure during pressing
    * Excess H2 time may also cause the investment to crack under pressure and the porcelain will fill the cracks
    * The use of excess debubblizer may weaken the surface of the investment. This may cause the investment to crack and the cracks would fill with porcelain and cause the fin.
    * Placing the pattern in the center of the mold positions it in the coldest part of the mold. This could cause the porcelain to be sluggish and not flow as readily.

Additional Investigation:

    * Check Waxing Techniques: The waxers were heating and sealing the sprue attachment without adding additional wax. The waxers were attaching sprues to the cusp tips. The philosophy was if two sprues produced a miscast then four will solve the problem. The use of four sprues attached to the cusp tips (continued)
    * positioned the crown in the center of the mold. The center of the mold is the coldest part of the mold because the heat from the muffle has to penetrate through the investment from the outside. Also there is no heat source on the platform so it is heated from the side only.
    * Check investing techniques, check use of excessive debubblizer.
    * Check vacuum level on investing machine.
    * Check vacuum strainer on investing machine to assure it is not blocked.
    * Check burnout time.
    * Check air pressure.

Additional Solutions:

    * Instruct all waxers to attach the sprue with a drop of sticky wax, next flute the sprue attachment using a lower melting red utility wax. This will assure rigidity yet will not melt the inlay wax.
    * Change sprue attachment positioning to eliminate Bubble entrapment.
    * Reduce the number of the sprues so the pattern can be positioned at a 30∫ angle off centerline and toward the outside of the mold.
    * Service vacuum pump on the vacuum investor, clean vacuum strainer, Verify minimum 28î Hg.
    * Eliminate use or significantly reduce use of debubblizer prior to investing.
    * Check burnout cycle and adjust as necessary.
    * Check air pressure and adjust accordingly.

By looking at the above evaluations we can determine that several different problems can be affected by the same solution. If we follow the four-phase failure analysis approach we can easily pinpoint the root cause of any failure in the laboratory. This will ultimately reduce the number of remakes and improve productivity.

Every member of your laboratory team should understand how to evaluate potential problems so they can be eliminated before they occur at each manufacturing phase.