Selecting resin the GM way.

Inside This Story:

* Resin systems have specific strengths and weaknesses, often forcing a balanced compromise to be made for a particular molding system.

* GM's Gen IV V8 aluminum block project needed to select a resin system, but the requirements made finding a proper compromise complicated.

* Detailed in this story is how a selection team broke down the complex issues with resin system selection to make a logical fact-based decision.

Metalcasters must make daily decisions on which products should be utilized in their plant's production process. One of the most difficult decisions is the selection of a sand resin system and supplier. Each resin system generates a combination of properties that the supplier will tailor to suit the needs of the customer's particular casting application. A perfect system does not exist, however. Each system has specific strengths and weaknesses that must be balanced to achieve the best "compromise" of performance properties for the customer.

General Motors (GM) Powertrain Div., Saginaw, Mich., faced such a decision for its Gen IV V8 aluminum block project. The block will be produced at GM's Saginaw Metal Casting Operations (SMCO) starting in the 2007 model year using the precision sand process, which requires an integral cured sand package. The project team selected the phenolic urethane coldbox system due to its high productivity and GM Powertrain's production experience with the system. The mold package is projected to weigh 700 lbs. (318 kg), and production is estimated to be 1,500 blocks/day.

A resin system had been selected for preliminary development work to establish initial resin performance criteria, but a resin system intended for actual production still needed to be designated. Due to the complexity of selecting the proper resin system, a multidiscipline engineering team from environmental engineering, purchasing, manufacturing engineering and production was formed to complete the assessment of the various coldbox systems available.

Project Initiation

Because some properties of resin systems are affected adversely by the enhancement of other desirable characteristics, the final product is a balancing act of finding a system that can deliver the critical characteristics, while maintaining adequate properties in other areas.

For the Gen IV block project, the team established criteria to be used in the selection process and grouped them into four major categories:

* quality (resin performance, shakeout characteristics, etc.);

* safety (emissions and odor issues, etc.);

* people/responsiveness (quality of the supplier's technical support);

* cost (price/lb.).

It was necessary for all important criteria to be identified and to establish a selection process that allowed the significance of each variable to be incorporated.

The team adopted a "modified Pugh" analysis methodology for selection activities. In this method, the relative weight of each of the separate criteria is established using a 1-10 scale, with 10 representing the most important criteria. After the criteria and the relative weight of each are determined, different concepts are evaluated to assess their relative capability to meet the needs of the customer. As each concept is assessed, it is assigned a rating with respect to the concept's ability to address the datum, or acceptable target range of criteria.

In the modified Pugh system, each concept is given a score of 1 (for criteria that does not meet the datum), 3 (for meeting the datum) or 5 (for exceeding the datum). A "show stopper" rule also was established. This dictates any criteria that is rated a 10 on importance must be met in order for that concept to be considered. The concept that achieved the highest cumulative score then would be selected.

Developing the Criteria

A critical step to using this approach was to develop good criteria. Based on the customer's needs, 59 criteria were developed, including tensile strength, humidity resistance, flowability, price per pound, technical lab capability and Material Safety Data Sheet (MSDS) component rating.

Each team member submitted his or her individual weighting of the criteria. The individual weightings then were reviewed, and a consensus weighting for each criteria was created. The datum was developed with the use of prototype castings that allowed many of the datum to be based on the development resin system performance.

Narrowing the List

After the purchasing department reviewed a preliminary list of resin system suppliers, it identified two that had the capability to meet GM's requirements for delivery and cost. Each resin supplier was allowed to recommend two resin systems for the evaluation. The existing development resin system was identified as the baseline system for the evaluation. The selection team suggested the suppliers provide one resin system designed for optimal performance and one resin system developed to be more environmentally friendly.

Two phases were used to test each system. In Phase 1, data was collected for most of the criteria on all five resin systems (two systems from each company and GM's initial resin system). This would include most of the lab testing. For the second phase, the selection team chose the two resin systems that scored the highest and tested those systems for the criteria that required actual production of Gen IV mold assemblies and castings.

Testing Begins

Tensile tests were run at the Univ. of Northern Iowa, Cedar Falls, Iowa, and consisted of 30 samples for each of the criteria. This required 100 total batches of sand and 1,800 tensile specimens. In Table 1, which reveals the tensile test results, the baseline resin system is coded as "C." This baseline system had an average of 168 psi and a standard deviation of 15 psi in the immediate benchlife test with a 5-min. break. This resulted in a datum of 153183 psi. Resin D had an average of 201 psi, which is better than the datum and results in a rating of 5, while Resin E had an average of 171 psi, which is within the datum and results in a rating of 3. This same procedure was applied to each separate category to establish the ratings for each criteria.

Not all tests for each criteria revealed a marked difference among the resin systems tested. For instance, criteria involving expansion, erosion resistance, casting surface finish, casting gas defects, casting dimensions and gas evolution all exhibited such small differences that no significant contrast could be found. However, the resin systems performed differently in the flowability evaluation test.

Checking the Flow

Due to the thin web section on the block water jacket, the selection team wanted to evaluate the ability of each resin system to produce an acceptable thin section core, especially at a longer benchlife. Instead of a 1 x 1-in. (2.54 x 2.54-cm) tensile biscuit, a two-cylinder water jacket corebox was produced that could be utilized on a standard core blower (Fig. 1). The dimensions were similar to the design of the proposed Gen IV production jacket core. This tool allowed performance testing to evaluate the flowability of the sand mixture at various resin levels and benchlife times (Fig. 3).

[FIGURES 1 & 3 OMITTED]

The team also wanted a feel for the strength of the actual core. A prototype core breaker was designed using an inflatable rubber bladder to break the test jacket core (Figs. 2 and 4). A simple regulator was installed into the compressed air supply line to help in the control of air pressure to the bladder. The pressure was increased slowly until the core broke.

[FIGURES 2 & 4 OMITTED]

In the water jacket evaluation (which lasted 4 hrs.), a batch of test sand was mixed and blown into the water jacket corebox every 15 min. The core was visually examined for complete fill and then weighed. After these checks were completed, the core was broken on the core breaker. In this test, Resin A had the best performance. Resins D and E were consistent in performance, but Resins B and C exhibited lower strength values than the other three systems (Fig. 5).

[FIGURE 5 OMITTED]

Assessing Environmental Performance

The environmental performance of a resin system is just as important as a technical performance. A resin system with great strength, flowability and benchlife is no good to a metalcasting facility if it poses a hazard to the workforce. Therefore, the initial evaluation of environmental performance was a review of the MSDS by the GM environmental department. Department members looked at each component of the five resin systems and rated them 1-5 with regard to volatiles, semi-volatiles and naphthalene.

Two levels of emission tests were completed. During Phase 1 of the testing, total volatile organic compounds (VOCs) were measured via EPA method 25A on all the test resins during the pouring of test cores. The tendency to generate smoke was measured using the core opacity test, and gas evolution tests also were run. During Phase 2 casting trials, total VOCs and speciation tests were conducted on the actual core packages produced from the two resins selected for the final evaluation.

Another environmental issue was the odor of the resin system. While conducting casting trials in Phase 2, odor samples were collected during pouring and shakeout and sent to an odor panel for evaluation.

Customer Service Matters

In order to best assist GM in the development of the Gen IV casting process, the resin system supplier needed to possess technical problem-solving skills, knowledge of the precision sand process and the ability to respond quickly to technical issues. Each supplier made a presentation to the Gen IV team outlining their capabilities, which was then correlated to the Pugh analysis spreadsheet.

Another major criteria for the selection process was the resin cost and the per-piece cost of the castings. While the other evaluations were being completed, GM purchasing worked with the suppliers to establish the final cost for each resin. After the final quotes were confirmed, the data was added to the Pugh spreadsheet.

Picking the Supplier

Once all of the data was entered into the selection spreadsheet, the total points for each major section were calculated. The team then had enough information to make a logical decision based on data and facts.

The methods used in the resin selection for the Gen IV block project have been used successfully for several GM projects. As demonstrated in the resin selection process, large complex issues can be broken down into small measurable criteria to achieve an effective decision.

This article was adapted from a paper (05-171) presented at CastExpo '05 in St. Louis.

About the Authors

Larry Stahl is a development engineer and D.J. Couture is a process manager at General Motors Corp., Saginaw, Mich.

For More Information

"Evaluating Coldbox Binders: A Practical Approach," S.G. Baker, MODERN CASTING, May 2001.

"Evaluation of Sands for Coremaking Process: A Practical Approach," J. Werling, AFS Transactions (02-030).

Table 6. Pricing Methods Identified by Responding U.S.
Metalcasters

                                              Number of
Method used to determine price                responses

Negotiated between producer and
  customer for each job                       250
Negotiated between producer and
  customer on a 1-yr. or less contract        124
Negotiated between producer and
  customer on a contract of more than 1 yr.   96
Price set by producer
Price set by customer                         91
Price set based on imported product price     35
  or by imported product offer price          30

Note: A total of 403 firms responded. Many indicated more
than one price determination method.