What Is Meldin?

15 Aug


Porous Polyimide Retainer Materials: Meldin™9000 and Minapore™

Originally issued Date April 1992
William E. Titterton (Technical Director, The Furon Co.)
Peter C. Ward (Product Engineering Manager, Timken Super Precision)

As early as the late 1950’s, a need was identified to develop a porous retainer material that would overcome the shortcomings of phenolic retainers. These problems included low oil retention, short service life, and some tendency to develop debris.

The Evolution and Refinement

In 1968-1969, Charles Stark Draper Laboratories (CSDL) proposed porous polyimide as a candidate for retainers, and they demonstrated the capability of manufacturing this structure. The criteria that was used that lead to this choice was:

  • Stable at high temperature

  • Chemically resistant and compatible with all lubricants

  • No foreign chemistries

  • No leach outs

  • No out gassing

  • Manufactured using concepts from powdered metal technology

  • Tight micro pore diameter and total pore volume control (e.g. .75 microns and 12% porous)

  • Sufficient strength

In 1970, CSDL began work with Dixon Corporation on the U.S. Air Force sponsored research project. Meldin™ 9000 was the resulting porous polyimide. By 1980 period it was qualified for use on a few United States military programs as a result of the product’s successfully completing 3,000 hours of life testing at Customer’s testing facilities. In 1980, Singer-Kearfott, together with Dixon Corporation, began work on the U.S. Navy sponsored Program to develop a stronger, more easily machinable Meldin™ 9000 product that would also have tighter controls on the pore size and volume. In order to achieve these results, Dixon went back to the basic chemistry of the process and made several significant modifications in the material workup.

In 1987, the Material Development group at Dixon began an exploratory project to evaluate materials and process techniques that would lead to a commercial porous bearing product. Timken Super Precision and Dixon both felt that an industry need existed for a less expensive retainer material than the Meldin™ 9000 with performance superior to the presently employed composites, the uses to be in non-critical products and for commercial applications.

A shopping list of desirable properties was drawn up to provide for a focus for what we hoped to achieve from the materials evaluated. These characteristics included the following:

  • Temperature range -60 degrees F to +350 degrees F

  • Chemically inert

  • Low thermal expansion

  • High tensile strength

  • Stiffness/toughness

  • Low out gassing

  • Controlled porosity

  • Manufactured by one or more known techniques

  • Resistant to cleaning solvents

  • Resistant to radiation

  • Oil wettable

  • Low water absorption

  • Easily machined

In 1992, after reviewing dozens of different possible candidate resins and varying combinations it was again determined that polyimide is the best choice. The resultant product of this effort is named Meldin™ 8100, Minapore™. It became necessary to innovate our approach to manufacturing and later the test parameters for the product in order to achieve the primary objective of the project; that of producing a lower cost alternative to the materials presently used.

Minapore™ its properties and performance

The success of Meldin™ 9000 over the years to its mature state has just been described. The following sections of this paper relate to work done evolving a more commercial through-porous material .

Miniature bearings have many retainer types historically available for the designer to use. Pressed stainless steel ribbon and “C” type retainers, molded plastic retainers, a category of oil reservoir types, phenolic and Meldin™ 9000. This last category was felt to need a third choice. Graphically speaking, on one end of the cost scale, we have the very mature use of phenolic retainers, which are very economical and can hold 1-4% of oil by weight. But suffering from relatively high cost causes it to be used only in the more sophisticated applications. The other parameter represented on the scale is propensity for debris generation. Phenolic retainers suffer from tendency towards particle generation, just due to the nature of the two phase material. Meldin™ 9000, a single phase, homogeneous material can be processed such that the likelihood of foreign material and particle shedding is eliminated. All bearings that run by necessity with very small amounts of lubricant in the ball race interface for torque considerations, cry out for life extension which through-porous, high oil retention materials could offer.

Dixon Industries and Timken Super Precision have jointly pursued the search for an economical through-porous plastic that would fill this gap. This paper endeavors to report the above properties of the evolved, commercially processed material, demonstrating its application to miniature bearing retainer needs and compares ring tensile strengths, oil retention ability, oil bleed rate propensity, cleanliness compared to finished phenolic, and finally data will be described that shows improvement in ball bearing performance in a typical gyro application.

Ring tensile strength

The data in Figure 1 illustrates the ring tensile strength in pounds per square inch measured using the procedure outlined in the IBWG porous polyimide military specification for Meldin™ 9000. The tensile strength results are plotted against the total pore volume in cc/gram and illustrates the (suspected) relationship of strength to percentage of holes. The salient point is that the commercial product has been tailored in terms of pore volume to give a minimum ring tensile strength of 2700 lbs per square inch corresponding to a pore volume maximum of approximately .18cc/gram. The new process for making this material was tailored to increase the strength over Meldin 9000 by shifting the total pore volume down below that normally obtained in Meldin™ 9000. The result is minimum ring tensile strength of 2700 PSI where Meldin™ 9000 is 2000 PSI. This increase was felt to be advantageous for commercial applications.

Figure 1: Ring Tensile Strength
versus Pore Volume (cc/gm)

Oil Retention

In Figure 2, oil retention is plotted against total pore volume similar to the ring tensile data and as expected shows a direct correlation. At the extremes a .20 cc/gram retainer can hold up to 16 or 17% oil by weight and a .14 cc/gram retainer can hold as little as 8 or 9% oil. The oil used in these tests was MIL-L-6085A, vacuum impregnated in an R3 size (“M”) type retainer. After vacuum impregnation, all specimens were centrifuged. One interesting note is the surface oil retention of the .1 cc/gram specimen. It was reported in earlier work (1) that the pore size and the amount of holes in that specimen was too small to accept MIL-L-6085A oil. In effect it approximates the amount of oil clinging to the surface and is in the same order of magnitude as a phenolic retainer, approximately 3% by weight.

Figure 2: Graph of Oil Retention versus Pore Volume

For the commercial product, a balance between tensile strength and oil retention was reached. 8% minimum oil retention was chosen and controls the minimum point of pore volume. Measuring and controlling oil retention in a standardized manner, one can guarantee getting pore volumes bigger than .14-.15 cc/gram, thus getting more than twice as much oil for bearing life as compared to phenolic retainers.

Interest in other types of oil has lead to preliminary data illustrated in Figure 2, that heavy oils such as Bray 815Z, are retained in the retainer to the same volumetric extent and because of their molecular weight, result in much larger oil retention percentage numbers.

Oil Bleed Rate

Timken Super Precision measures oil bleed rate by consecutive centrifuging of an impregnated retainer at 450G’s, at room temperature. Figure 3 illustrates the results of four different Minapore™ R3 retainers, Meldin™ 9000 and phenolic of the same size for comparison. The oil retention is measured every five minutes and is an indication of oil leaving the retainers. The total oil loss in 25 minutes seems to be quite consistent whether phenolic, Minapore™ or Meldin™ 9000 is used. This means that bearing performance should be predictable and similar to the other two retainer materials. Further tests were done with three Minapore™ retainers of the same size but impregnated with Bray 815Z oil. Predictably, the original oil retention numbers start high but the bleed rate is on the same order of magnitude as the light MIL-6085A oil. This data allows one to expect the Minapore™ retainer to hold large amounts of oil in reserve for use in the bearing but meter it out to the contact surfaces in the same predictable way that phenolic and Meldin™ 9000 have done for years.

MINAPORE™ Bleed Rate – with Mil- L-6085 Oil

OIL % vs. Time

Sample Number

Time in Minutes







Oil Loss (mgs)











































Centrifuge at 450 G’s at Ambient Temperature

Figure 3. Oil Bleed Rate – Mil-L-6085 Oil

Shedding propensity

Polyimide material is inherently clean, as shown in a commonly used test. This test involves a simple procedure using a hot water ultrasonic process on a sample batch of phenolic yields, even after good cleaning techniques, still a slight amount of particles being removed from the retainer surface. In contrast, the Minapore™ retainer, after proper washing and being subjected to this test, shows absolutely no regeneration of particles due to the ultrasonic energy. This rather qualitative result has been quite typical over the years of Meldin™ 9000 and it is not surprising that the nature of the Minapore™ material is very similar. Many lots of Minapore™ have been tested with identical results.

Northrop Corporation in Norwood, Massachusetts has quantified this kind of data by passing a carrier liquid which has been exposed to the retainers through a laser particle counter. Of particular interest is the very dramatic improvement in particle sizes bigger than 15 microns, which must be fairly substantial chunks of resin or cotton breaking off the retainer.

Chemical and Thermal Stability

Thermographic analysis was used to measure the stability of Minapore™ material by Northrop. The weight loss vs. temperature in nitrogen was measured. The results indicate phenolic starting to lose weight in the 250 degree C region whereas the polyimide Minapore™ material was stable up through 640 degree C, at which time the test was stopped, and only 1% loss was recorded.
Differential scanning calorimetry was also used and similar results show the phenolic with two major swings in heat flow, the first thought to be the complete curing of the leftover uncured phenolic resin and the second excursion showing the carbonizing that was indicated on the TGA analysis. No heat flow swings were noticed with the Minapore™.

Bearing Life

Bearing performance degradation is the measure of life in many sophisticated miniature bearing applications. Testing in a customer’s actual system and measuring parameters significant to the customer’s is the true test. Northrop Corporation of Norwood, Massachusetts has been working with Minapore™ retainers in gyros and shares data. It represents a gyro parameter, drift versus time, as an indication of gyro bearing torque perturbations. (Figure 4).Northrop engineers feel that the inherent cleanliness and extra reserve of lubricant contribute to continued accepted drift performance in Minapore™ retainers.

Figure 4: Minapore™ and Phenolic Retainer Drift


The purpose of this work was to show the basic parameters that produce a superior oil retaining material for miniature bearing retainers. These parameters, strength, oil retention, bleed rate, shedding propensity and stability have been compared to Meldin™ 9000 and phenolic.
Oil retention, shedding propensity and temperature capability are superior to phenolic, strength and cost are better than Meldin™ 9000, but not quite as good as phenolic. This new material, Meldin™ 8100, as rod stock from Dixon, and Minapore™, as retainers from
Timken Super Precision, has been found to meet the goal of finding an economical porous retainer material that fits between phenolic and Meldin™ 9000.

Please note that there are size limitations to porous polyimide due to the available rod stock. Please contact our sales representatives to determine if a retainer availability.


1. “Preliminary Work on a New Retainer Material”, P. Ward and T. Messa, IBWG Meeting 1990.
2. Development Laboratory Reports, The Dixon Facility of the Furon Company, 1989, 1990, 1991.
3. Mantech Project, Dixon Industries and Singer-Kearfott, 1989.


The authors would like to thank Mike Leary of the Dixon Facility, Furon Company, Dana Caldwell of Timken Super Precision, Tony Messa, formerly of Dixon Industries, and Jim Gingrich and Mark Schwartz of the Precision Products Division of Northrop Corporation, for their contributions and cooperation.

1 Comment

Posted by on August 15, 2007 in General Knowledge, Mechanical Plastics


One response to “What Is Meldin?

  1. Dr. Giuliano D'Andrea

    June 25, 2009 at 2:01 am

    I enjoyed reading your article. I would like very much to know more about this product particularly its cost and availability.
    Thank you.
    Giuliano D’Andrea, Ph.D.
    USASETAF (A) Science Advisor
    CONUS COM: 011 39 0444 717645
    Italy COM: 0444 717645
    DSN: (314)634-7645


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: