Back to Basics

A Surprise Fix to a Sticky Problem

by Emmanuel P. Papadakis^*

Everyone remembers a parent peeling the backing from an adhesive bandage and applying the bandage over a scrape or a cut. This simple process has immediate relevance to an NDT problem presented to me when I led the NDT Research and Development Group at the Ford Manufacturing Processes Laboratory.

 

The Problem
My group was informed by the Transmission Division that the friction strips on automatic transmission bands were coming unstuck in driving tests. The bands are about 51 mm (2 in.) wide strips of steel about 1.6 mm (0.06 in.) thick. They have small attachment brackets on both ends. The friction strips are a sort of asbestos cardboard adhesively bonded to the inside of the bands. The bands are like hoops on a barrel; they go about 90% of the way around a circle of about 152 mm (6 in.) diameter. In transmission fluid, this type of cardboard is durable and has the proper friction against the inner steel cylinder.

---

The failure mode was disbonding of the cardboard from the steel strip.

---

The failure mode was disbonding of the cardboard from the steel strip. The transmission lost torque. The problem was posed to my NDT group as a requirement to develop a test for adhesion of the cardboard to the steel band.

The assembly method in the factory was to place the cardboard inside the steel band and hold it in place with the adhesive. This adhesive was then cured and made permanent in an oven. The adhesive, before curing, was a tacky, sticky layer on one side of the cardboard as supplied by a vendor. The tacky layer was protected by an impermeable layer which was peeled off just before the cardboard was placed on the steel band. This whole operation was manual.

 

Searching for an Answer
The engineers brainstormed; possible causes of failure such as missing adhesive, poor quality adhesive, unclean steel bands, improper oven temperatures and excessive or inadequate time in the oven were considered. What was needed from us in the NDT group was a feasible test for lack of adhesion.

We approached the problem from the point of view of damping of ultrasonic waves in the steel by the cardboard when well bonded. Pulses (almost as short as impulses) from broadband ultrasonic transducers were found to contain the spectrum of frequencies needed to excite the bands into their fundamental thickness resonance at the location of placement of the transducer on the outside of the band. The resonance lasts a long time with no cardboard but is highly damped with adhered cardboard. This difference provided a solution to the testing problem. The areas of poor adhesion could be mapped by moving the transducer. Research and development took about one day, as we had all the requisite equipment in our well equipped laboratory. After that, we investigated vendors who could provide automated systems to test the bands. The resulting research and development was reported in Materials Evaluation in February 1978 (Figure 1).

Figure 1 — Ultrasonic system for NDT of parts using induced resonance with damping.  The enclosed region is the instrument designed for the NDT work described in the article.

 

A Surprise Ending
However, before the development of the automated testing system could get off the ground, the engineers in the Transmission Division canceled the project. They told us that our idea was excellent but was not needed, as they had discovered the root cause of the disbonding and rectified it.

The cause, as it turned out, was not faulty engineering, poor materials or failure of factory equipment. The cause was lack of training. Everything was perfect except for one factory employee. He was a substitute for the permanent employee who had called in sick. The new man had been trained to place the cardboard inside the band, but he had not been trained to peel off the protective layer first.

Is it any wonder that the bandage did not stick? And is it any wonder that the problem cured itself when the sick man returned to the production line?

Our NDT Research and Development Group, going beyond the state of the art and far ahead of the state of the industry, had invented a useful testing method for a problem that didn't really exist. Essentially it was a waste of time. However, sometime, somewhere, the work should help someone.

 

REFERENCES
Papadakis, Emmanuel, "Ultrasonic Impulse-Induced-Resonance Utilizing Damping for Adhesive Disbond Detection," Materials Evaluation, Vol. 36, No. 2, February 1978, pp. 37-40.