A team of human smell-testers, a $75,000 computerized 'electronic nose' and a row of household canning jars all play key roles in ensuring that while customers may enjoy the 'new car' smell of a Ford Focus, there should never be any unacceptable odors coming from the car's interior.

For 25 years, Ford specialist have evaluated the aromas from potential trim materials at the Company's research laboratories in Germany and Britain. Now, for the Ford Focus, this expertise will be used to 'teach' a computer-based electronic 'nose,' which will be able to check routinely that components from suppliers meet agreed low-odor specifications.

"We believe that Ford is the first automotive company in the world to use one of these electronic noses," says Dr. Jorg Sassmannshausen, who heads the analytical section of Ford's chemical labs in Cologne. "The electronic nose is just one example of Ford's response to the fact that customers are becoming more and more conscious of quality in every aspect of the vehicle. Unwanted odors create a poor impression."

The team of 'super noses'

 

Dr. Jorg Sassmannshausen heads the analytical section of Ford's chemical labs in Cologne

Even before considering smell, Ford checks to ensure that no potentially harmful materials are proposed for the interior. A regularly updated list of all substances known to be poisonous, carcinogenic or otherwise harmful are banned from use, more than 2,000 in all. "That still leaves a long list of materials that we could use, for seating foam, textiles, thermoplastic trim parts, rubber seals and so forth. Some may naturally have little or no odor, but for others, it's more of a problem. For example, the foams and padding materials used deaden unwanted sounds are some of the most sensitive for smell, explains Dr. Sassmannshausen.

The Ford lab in Cologne has also done extensive testing with recycled materials, where minute quantities of foreign matter, such as paint, can give rise to unpleasant smells. Such a material may be acceptable for a wheel-arch liner, for example, but would be disastrous in the heating and ventilation ducting.

Work on initial sampling of potential materials for the Ford Focus was concentrated at the Company's lab at Dunton, England, though the same test procedures are regularly used in the Cologne facilities, too.

Each material undergoes a standard process in the lab, designed to simulate real-life conditions in a car on a hot, humid day, which is when smells are most noticeable. Relatively large samples of the materials-about 10 cm by 20 cm-are sealed in airtight containers, and heated to 40 degrees C for 16 hours. Ford's experience has proved this is the optimum temperature, as anything higher actually renders the sensors in the human nose less responsive to smell.

A small quantity of water is also included in each jar. Water vapor is a good transmitter of odors, making the nose more sensitive to any traces that may be there. The test simulates the real-life situation of driving through summer rain, with high temperature and high humidity, when any smells are most likely to be evident.

Once the test samples have been heated, the expert 'noses' go to work. A panel of seven people from the lab evaluate each sample in turn for a few seconds, and rate it on a four-point scale. One point means no smell detected, two means a smell is detectable but that it would be acceptable to the customer, three is unacceptable, and four is disgusting!

The panel of noses is chosen to provide a representative range of customers. The best and worst of their individual ratings are discarded and remainder averaged. If the result is an overall rating of two or less, the material is judged acceptable.

"Though necessarily a subjective test, it's one with which we've developed a lot of experience, so we can ensure consistency," comments Sassmannshausen. "It's also important that we run tests in the clean atmosphere of the lab. On the factory floor, for example, the nose is likely to be saturated with smells and much less sensitive."

In Ford's high-tech labs, it is a surprise to find that the containers used for these tests are standard 3-liter canning jars, like those used in the kitchen at home. Sassmannshausen says that they are ideal. "They're glass, so they produce no smell of their own to interfere with the test, and the wide openings easily take the samples. We've never found any standard chemical beakers as good," he says.

Electronic nose a 'first' in Europe

During the last two years, Ford has been working to extend its expertise with material smell-testing, and develop a routine process which could easily be used as part of the quality control system by its component suppliers.

Human testing in this case would be impractical for reasons of complexity, consistency and cost, so Ford turned its attention to the computer-based electronic smell detectors used in industries such as food, cosmetics and packaging. Their normal tasks range from checking water quality at breweries across Europe to establishing the freshness of fish in Japan.

These 'electronic noses' can even differentiate between one- and three-day old fish, and were just as discriminating in identifying good and bad samples of automotive materials in Ford's tests. In fact, the machines have proved even more consistent than their human equivalents.

"The crucial difference with the electronic nose is that it has to be 'trained' step-by-step to recognize what's good and what's bad. A person reacts instinctively, whereas the machine can only compare each new sample with its previous results. But once trained by our human noses, the electronic one has proved very reliable. It never has a head cold or a bad day," says Sassmannshausen.

The electronic nose in the Ford lab uses an array of 12 chemical sensors, which each respond to different components within an aroma to produce a 'fingerprint' which identifies the material under test. The machine can then seek a match with other 'fingerprints' in its memory.

The individual sensors have a polymer surface which acts as a conductor between two electrodes. The polymers react with the aroma molecules in an air sample, varying their electrical resistance. The change in voltage across the polymer is then simply measured by passing a current between the two electrodes. Each of the 12 conductive polymer sensors has a different structure and so responds to different aroma molecules.

In the Ford lab, the electronic nose operates in conjunction with an autosampler, where test materials in small glass vials are heated for three hours at 80 degrees C. A needle is then automatically inserted through the rubber seal in the vial, to draw off a sample of air, including aroma molecules, and feed it to the sensor head for testing. The autosampler can handle up to 48 samples at a time.

"The trials have proved so successful that we are now working with our suppliers, so that in the very near future they can start routinely using electronic noses in the production of components for the Ford Focus and other cars," says Sassmannshausen. "That would be a 'first' within the European automotive industry."

From the various electronic noses on the market, Ford has identified as most suitable one produced by the British-based company EEV/CSS (formerly Neotronics Scientific), part of GEC Marconi. They have successfully collaborated with Ford in developing processes specifically for the automotive industry.

But despite the potential Dr. Sassmannshausen sees for the electronic nose, it will never entirely replace his specially-trained team.

"The machine is still a poor copy of the human nose, which has more than 20 million sensory cells," he says. "So there will always be a key role for our human 'super noses' in the lab. We not only need their skills to train the electronic nose, but also to cope with the unexpected."

Autoworld.com

Back to Today's News