For the past 400 or 500 years, heat treating has had somewhat of a bad name. The heat-treating department was a private little hell hidden in the back of the shop, and the people who inhabited that murky inferno were considered alchemists. You gave them the parts you had sculpted out of metal, and they performed some mysterious magic on them, and lo and behold, the parts were indeed better, but nobody really understood why.
Remarkably, just in the past few years, heat treating has made some bold leaps into the 20th Century, and it is now ready to pounce into the 21st. Heat treating has become a science. It is finally being accepted as an equal in the metalworking fraternity, and it is now as ready for the factory of the future as the rest of the metalworking processes.
Heat treating is no longer magic, the process and its metallurgy are understood. To meet today's demands, it has become flexible, it has become productive, it has become competitive. And solid-state controls and sensor technology have given it the most desirable property of all: repeatability. A batch today will have identical characteristics with a batch run last week or next year. Heat treating has arrived!
Induction means production
Jack Cachat has been president of Tocco Inc, Cleveland, OH, for about 150 years or so. He knows just about everyone in this business. He knows induction heating's past and present, and is in a great position to predict its future.
As he sees it, the key trend today is using more robotics to automate heat-treating systems and make them flexible enough to process different types of parts. Although induction heating systems are usually in high-volume, dedicated lines, handling at best no more than three or four closely related parts, they now offer the flexibility for mid-volume tasks.
"The challenge today," Cachat reports, "is to design in more workhandling ability and greater flexibility than in the recent past because now we're seeing more variation in parts and smaller batch sizes. Also, with the effects of the new just-in-time philosophy coming to the auto industry from overseas, induction heating is a natural. Because you can shut it off and start it up again so easily, you can run smaller batches, and the problem of different sized parts is simply a matter of changing a certain amount of tooling--an inductor, locator, or transfer device that merely requires an adjustment.
"With induction heating, you can get the type of temperature you want, when you want it, and vary it readily. You can get good quality control with very precise timing of the heat-energy cycle in kilowatt seconds.
"But there are sill gaps remaining," Cachat admits. "One thing we don't have quite yet is a method to nondestructively test for hardness and depth of case as the part comes off the inductor, to prove that the part is satisfactory. Right now, most of that is done visually. We would like to be able to determine grain size and structure without having to cut and polish the part. We would love real-time metallurgical feedback, but there's nothing like that on the horizon yet."
"We are seeing the beginning of the use of radiation pyrometry to sense part temperature and produce a temperature-control signal that modulates either induction-heating time or power levels. This is something that will come. We're using kilowatt-second units to meter cycle time and power level, although we can add a radiometric temperature measurement to show that good control is being maintained. This confirms the operator's suspicion that temperature is being held within limits."
One of the problems is not being able to actually sight on some parts as they are being heated, although others are easy to see. "We will soon be able to integrate temperature readings to adjust operating parameters and gain finer control. We anticipate an increased demand for this more sophisticated level of control.
"Our kilowatt-second control is a digital control that precisely regulates heat input. It's been out for about three years and is in broad use. With the advent of high-power solid-state controls, we can get very rapid response controls for power level, time, and frequency--down to the fraction of a millisecond. It's both sharp and quick. In fact, in some cases, our switching time is actually too short and we have to be careful to apply the power gradually and not shock the part.
"The heating pattern is often custom programmed to the part. The basic programmable controller is ideal for our process, and it can be Allen-Bradley, GE, Modicon, Westinghouse, or whatever you want."
Energy densities are going slightly higher in capabilities. "Where we once used between 20 kW/sq in and 30 kW/sq in, we now see some cases where we go to 40 kW/sq in or even 50 kW/sq in. We use these higher power densities to produce shallower, finer cases by bringing the surface temperature up faster. In one example, we got the very shallow case we wanted with a heating cycle of only 0.4 sec."
Heat treaters are
accepted as equals
These newer capabilities affect part design and the relationship between heat treater and designer. "We try to work with design engineers to get them to stop specifying any more hardening than they really need," Jack explains, "but they are usually difficult to persuade. You have to go through a great many tests to prove to them that what you're telling them is true. They're the ones who take the rap if a part fails for insufficient hardening, so they'd rather be safe than sorry. And you can't blame them for that.
"I think the heat treating profession has gained a lot of respectability today. People listen to us a little more. There are several very good books out on the basics of this science. Sure, at one time they thought we were magicians. We're no longer alchemists to them; we're accepted as equals."
The integration of the heat-treating process into the machining process has been particularly successful in valve-seat hardening on cylinder heads. "Our equipment is fully integrated into a transfer line. They machine the valve seats; transfer the heads through burnishing operations to our equipment, which hardens the seats; and then they move on to other operations."
Yet, despite this example of success, most spindles for the new front-wheel drive cars are hardened today in separate operations. "There is a large potential market here for automation to combine these operations," Jack observes, "for both the hollow spindle used on front wheels and the solid spindles used on rear wheels."
The competition faced by induction heating today includes the people promoting lasers, but Cachat doesn't think they will prove to be economically superior, except for very small parts. "And even there, I would recommend electron-beam methods, even over our own induction-heating methods, if the part's the right size.
"The laser has shown superiority over induction heating for rather small-bore cylinders (1" dia, for example, 6" to 8" deep), where it can produce a spiral-scan hardening and for a few larger bores where induction heating could not be done. This is obviously a design compromise; they don't really want a spiral hardening pattern, but they must take what they can get.
"Another potential competitor, but not very broadly used, is a welding system using an inert gas or heliarc type of welding head. This yields a very shallow pattern and has some inherent problems with starting and controlling the arc. Also, we've been experimenting with plasma approaches, but we have been unable to get the heat source into a small enough area to do any good. So far, it is a big torch and big flame that is not very useful or economically competitive with existing induction methods.
"There is very little work going on combining atmosphere with induction heating. The newer areas are ion nitriding and ion carburizing, but I don't see much practical development work there. The aerospace people are trying to squeeze extra performance out of critical parts, but the rest of the metalworking world doesn't really need to worry about these processes."
What's the future for induction heating look like? "The heat-treating industry, generally, will probably see some growth, but in a lot of specific...