What Temperature Is It, Really?

In heat processing, we regard temperature as one of our most critical control variables. Get it wrong, and you're in trouble. Yet, a couple of recent discussions with customers drove home the realization that the correct process temperature is what we perceive it to be.

One of those conversations went something like this:

Customer: “The product needs to be 190oF, and I have the oven controller set at 220. It can't stay up with production.”
Consultant: “That's because you need to set the oven at 450oF, and when you run the heavier stuff, it'll have to be even higher.”
Customer: “Huh?”

Here's the gist of another one:

Consultant: “What product temperature do you need?”
Customer: “I need a core temperature of about 200oF.”
Consultant: “What's that equate to in terms of product skin temperature and oven temperature? We need to determine a control setpoint.”
Customer: “I have no idea.”

Heat, like water, flows downhill. In a convection oven, the hottest place is the burner flame or heating elements. Their high temperature is diluted by the circulating airstream, which, in turn, is hotter than the oven heating chamber and the product. That's how you get heat to flow to the product. In high temperature direct-fired furnaces, the process is much the same, except that you usually depend on flame-to-furnace-to-load radiation to do the bulk of the heat transfer work.

When determining the appropriate process temperature control point, the ideal method is to measure the actual product temperature, but it often isn’t feasible. Instead, look to establish a reliable cause-and-effect relationship between good, consistent product quality and the oven air temperature (or some other easily measured process temperature) and heating time.
And that's where the confusion often arises. As the temperature goes bumping down the steps from source to destination, where do you measure it? Ideally, you'd like to measure the actual product temperature -- that's what matters, after all -- but it often isn't feasible. How can you probe the temperatures of 500 pieces per hour as they zip by on a conveyor? Contact thermocouples wouldn't be practical. Maybe an infrared scanner, but if the temperature varies significantly from piece to piece, how do you keep the temperature control system from thrashing itself to death trying to make all those corrections? And what about processes where product core temperature is all-important? Drilling holes in each piece is out.

The relationship between surface and core temperatures will vary widely, depending on the thermal properties of the load, its dimensions and whether it contains moisture or solvents that have to be driven off.

It's because of considerations like these that you often have to settle for implied temperature control -- if the oven air temperature is A, then the product will be B degrees at the right time. And how do you know it will? Experience and cut-and-try testing. From this, you soon learn that at two different loading rates, products that need to be heated to the same temperature may require different oven setpoint temperatures. Because heat transfer rates are limited, the only way to get the product to its desired temperature in the time available is to run the oven at an elevated thermal head. The higher the loading, the higher the thermal head must be.

If you're having trouble grasping this concept with convection ovens, think infrared heating instead. Most conventional infrared heaters operate at temperatures between 1,400 and 1,800oF (760 to 980oC). They're commonly used to heat products to temperatures of up to 800oF (425oC). You don't want the product to come up to anything approaching the heater temperature, so how do you keep it from overheating? You move it away as soon as it comes to temperature; otherwise, you'll cook it.

This points out the other critical variable in heating -- time. Give a load enough time in the presence of a heat source, and eventually it will approach the source temperature. Where you have a product that absorbs heat quickly, you have to pay close attention to the heating time to avoid overexposure. Oven cycle times and conveyor speeds, however, are often driven by the needs of other parts of the manufacturing process -- you may not be able to vary them at will. In situations like these, your only recourse is to adjust the oven's control setpoint.

The bottom line is this: In the majority of cases, it isn't practical to measure the actual product temperature on the fly. In many cases, you're not even sure what it should be. However, if you can establish a reliable cause-and-effect relationship between good, consistent product quality and the oven air temperature (or some other easily measured process temperature) and heating time, you've done your job.

Links