原创 Oven reveals flip side of thermal management

This became very clear after I read a fascinating article in a recent issue of IEEE Spectrum about the ineffectiveness of the standard home oven. The article, "Recipe for a Better Oven" by Nathan Myhrvold (yes, the former chief scientist at Microsoft) and W. Wayt Gibbs, discussed the many shortcomings of the standard home and even commercial baking oven.

It was both illuminating and somewhat discouraging for many reasons. It's not just that the oven is wasteful in terms of energy use, it's that it also does a poor job despite that inefficiency; whereas I had hoped that the inefficiency was a necessary part of the tradeoff for getting the baking done right.

I already knew that the simple on/off control of the gas line or electric power was a marginal approach. Anyone who has studied any control theory knows that a loop with on/off control, rather than true proportional control and a PID (proportional-integral-derivative) control strategy, is not good at keeping the error between setpoint and actual value small, unless you are willing to switch that on/off valve at a fairly high rate.

While it is true that thermal situations tend to have relatively long time constants and so can accept a somewhat lower cycling rate than some other situations, oven vendors like to keep the valve's cycle time to a few minutes. So it is common to see swings of ±50°F (about ±25°C) around the setpoint. That's just not good enough for many recipes or foods.

The problems of the standard oven design and subsequent performance go well beyond the basic temperature-control loop. Only after I read the article did I realize how much I hadn't known about the implications of oven design and the cooking process.

For example, the heat source doesn't cool the food directly; instead, it heats the oven walls (metal or brick), which then re-radiate the heat to the food. The article explained how this leads to very uneven heat zones in the oven for conventional metal-wall ovens, and numerous other problems. (The article did highlight some ways to improve an oven's performance, which was nice to see.)

These other problem are not trivial, either, as they have to do with how different foods are affected by the radiant heat of the oven walls versus direct heating; not surprisingly, it does make a difference, just as different types of electrical load impedances affect power supply performance. (There are times when I felt the whole discussion of oven idiosyncrasies was akin to the mysteries of black-body radiation, which were finally resolved with the development of quantum theory in the 1920s by Max Planck and others.)

The irony of the oven problem is that for most engineers, the primary concerns regarding heat are twofold: minimizing generation of it in the first place, through the use of low-power circuits and high-efficiency supplies; and maximizing its removal, using convection, conduction, and radiation via use of fans, heat sinks, cold plates, and even advanced active techniques. It takes a 90° or even 180° shift in thinking to begin to understand heat and its effective delivery as an outcome rather than as an obstacle.

Have you ever been faced with a design challenge that required that you reroute your established way of looking at something, to a very different course? Did you realize the differences right away, or did you have to really step back to understand the many inherent assumptions you were making and how they had to be changed as well?