"There will be 100-W surface-mount converters within 18 months . . . but not exactly what we recognize as traditional surface-mount," said a coy observer 18 months ago. A telling statement indeed, for what we've really seen is a cascade of comment touting components as "surface-mountable" vs. "surface mount," a horse of an entirely different power.

The one definition we believe holds up, and should, says a true surface-mount device comes in tape-and-reel, uses pick-and-place production equipment, and can withstand 280 degrees C reflow soldering.

If it doesn't meet these requirements, it doesn't qualify. Even neglecting the first criterion still places the ball more into the hands of packaging gurus and pick-and-place equipment makers than power IC/module technologists per se, for 125 W can be packed into a module measuring only 2.3 by 2.4 by 0.5 inches right now.

But there are questions: Can the module's weight be held to 70 grams (the "maximum" weight being bandied about by converter makers) and package area be limited to about 2 by 2 inches for the equipment's vacuum pickup to handle? (Present equipment can manage upwards of 100 g.) Will the larger package surfaces be sufficiently flat for pickup, and the coplanarity specs be within limits for sitting properly on board? Will the weighted-down pc board and its pads and traces be sufficiently rugged to survive the drop test?

Despite some optimistic talk and a bit of hype, we think the answer overall is no, not for regulators (non-isolated) or dc/dc (isolated) devices of substantial power.

But that still doesn't change events near-term, which relates to a few 25-W+ power converters touted as "surface-mount" we think could arrive before summer's end. We're not clear how their physical appearance will stack up, but if carried off, they could represent a crossroads in the evolution of true surface-mount dc/dc converters. In the best scenario, they could lay out a new philosophy for surface-mounting denser power sources.