Electronic and thermal feedback loops

Phil, our Undergrad Text is the latest Edition of "Fundamentals of Momentum, Heat, and Mass Transfer", Welty, Wicks, Wilson, and Rorrer

ISBN-13: 978-0470128688

ISBN-10: 0470128682

Older editions abound.

Steve

No. There's no TEM mode in waveguide because there's no centre conductor. Both TE and TM have axial components of the field.

You can get them in free space too, e.g. near a focus. At high numerical aperture (i.e. a fat converging cone), the field components coming in from different directions aren't collinear. If you imagine a field component near the edge of the cone, coming in with its E field along a radius of the cone, it has a huge component along the axis of the focused beam.

This also makes the resolution dependent on direction and polarization, which is very important in high-resolution lithography, for instance.

I once designed a microscope to work at a numerical aperture of 6.4 in silicon. It used a 50-nm air bearing to avoid touching the surface. The light was of course horribly evanescent in the gap, but using tangential polarization (E always perpendicular to the cone radius) improved that out of all recognition.

Then IBM lost 8 billion dollars in one year, and both my budget and my customer went away. :(

Lots of people use the general technique now, anyway.

Cheers

Phil Hobbs

Thanks, Steve, I'll have a squint. Probably I don't need the cutting edge treatment of water-cooled heat sinks. ;)

Cheers

Phil

I was unclear. What I meant to say was, "with its E field in the plane defined by the beam axis and the cone radius", i.e. P-polarized. Sending a linearly-polarized laser beam through a high-NA lens produces P-polarization on two sides of the cone and S-polarization on the perpendicular sides, with various linear combinations in between.