multilayer PCB

But did you add a 6 axis accelerometer and compass sensor to keep the worm hole centered during rotation?

For some obscure reason it reminded me of this:

Wondering what it is, is this a microwave beam position sensor?

This because I once did some work on a beam position sensor for a lineac, reminds me of that, but that had cavities and detectors around the beam.

Curious.

No, it's part of the beam modulator system for the NIF laser. There's a circular delay line that generates impulses every 250 ps. That particular version of the board didn't work, but I kept it because it spins so well.

The linear version did work:

but nowadays we could do it with an 8 Gs/s DAC.

Nice, that is the real high-tech!

Them aliens will like it too.

Is Buttkiss your name or your fetish?

Is that what you call a new board spin?

Who makes such frankenboards and how many kidneys does one need to sell?

Best regards, Piotr

That might hjave been state of the art back in 1990.

And it has come out bowl-shaped?

If you wanted non-dispersive strip-line transmission lines, it makes sense to make the two layers in the middle of the stack microwave laminates.

You can only get even numbers of layers, so the stack isn't completely symm etrical, but FR4 bonded glass fibre is pretty stiff.

Back in 1988 we used isocynate bonded teflon cloth as our "microwave lamina te" which is pretty soft, and didn't have any trouble on a triple extended Eurocard, but we confined it to the two outer layers - microstrip was good enough for what we were doing.

This didn't stop some idiot at the printed-circuit shop deciding that our s tack up order would make the board bend and shuffling it, which meant that out first board didn't work - and it took us a few weeks to find out why - but the subsequent boards where the layers had been stacked up correctly di dn't bow either.

We did have one small board with a teflon-bonded alumina substrate, but tha t was a single layer double sided board.

Most board houses will do this. The good ones will warn you not to do this.

It would probably be OK if the stackup were symmetric. This wasn't.

The production board was all FR4.

No, more like a potato chip.

huh? if it resembles a hyperbolic surface how does it have a single point intersection with the plane (of the benchtop)? is there's something else under it?

Does this mean you had to respin the board?

A couple of times.