Popular article about it:
Paper:
Like the RF stuff, like I stated last week, let the young generation do things at why not 50 GHz, well they went an order of magnitude higher!
Now this could change a few things!
Hmm..my trusty 60W lightbulb emits the same Terrors-Hurts frequencies, and there are so many sensors that detect same that have been around for the same length of time, that your daddy could have done the same thing(s) when he was young.
On a sunny day (Tue, 11 Dec 2012 20:02:49 -0800) it happened Robert Baer wrote in :
Could you please decrypt that a bit?
Thanks for posting this. If it can be manufactured in a sensible fashion then it is in effect a 4x4 phased array radar implementation using microstrip (maybe that should be nano strip) techniques.
It looks *very* clever if they can beamsteer and maintain a compact beam psf then they do have a cheap cost effective terahertz imager. Actually imaging in terahertz is so difficult it probably won't matter if it is cheap so long as it works! I noticed that terahertz astronomy images that have been made are not for public consumption yet.
They have certainly acheived something quite amazing using silicon instead of exotic III-V materials.
Can anyone like JimT who understands these things explain how they are getting useful gain from small paralleled? transistors at above f(max)? I find it more than a bit boggling.
Regards, Martin Brown
From what I could understand, it's more or less based on distributed amplification (the bit about many small transistors working together)
Ft and Fmax are given as some gm/f(Cgs,Cgd) ratios. Absorbing the parasitics in T lines allow higher working frequencies. It's only 280GHz amplifiers after all...
This is getting us back to the circa 100MHz Tek 585A...
-- Thanks, Fred.
I guess the proper hint is: wavelength of visible light.
-- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico@nctdevpuntnl (punt=.) --------------------------------------------------------------
On a sunny day (Wed, 12 Dec 2012 15:41:46 GMT) it happened snipped-for-privacy@puntnl.niks (Nico Coesel) wrote in :
Yes, but does that penetrate like THz waves to make a pictire of hidden objects? I think not.
But you can make your own airport scanner:
In one word, "incandescent".
Yes. OK I can see how that might work given a high degree of native cunning and some precision layout. Another idea that is exactly of its time (like the Deep Blue chess hw).
Terahertz imagers are potential mould breakers in body scanner and diagnostic capabilities because so many otherwise opaque materials are transparent at that frequency and it is non-ionising.
In astronomy the novelty is that whenever you open up a new waveband to high resolution imaging you inevitably see things that are unexpected. Chandra demostrated this is spades by hires imaging in X-rays.
Regards, Martin Brown
objects?
After reading the article it reminded me of the body scanner demonstration I've seen at TNO. That technology is used in body scanners on airports. I guess people now managed to integrate it on a chip. You'll still need an antenna though. On the TNO demonstator a rotating aluminum plate was used to deflect the signals into a microwave antenna (probably some sort of detector).
-- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico@nctdevpuntnl (punt=.) --------------------------------------------------------------
On a sunny day (Thu, 13 Dec 2012 21:03:22 GMT) it happened snipped-for-privacy@puntnl.niks (Nico Coesel) wrote in :
Clever idea. Perhaps with modern chips you can have many chip like antennas. Yesterday I ran some tests with my microwave field strength meter
ks
ild.
hires,
foil,
foil
Aluminium foil is covered with a thin layer of aluminium oxide - it probably doesn't make good Faraday cages. Slots in Faraday cages make slot antennas which can be quite efficient at frequencies where the slot is about a quarter-wavelength long. Stainless steel is just as bad; it oxidises almost as fast as aluminium, but it is protected by a thin and coherent layer of chromium oxide, which - like the protective layer of alumina - is electrically conductive but nowhere near as conductive as bare metal.
If you want a good Faraday cage, make it out of copper foil and solder all the joints.
Serious screened boxes come with gold-plated beryllium copper finger strips to close the gaps. Gold doesn't oxidise but if there is sulphur around it will eventually tarnish - when I was young we had to polish the gold-plated fingers on our printed circuit cards about twice a year to keep the racks working. Modern DIN 41612 two part connectors seem to generate high enough forces between the mating bits of metal to produce a bit of micro-welding, and seem to stay reliable forever.
-- Bill Sloman, Sydney
On a sunny day (Fri, 14 Dec 2012 04:00:26 -0800 (PST)) it happened Bill Sloman wrote in :
Yes, you may be right, I have been looking for suitable boxes, this is sort of my reference paper on the design ATM:
loman
.ni=
bu=
oug=
g w=
en =
alu=
..
n page 107,
Looking at their picture of their box, it looks as if they over-killed on metal thickness to save having to think hard about electrically bonding the body of their box to its lid. If you've got access to a numerically controlled milling machine, chewing up aluminium blocks is quick and easy, if not cheap.
Three mm of metal should be plenty if you can do a bit better on the bonding across the joint.
-- Bill Sloman, Sydney
On a sunny day (Fri, 14 Dec 2012 06:39:52 -0800 (PST)) it happened Bill Sloman wrote in :
Possible, I may consider to use 2 boxes inside each other like a Russian doll. One small one housing the RF with some feed through caps for power, inside a bit bigger standard eurocard alu box with 1 mm thickness like I use for many other projects. If that test fails then I will try something else. Boxes come from all different sources, maybe here for Christmas.
Mmm, (2n + 1)/2 wavelength (1/2, 3/2, ...), not quarter.
But slot effectiveness drops off sharply with width. And that's for a slot in zero (or near zero) thickness. What if you overlap the seams? Now the slot is effectively very deep/ What if you overlap by 1/4 or 1/2 wavelength? Now it's deep by a magic number!
If you actually insulate between sheets, an overlap of 1/4 wave should be ideal, while an overlap of 1/2 wave should act like nothing at all (making a good, if high-Q, slot). Be mindful of the wave velocity in such a joint, which is probably less than c, so the actual width probably won't be f/4c but something shorter.
If you ensure contact just at the edges of the overlap, and nothing inbetween, you'll do quite well already; any leakage will proceed best in
1/4 wave widths, since a 1/2 wave is shorted at both ends. In either case, the mismatch will be quite large (i.e., reflective).That adhesive aluminum tape stuff is probably quite good for this. Even if the metal part doesn't make contact, the adhesive probably has enough dielectric constant, combined with a lot of losses, that if you spread it over a lap joint, whether or not that joint itself makes continuity, the "opening" of that slot will be thoroughly shunted by the tape and adhesive. Again, it would be handy for the tape to be cut into 1/2 wavelength widths, centered over the joint.
If this cage is needed for 2.45GHz or something like that, great; if it's for more than 30% or so bandwidth, you'll have to be more aggressive at trapping those resonances.
Tim
-- Deep Friar: a very philosophical monk. Website: http://seventransistorlabs.com
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.