Machining a 10 GHz Lens Antenna from Rexolite 1422 Dielectric Polystyrene 

Mmm, coffee...

You have great taste in RU-vid

@@sambrose1 thanks. See also Clickspring

I just wish I had more time to edit and post videos!

@@MachiningandMicrowaves Do what you can. We will watch, learn, and enjoy.

I'll try to find time to do some better explanatory videos. This clip is for the lecture I'm giving on Sunday. The recording of the lecture should appear on RU-vid at some point. I'm rubbish in real time/real life though, lots of arm waving and staring into space trying to think of the right word while I should be explaining stuff. Nominal Aphasia. I've always had it, which is why I prefer to do voiceover videos. My blooper reels would be even more hilarious than Kurtis from @CuttingEdgeEngineeringAustralia but with less sweariness and more Chihuahuas

@@MachiningandMicrowaves your explanation actually are very good. Radio frequency filtering and engineering is just so far out of the realm of what I know. Like explaining trig to a 4th grader. I like listening and pretending I'm smart lol. Keep up the great work.

@@evzone84 I like those videos where they get a world-class expert to explain something to a youngster, a high schooler, a college student, a grad student and another expert. It would be an interesting challenge to do perhaps a four-level explanation. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-QcUey-DVYjk.html

There's a surprising number of radio hams who, like me, don't actually enjoy the "talking to strangers on the air about nothing of note". My late wife used to give me one of her Looks and ask "But do all those people have anything interesting to SAY to each other?". The hobby for me is about making things, testing stuff, exploiting weird and unusual propagation mechanisms and helping other folks to do what THEY are interested in. I'm always attracted to the edges of things, so the 8270 Hz band and the Terahertz zone and longwave infra-red is all exciting right now. Yes, 8.27 kHz. Microwatts of effective radiated power at ultra-slow data rates and bandwidths of 80 microhertz but still able to get signals across the Atlantic. Also working on 470 THz red light using Phlatlight ultra-bright LEDs running 27 amps at 3 volts into large Fresnel lenses for communications over 150+ km. Cloudbounce with red light is something else I'm messing with, rain scatter on 47 GHz, bouncing signals off aircraft to get microwave contacts into Europe over 600 km and more than 1500 km via passive reflections from the International Space Station. Bouncing radio signals off the moon is my main focus at present. That's so close to the limitations of Physics, and the sheer thrill of hearing your own doppler-shifted echoes off the moon after the 2.5 second light-travel-time delay takes some beating.

I wish I had more time to make videos, but Project Swordfish is eating all of my spare time. It's going to be really really cool if it all works out, but the stress level is extreme! Also I can't discuss what it's about, which is a nightmare, because I have SO MANY questions....

@@MachiningandMicrowaves' Project Swordfish'😱- sounds like an 'off the books' energy weapon prototype - if putins head explodes inexplicably I'm going to start wearing a tinfoil hat😜

@@Mwwwwwwwwe ... I never take mine off, even to sleep. You never know when they will get you... 🤯🤖👍

@@wizrom3046 be careful... project jabberwocky can vaporise tinfoil hats- best to upgrade to a graphene/ceramic composite with earthing via a gold lightning rod

@@Mwwwwwwwwe ... got it! Will do right now. 👍

It's remarkably stiff material and I've never managed to rip it out of the chuck even when I got a bit over-enthusiastic. It chipped out when I was roughing, but the chipped section was removed when I made the taper. When I'm making a batch, I set the compound the other way and turn the taper backwards towards the chuck, roughly to size, but this was a one-off done in a hurry, so I just went for it

I've been thinking about making some combiners/splitters machined from solid dielectric waveguide, as the only was of making the intricate internal features is by 3D printing and investment casting. I could machine a negative of the space, then coat it with a layer of conductive material and electroform copper over that, but it's also possible to make waveguide just from a dielectric with no metallization. I'm thinking about combining multiple 24 GHz amplifiers without using magic-tees or cavity couplers.

[blush]

That would be nice!

I'm working on some new lens designs so there will be a proper length video about those at some point.

I've just been reading a long and complex thread about Kalman filters in phase locked loop control circuits and how to control 1/f and flicker noise. I love the time-nuts forum because they don't make any attempt to simplify the language, it's just the hardest of the hardcore time specialists riffing away and having fun. Spectacular stuff when you are looking for precision beyond 1e-13 and controlling the jitter and atmospheric variation of GPS to the point where seasonal ground heave is one of the major noise sources as the subsoil warms and cools. Labs with 0.1C temperature control, folks with multiple caesium atomic clocks and even hydrogen masers in their basements. I only have Rubidium clocks and a number of double-ovened Morion oscillators mounted in anti-vibration enclosures inside temperature-controlled and insulated rack cases in the three 42U data racks in my study., plus a number of GPS units. My timing is pretty much GPS quality over periods of hours, but short term it's perhaps not much better than 0.1 ppb (1e-10). That does mean I can generate a signal to within 1Hz at 10 GHz, but it feels rather sloppy and imprecise to me! I did a load of work with MECL III back in the 1970s, clocking it at 280 MHz and that felt ridiculously fast. Now I'm reading an article about 100 Gbps over ultra-wideband radio. How things change...

That's the plan, but time constraints are a huge problem as I'm working through a backlog of 13 months of machining and electronics jobs, some of which are just not photogenic or interesting. Also, it's 2.30 AM local time and I have to get up for work in less than six hours! Bedtime...

I watch everything @thesignalpath produces. Super channel

It's interesting to map out the signal levels in two dimensions if you can arrange that as well. You need to think about how to handle reflections from objects, buildings and the ground, so if you can get the thing in the clear so there's at least 200 wavelengths of free space around it and it's at perhaps 5 metres above ground, that helps to get the real pattern measured. Of course you might be more interested in knowing the pattern when it's working inside an environment, but that effectively overlays the pattern of the receiving system on to the reflections. Someone I know is using drones to do this at lower frequencies, and another is using a helicopter to measure digital broadcast radio antenna patterns at around 200 MHz. I use two pneumatic masts, with the antenna under test on one and the sensing antenna on the other. I can pump them up to 13 metres above ground to ensure there are no reflections. I use tape rulers to measure the height of the transmit and receive antennas and a calibrated rotator on the transmit antenna to test the pattern at multiple different heights. It's huge fun. If you want to get REALLY fancy, set up an array of 25 BT receivers (5x5), then sweep the transmit antenna around and see if you can see the sidelobes passing over the sensor array when you play back the signal level as the colour or brightness of a display. If you can get hold of even more, you can almost make a "camera" for 2.4 GHz and try diffraction and lensing and get up to all sorts of high jinks.

Rexolite is more brittle than Perspex but takes a nice polish. The lens fits in the end of a round waveguide tube. Waveguides can carry microwave energy with extremely low loss, but when you want to illuminate a parabolic dish, you need to form the beam into a nice clean lobe without any responses from off the main axis. Target it to illuminate the dish with just enough of the available beam power to give good efficiency without any spillover. First issue is to get from the air dielectric into the plastic without reflection mismatch, so the rear of the lens smooths the transition. Then the tapered section, which has a relative permittivity of about 2.6, carries the wave at a lower speed than the air and acts to focus the energy into a beam with about 100 degrees beamwidth at -12 dB (about 7% of the peak). The dimension is only about 0.7 of a wavelength, so it behaves like an optical lens that's about 300 nanometres diameter. More of a diffraction than anything. I'm working on a project with graded index lenses at the moment. I'll try to do a proper video on these lenses. A similar finished antenna feed is on my blog at www.g4dbn.uk/?p=1644

Nice thing about them is they are totally sealed without needing a radome. They have a good pattern, but there's some sidelobes in the H -plane that you can't get rid of with a rotationally-symmetric lens. Usual approach is to mould the dielspike with knobs on the sides, but machining that is tricky, I really want to try a smaller diameter tube totally filled with dielectric to see how it performs, but my PC runs out of memory doing the simulations as it only has 16 GB of RAM. I checked with someone this evening who runs 256 GB for his OpenEMS simulations. I need to go LARGE! Second good thing is that if you make one for 5.7 GHz and one for 10 GHz, they can mount very close together so you can run two bands just by offsetting he dish by 2 degrees without compromising the pattern too much. I run then over and under and use the elevation drive to change between bands. I still prefer a Pickett-Potter dual-mode feed for best performance and sidelobe control. I'll be making some more of those in an upcoming video too.

@@MachiningandMicrowaves Oh nice, that elevation for band change is a nice idea.

Rexolite is quoted for use from -60C to +100C. I don't know of any polarisation rotation or other effects. Spec sheet is at www.rexolite.com/specifications/ I think it would be fine with a stepped septum or dual probe RHCP in the right sized tube at 8 GHz

Not sure about long term UV survival

I think she's keeping her powder dry, so she'd ready to issue a few salvoes in the next video!

I feel like that when I'm watching the most excellent @SarahNTuned when she's working on cars. I recognise some of the words and the occasional sentence, but somehow it doesn't really matter that I don't know what she's talking about some of the time. I'm going to try to emulate @Mathologer and @Numberphile to see if I can explain complex concepts using images and diagrams and not too many equations - except when they are stunningly simple and elegant. I don't want to dumb things down though. I'm still trying to work out how all of this YT thing works. For someone as radically introverted as I am, it's all scary as h*ck.

@@MachiningandMicrowaves Oh this wasn't me dissing your content or the way you do things. It was still very enjoyable and I got some knowledge about a lot of things I never knew would be fascinating. If you don't want to dumb things down (which I thing would actually be a negative in this case) or plaster diagrams and spend more time explaining than creating, you could always leave some sources and those diagrams in the descriptions for people who do want to delve deeper into the technical side of what you do.

I get it from Specialist Engineering Plastics in Grantham, UK. They only do 6 foot long round bars, no short lengths. What size and quantity do you need?

@@MachiningandMicrowaves I need a piece of rod diam 20mm by 100mm long

@@stanjobar I have some offcuts of 25.4mm dia that I can cut to length and send to you if that helps. email me neil@g4dbn.uk

With a loss tangent of 0.00066, it would be dissipating around 1 watt for a kilowatt of RF. I think a watt might be fine, but the thermal conductivity is really the limitation. If hotspots were to form and not be able to shift the heat out by conduction , that might cause problems. I don't have a travelling wave tube amplifier, so the most I'm likely to be able to generate at 10 GHz is about 80 watts of carrier from a pair of Wolfspeed CMPA801B030s, so I'm nowhere near the limits. It should be able to take 10 kW pulses though at a few percent duty cycle.

His would have been for Long Wave and weighing 473 pounds so he had to use all of his cranes at once!

It's getting very difficult in the UK with safety regulations about what you can touch and use while driving, plus it's painfully hard to install radios in most modern cars, and they are so full of electro-smog that getting good reception is tough. My old motorhome had a diesel engine and almost no electronics, it was a fine vehicle for radio contests and expeditions. Even my 2013 Mercedes diesel is electrically noisy and tricky to fix a radio into. I use a trailer to carry a 12 metre inflatable SCAM mast, 230V generator, deep cycle batteries and compressor these days, and set up the radios in the back seat.

@@MachiningandMicrowaves We used to have vhf radios in our service trucks. Electronic diesels, good grounding and a line filter helps a lot!

@@NavySturmGewehr Modern cars are a nightmare unless you turn them off completely. Warm and comfortable inside, so you put the antennas outside in the weather and get on with the important "playing on the radio" bit in comfort.

@@MachiningandMicrowaves Years ago on our 7.3L Powerstroke diesel service truck, if used the handheld, there was a span of rpm that the radio would pick up. When I installed a radio into the truck, improved the trucks grounding, the installed radio did not pick up that artifact. The hand helds still did. vhf, but I don't remember the frequency.

Spookily, I had an enquiry from a company who work with video machine intelligence and intruder identification about high power 100 GHz steerable antenna systems. I'm assuming it was for "area denial" purposes. I think it was for slightly more sinister purposes than my original idea, which was to stop marauding cats from digging up the flowers in my garden by giving them a sharp blast of microwave energy! However, I have two Chihuahuas who find cats highly distasteful and chase them away. Fully automatic, with machine vision, audible alerting, able to work behind trees and bushes, no batteries required. the only microwaves involved are those I use to defrost the raw chicken wings they eat for supper.

This is pretty dialled-back on how I usually talk to other specialists. I'll try to keep a mix of serious technobabble and explanation using sidebars like Quinn does.

Heh heh....

My electroforming equipment should be arriving next week, that's going to be an interesting new technique for making things that you an't machine or 3D print. I'm a bit worried about whether the supply of Preperm 3D printing filament is going to resume after the company was taken over recently and the online shop was closed. I wanted to try variable refractive index metamaterial printing. Fingers crossed....

Noooo! He's fallen to the CNC! All hope is lost! Lol. Though it looks like I've got a bit of homework. I'm not familiar with material that is variably refractive. I mean to photons sure the reflector can be dulled or oxidized unevenly. But that's not been a benefit to anything I've seen. Having a plastic/polymer that is variably refractive to electron waves on the other hand... is it used in conjunction with turning crystals or an alternate to? Is that even what's going on with the refractor?

@@daretodreamtofly3288 Imagine a sphere of glass. It doesn't focus light very well, lots of distortions. Now if you could vary the refractive index with diameter, you can fix the aberrations and make an effective lens. Not practical. Now radio waves and light waves are the same thing. Still photons, still electromagnetic waves, just at a lower frequency. Same as gamma waves are higher frequency. Lenses behave similarly so long as they are large compared with the wavelength. If there are air gaps inside the material that are less than about 1/20th of a wavelength, they don't have a large effect on the performance of the lens. Now, for light, that means less than 20 nanometres, but for 47 GHz radio, it's about 0.3 mm. Next, imagine 3D printing a 100mm diameter ball from low-loss plastic with 100% fill. It would be a poor lens. Now change the fill ratio using smooth swirly shaped voids and vary that fill ratio continuously with distance from the centre of the ball. If the voids are less than 0.3 mm, you've just made a graded-refractive-index Luneberg lens. Now take is a few steps further and change the shape of the printed object and vary the density and effective refractive index, and throw in different dielectric materials with different relative permittivity and suddenly you are in a completely new world of antenna design, where you can customise the coverage pattern of a lens antenna. Exciting times we live in.

Then it goes totally next-level, with metamaterials that have a NEGATIVE refractive index, where the phase velocity is greater than the speed of light. A convex lens made of that is divergent. Now that is spooooky.

@@MachiningandMicrowaves now that I understand a little about. And I've seen two running thoughts on this. One is that on our level of "normal" physics that it only appears to be traveling the speed of light. But that supper connectivity comes into play at some point. Sighting that the "speed of light" is in reference to the speed of a photon in q vacuum. I mention this not for you sake but just to give you reference on my understanding which could be wrong by all likelihood. But as for the second idea on "faster than light" transmission is down on a quantum level the particular is moving with so much energy that our universe "forgets" where that particular was and just puts it in the place approximately appropriate for it as it had been moving. I find that description way to basic but it's better than "the particle pops out of existence and then back in" then there's also the idea that the particles are just a function of Quantum strings and as that move faster than the speed of light that they revert to individual strands of energy then when they slow back down recollect into matter