Project ideas for 5 axis milling?

[Marsupilami]

Hi Folks,

And now for something completely different: what RF/mw related stuff would you cut on a small 5 axis milling machine?
I have plenty of 3 axis experience with enclosures and stuff but I was always fascinated by the idea of making something more 3 dimensionally complicated and now that I have the chance can't think of anything good. It can't be too big and it should do something measurable under 20GHz. I was thinking like some cool coax launch to horn or I'm wondering if there are more exotic cavity filter concepts with perpendicular cavities or something along those lines. What would you guys make?

[A.Z.]

what about a 3D fractal antenna ?

[cdev]

I don't realy understand what you mean by 5 axis. Maybe could you upload a picture of something you'd make and why you'd make it differently??

[Marsupilami]

what about a 3D fractal antenna ?

Ahaha I didn't even know this is a thing. Fascinating. Also great idea. I have to find a pattern that's machinable. This one liked is 3D printed and plated but it looks doable. Great tip, thank you!

[Marsupilami]

I don't realy understand what you mean by 5 axis. Maybe could you upload a picture of something you'd make and why you'd make it differently??

5 axis machining is when besides moving the cutting tool in 3D (x,y,z) relative to the workpiece you can also rotate in 2 axes so you can reach from different angles without multiple setups and/or cut continuous curved surfaces. The typical example is turbine blade.


I don't have anything specific on my mind at the moment, that's why I'm asking. I know that there are weird curved surface stuff in RF that might benefit from being machined from a single piece. I was maybe thinking exponential horn antenna, but probably I don't have the work envelope to do that in appropriate size.

[cdev]

At the beginning there is what looks like the agitator from a washing mashine..Or a Thai restaurant's shell design.. I could see having such an ability as being kind of useful.. kind a sort a

[coppercone2]

this stuff is probably ALOT easier to make then to design, especially the fancy patterns.

custom (often parabolic) reflectors come to mind for something that is somewhat conceptually easy and potentially useful but difficult to make with conventional tools. That is the first design that comes to mind that makes 'nice' use out of a 5 axis machine. And they actually look good too. plenty of machinable things here that are lower design skill/effort to entry

https://swamphen.co.uk/new-blog/2019/4/15/lets-talk-about-reflector-antennas

for some of the analysis behind reflectors (not as simple as one might think)
https://iopscience.iop.org/article/10.1088/1742-6596/2214/1/012028/pdf

*also think about 'nice' brackets/supports used to sturdily mount that shit, not like stamped sheet metal bendings, this could also be a complex 3d mechanical problem when weight, wind resistance and aesthetic is considered. An interesting reflector design might make RF communications towers stop degrading property value. Boresight adjustment systems that are sturdy and look nice for instance.

I think you might find more interest in optics then RF for small devices that are easy to design but complicated. I.e. metal flashlight reflector, not plated plastic flashlight reflector. Once you have multiple LED's and you want to make a few different beam widths (i.e. floodlight + beam light) that are 'nice' its already hard IMO

That's the MW build problem, what to make, and why. Easy to get caught up in esoteric design that is already designed and probably forgotten about in johnson space center engineering building aux storage 5.

There are probably cavities that deserve making too, but you need someone to design em... for 'screwing around' with no purpose, only a parabolic reflector comes to mind, for those tired of scraping dishnet logos off things. There are probobly interesting things that could be done to make antennas more wind resistant and lighter too with advanced design and machining. I think microwave is still really focused on MMIC circuits more then anything right now.

Also, advanced audio designs (think bose) also come to mind for making a multi axis machine useful. Cast iron speakers, so forth. Those could possibly make money off etsy or something.


Antennas are interesting because they get architectural. Consider the 'circus' that VLF transmitters are.


I bet there are people really interested on how to make sky scrapers look less bad because of comm equipment.

Maybe we can have that nice 'bladerunner' skyline, because it sure as hell is not gonna look at that good with how antennas look like nowadays lol (and how fast they are multiplying).. and that was supposed to be a dystopia.

[A.Z.]

what about a 3D fractal antenna ?

Ahaha I didn't even know this is a thing. Fascinating. Also great idea. I have to find a pattern that's machinable. This one liked is 3D printed and plated but it looks doable. Great tip, thank you!

check this

https://wcares.org/fractal-antennas-for-ham-radio/

[cdev]

That would be a broadband antenna, most certainly. I saved a copy of the nice rendering. Actually they all would be. I have a PCB layout for an archimedian spiral antenna. That makes a good antenna but its surpringly expensive. A two arm Archimedian spiral makes an excellent GPS antenna, a four arm one makes a truly spectacular multi-GNSS antenna.. One of the best GPS antennas around.. You need a 90 degree hybrid phase shifting RF MMIC. It pulls in a lot of signal too. Enough to use your GPS in your basement. And have it work.

[cdev]

I think you might find more interest in optics then RF for small devices that are easy to design but complicated. I.e. metal flashlight reflector, not plated plastic flashlight reflector. Once you have multiple LED's and you want to make a few different beam widths (i.e. floodlight + beam light) that are 'nice' its already hard IMO

That's the MW build problem, what to make, and why. Easy to get caught up in esoteric design that is already designed and probably forgotten about in johnson space center engineering building aux storage 5.

Those must be interesting students. I dont know how many photonics students are in engineering schools rightnow. Its an interesting field. Sort of merges electromagnetic theory and optical phenomena.. and boy is that a rabbit hole which one could find yourself sucked into.. A strange one..

I like that stuff.. Quantum light phenomena..

It would be very nice to have a high resolution 3D printer to make shapes which you could turn into real world objects by 3D printing them out of plastics or perhaps wax and then casting them in metals using something like sand,, perhaps?   Mathematically originated 3D shapes often make beautiful sculpture.. like minimal surfaces

[cdev]

what about a 3D fractal antenna ?

My guess is that you could use flashing copper to implement a fairly good working approximation . I think I have even heard something to that effect. It would be like a folded Sierpinski triangle? Its worth a try..It would likely be fairly fragile.. but it would be interesting.

[Coordonnée_chromatique]

hello, the isostatism will be a nightmare, you should choose a metallic SLS process.

[coppercone2]

I would actually just spot weld sheet metal pyramids together for that I think, then fill in with a lower flow braze.. it should approximate that shape OK but I see how for dimensional correctness without filing insanity then you would need to do a casting

That would probobly need to be vacuum cast. The easiest and strongest casting to do that might be Zinc, I think it plates easier then aluminum, melts at a lower temperature, but you need to deal with fumes. Casting it in solder might be OK since I have seen solder get copper plated, you could theoretically do that shape with a plating pen, but I have had poor experience with that, its a skill you need to learn. Also, very tedious.

And you need a kiln to fire those 3d printed waxes, they won't go at oven temperatures IIRC.. so thats a big barrier to entry.

Maybe a dip coat in shield paint would work for a 3d printed shape if you use conductive glue for the connections.. is there a STL file? If someone made a STL file available for download then we might actually get some data lol

[cdev]

We have some popsicle molds that look as if they were easy to make, they are made out of silicone rubber..It may be possible to make a 3D model of the right shape.. to cast using mandelbulber2  Were it not for the holes,  I bet one could do that without the ability to EM model it.. as it looks like the kind of antenna thats determinable by its geometry..If one could create a mold layer by layer that would be best.. Then connect them..They make Very forgiving antennas.


I make one that looked like two toilet floats.. two balls, That worked well.. I ended up making it out of aluminum foil, basically.

[LaserSteve]

KS-15676 in miniature for x-band.

[LaserSteve]

Metallic lenses of the Winston Kock style. Long forgotten but should make a comeback. Bell Labs Innovation.

Steve

[coppercone2]

that is a meta material lens of some kind

[Marsupilami]

Hey All,

There are some good ideas here, thanks for everyone. Work happened meanwhile but if I'll have some play time I'll dig into which one of these are realizable.

Thanks!

[cdev]

Could in theory be used for high gain UHF antenna?