Welcoem to posts!!
in the future - u will be able to do some more stuff here,,,!! like pat catgirl- i mean um yeah... for now u can only see others's posts :c
I had to buy another text book (David Griffiths excellent "Introduction to Elementary Particles") to help me understand a magnificent THREE HOUR deep-dive video by Richard Behiel. I first studied electromagnetism back in the 1970s, then again ten years ago when, just for fun, I did a degree with a lot of Astrophysics and Electromag and Quantum Woo and Hard Sums (or "Math(s)").
Now I need to refresh my poor, aching, ancient brain so I can understand a Youtube video. My life is a total train-wreck, isn't it! It's nice to meet old friends like the Euler-Lagrange equation and variational calculus that I'd kinda lost touch with. It's hard work, but I guess it's supposed to be. As Richard says, "Ask not for easier equations, but for stronger coffee"
14 - 3
More light reading for those sleepless 3 AM overactive brain times. I am not prepared to divulge the reason for this strange diversion from the straight and narrow path, but Marx generators, large capacitor banks, vacuum pumps, valves, gauges, chanbers, kilometres (miles?) of copper wire and a strange compulsion to try to outdo that young Styropyro feller are all involved. Sorry no video this week, it's been IT Security Audit time at the Day Job. Such fun.
65 - 11
A little light bedtime reading. A chap needs a particle accelerator in his life now and then.
122 - 28
First Chips on the new SYIL CNC! Next video drops at 15:00 UTC today (4 pm CET, 3pm GMT, 10 am EST, 7 am PST)
73 - 2
Question for folks who do jewellery or other fine-detailed investment castings. Is Polymaker Polycast a good choice for burn-off and layer-line smoothing in an investment casting? Is their solvent any good? Any better alternatives? Any really clean PLA materials that don't leave ash after burn-out?
Any other suggestions for how to make excellent surface-finish precise investment castings? After viewing @lundgrenbronzestudios video about comparing PLA and Polycast at https://www.youtube.com/watch?v=iSNH8... it looks like neither is any use at all for the sort of precision work I'm doing.
Is there a soluble or burnable filament that can give useable quality? How about resins, anything useful there? Is slurry and sand multi-dipping then firing a better technique than simple degassed investment casts with vibration and vacuum? He's using a simple vacuum casting technique, would I get better results with centrifugal casting, or even pressure die-casting? How about the metal choice? I need something highly fluid/low viscosity, that either has surface conductivity close to copper or silver, or can be electropolished and plated with copper/silver/gold to give at least six skin-depths at 50 GHz without any nickel or other terrible RF conductors.
So far, it looks like machining from solid brass, then electroplating, is going to outperform any sort of 3D print/cast process by at least an order of magnitude in terms of surface roughness, but the requirement of split faces to permit machining causes just as many issues as the problems of investment casting or metal 3D printing do. Electroforming on a solid aluminium mandrel looks like one of the only viable home-shop techniques above 50 GHz, amirite?
11 - 18
Sneak peek at a fixture I make in next video, due out on Friday, but more important, a question for the Machining & Microwaves community!
I'm trying to think of a project where a METAL 3-D PRINTED mechanism would make possible some sort of functionality that can't be achieved by conventional machining. The huge problem is always the same with any 3-D printing system and that's getting internal voids with no layer artifacts and surface roughness of a few micrometres RMS to avoid scattering and grating effects. several of the fab shops which do metal 3-D printing have asked me to do sponsored videos, but without an engineering application that makes sense, it's hard to develop a project which isn't just "look at this thing that we can make that's *almost* as good as we could if we used traditional machining."
I'm not very keen on that idea. so I'm really looking for suggestions from the community about what type of RF structures could be printed using common metal sintering processes that could not be done on a lathe, or that will outperform parts made in that way. The beautiful bronze centred handle that Tom Lipton had made for his sensitive tail stock drill chuck was the first example I saw where a metal 3-D process produced a result that was potentially cheaper and better than a machined version. In that case, I think it would've required a five access mill to get the organic shape, but for the 3-D process it was trivial to achieve that.
If the machine used a vacuum (or inert gas) sintering process that worked with copper particles, and the affective surface resistance was similar to that of machined copper, and shrinkage was well defined and controlled, then it might be possible to print some wave guide combiner components. It would be very difficult to beat investment casting though. Those traditional techniques have a 50 year plus head start on additive processing, so it will take some time to displace them.
I would like to try 3-D electroforming, which would produce entirely solid copper forms in a similar manner to those that are 3-D printed, except it would be far too time-consuming to use single electrodes. I've seen an example using a matrix array of electroforming elements which then copper onto a substrate using that 2-D array of electrodes. printing ceramic 3-D form seems to me a much more viable solution in the short term and it certainly possible to make 3-D forms from ceramics that are impossible to machine or cast, so I don't imagine it will be too long before we have systems which can produce 3-D printed, aluminium copper or silver millimetre wave components.
Ideas please! There are sponsors out there waiting for a decent 3D metal printing project.
58 - 18
Best Compliment EVER!
Well folks, yesterday was in the top ten Weirdest Days of My Life (so far). I found myself in a church hall in Camberwell, South London, UK, on a stage in front of a red velvet curtain with TWO of the replica historic "mechanisms" I've been machining for the last few months as part of Project Narwhal.
A BBC TV crew were filming me being interviewed by the most excellent Professor Hannah Fry (Numberphile, StandupMaths,TomRocksMaths and all the cool channels, plus radio and TV etc). Then we examined the "mechanisms" and Hannah did an explainer about how they work.
During the final segment, the Prof told me I am THE MOST MASSIVE NERD SHE KNOWS. My work here is done!
184 - 19
Here's some parts I made this evening for Project Narwhal. I don't think I'm giving away any clues here!
76 - 29
ARDUINO ADVICE NEEDED PLEASE!!
For Project Narwhal, I need to implement a control system that synchronises two rotating machines at 2400 rpm to within a degree or two of rotation position, with only a few degrees of jitter. I can only use a magnetic pickup, and eventually I'll need to implement this using thermionic devices to make a rudimentary controller, but I need a very quick fix to get this working asap. There's a lot of rotational inertia and aerodynamic drag, so it should be simple enough to get the control loop stable. There's quite a mass of metal being spun.
I have some new Arduinos and a chunky DC motor drive shield that can run the 48 volt motor at up to 6 amps with PWM, so I'd like to use that platform, but it must be nearly 10 years since I last used anything Arduino-related.
Two questions:
1) Has anyone had RECENT EXPERIENCE of a decent PID motor speed control library for Arduino that can drive a PWM shield that they would recommend? I need to be able to take a pulse train from a simple mag pickup (with multiple magnets/coils if needed) on the motor shaft and sync the motor to an incoming pulse train from the remote motor (over a radio link or cable or fibre)
2) To aid debugging, I need a fast 8 bit ADC/DAC shield or off-the-shelf board that I can use to sample an incoming varying signal at about 50k samples per second, preferably a bit faster, and a DAC that can settle to 8 bits in a couple of microseconds. Again, has anyone had RECENT EXPERIENCE of success with such a shield or board that can be bought from a fast supplier like Farnell/RS/Mouser/Digikey or the UK robotics and maker shops?
I can't wait for anything more than a couple of days if I'm going to hit my deadline. I'd prefer it if there was an interface like analogRead() but at least 10 times faster. I don't need more than 8 bits, but if there's a fast 10 or 12 bit, I'll just use the MSBs.
The old PID libraries and ancient hardware I knew about have all vapourised or morphed into something else as a result of Progress.
Context: The original version of the 1920s "thing" I'm building used a synchronous AC motor on the local machine, with a set of magnet/coil pickups on the remote rotor, and sent the pulses from those down a sync channel, which was then amplified by a multi-stage vacuum-tube power amp that drove the local AC motor. Neat.
I don't have the luxury of an AC motor of the correct size and power at the moment, so I'm stuck with a brushed DC motor and have no time to get a three-phase AC motor and drive implemented, but I think I can hack some code to give me at least a fighting chance of getting the thing to synchronise. The old PID library was by Brett Beauregard if memory serves, but chances are there's something better that "everyone" uses and knows about. Except me.
I could perhaps make a PLL style hardware controller to lock the motor to the incoming pulse train, but time is of the essence, so pre-rolled and reliable code libraries are infinitely preferable to get me out of this deep hole. I still can't type well or use a screwdriver or soldering iron safely, but I can use voice input to cut some hacky kludged code, hence software solutions and prebuilt modules are hugely preferred.
Sorry I can't going into detail about the precise application as Project Narwhal is subject to an embargo until the program gets broadcast on BBC TV.
Right, back to making microwave radio bits with my fabulous killer robot.
22 - 59
Project Narwhal is progressing like an out of control freight train on a steep incline. Here's a small element of the project. I've suppressed 176 of the small components so as not to give too much away! There are two of these. The top plate is 250 mm (10 inches) square. Sorry I don't have a banana for scale. There is a ludicrous number of screws and more than 140 springs in the full system. Oh, the bevel gear transverse shaft is missing too, so make that about 185 components invisible in this small subsystem. As well as the second one of these, there are two similar smaller ones, plus motors, drive shafts, lots of electronics, and a mystery wrapped in an enigma. Deadline is closing in fast, I have no hope of getting it all done in time, but that's normal for me. I have a large bin of aluminium chips and a table piled with boxes full of parts. Colour-cycled and raytraced render here:
34 - 17
Machining and Microwave/mmWave Radio experiments. Spy devices. 3D printed antennas.
I design and make parts for experimental radio systems, scientific equipment and talk about unusual and cutting-edge topics related to high frequency electronics, antenna systems, waveguides, combiners, cavities, filters, digital comms, CAD, E-M simulation tools and microwave circuit design.
I hold a licence which lets me transmit between 136 kHz and 1 THz. I carry out experiments including RF, plasma, high voltage, hard vacuum, precision machining, metrology and electrochemistry. I've worked with commercial sponsors to produce collaborative vids on specific products and services within my subject area and welcome business enquiries. I've also appeared on BBC2 TV with Prof Hannah Fry, demonstrating my replica of the Great Seal Bug spying device.
Patreon patreon.com/MachiningandMicrowaves
Buy me a coffee??? ko-fi.com/machiningandmicrowaves
Neil Smith, Yorkshire, UK
AWIN