Talking to myself on 10 GHz with a Software-Defined Radio

[@hughthompson8018]

I have absolutely no idea what this is about or what 95% of the words mean, but I absolutely love it! Thank you very much!

[@someoneelse7629]

I know 95% of what he's talking about but since I'm not brittish I struggle a bit with some of the wording.

[@edgeeffect]

Writing your call sign in the SDR waterfall is BRILLIANT! I don't know much about radio so maybe everyone's doing this.... but I'm impressed!

[@MachiningandMicrowaves]

I just wrote a script to generate the tones at the right frequencies and levels, but there's a mode called Feld-Hell (Hellschreiber) which uses a similar approach to print text in a slightly slanted format across a waterfall. Silly things to do in the middle of the night....

[@danoneill8751]

Mental note: if putin sets fire to the atmosphere and I win the ark lottery and get a space for me and mine, must give up space of oldest son and give this northern bloke his spot - pretty sure he'll be needed to rebuild civilisation.

[@HAL_9001]

So the synopsis is: when you're up too late to talk to anyone else, talk to yourself?

[@MachiningandMicrowaves]

Got it in one. There is a Cunning Plan for a large network of receive stations similar to the ADS-B aircraft reporting systems. That would enable some very interesting wide area propagation experiments. Big project though.

[@bobaloo2012]

Talk about difficult signal paths, the first time I sent an "O" in morse, let up the key and heard it come back into my receiver 2.5 second later was one of the highlights of my radio careeer.

[@MachiningandMicrowaves]

I've never been loud enough on 144 MHz to hear my own echoes, but I've worked a few of the bigger stations on there with a single non-elevating yagi. The 3 metre solid dish on 10 GHz should work well even at 22 watts. I have some GaN MMIC SSPA devices that should do 37 and 45 watts on 3cm, and some Wavelab units that I hope to pair up and run at higher voltage to get 2 x 4 watts on 24 GHz. Also I have some 80 watt GaN HEMTs for 5.7 GHz, and plenty of power on the lower bands, so there should be fun times ahead. I don't think I'll get the 70cm array built this year, but I've been laying the prepwork for an 8-yagi EME array with welded open-wire feeders and a big watercooled NEC PA that can run full legal limit 24/7 without breaking sweat. Looking forward to hearing my first moon echoes for sure.

[@someoneelse7629]

@@MachiningandMicrowaves I have heard some of the big guns on 144MHz on my single Yagi, but never had power enough

[@MachiningandMicrowaves]

@@someoneelse7629 First one I worked was 5 degrees before moonset and it was Gary KB8RQ, 200 watts to a 9-ele, no elevation. He do 99% of the work: www.spasalon.com/kb8rq/

[@jamespray]

I for one really appreciate a look into the application of all these machined gubbins! I came here to expand my repertoire of "things I've at least heard about in passing" and by golly I'm getting my money's worth!

[@MachiningandMicrowaves]

I'm rather looking forward to seeing the next one, it looks interesting. I mean, I should know what's in it cos I kind of scripted and filmed it, but there's a mental disconnect after I press the PRODUCE button and it's almost like watching someone else's videos. Is that weird? Yeah, it's weird. Hey ho. Weird is good.

[@jamespray]

@@MachiningandMicrowaves I think it's a good sign when you like your own creative output enough to sit through it like your audience would!

[@someoneelse7629]

All of you that are slightly intrested in the radio side of this, dive in, get a license and start fiddeling around, to build antennas does not have to be this complicated and high precicion, just start on another band, on 2m you can fabricate antennas with an axe and scrap metal and still be within tolerances. On shortwawe, it is even less critical. Microwawes are a challenge, and tolerances are small, so that's not as beginner friendly.

[@MachiningandMicrowaves]

I've held a ham licence since 1973 and it's given me the opportunity to carry out an enormous range of experimental work including moonbounce, meteorscatter, passive reflection form the International Space Station, aircraft reflection, rain/snow/hail scatter, troposcatter, tropo ducting, sporadic E, auroral curtain reflection, chordal hop ionospheric refraction, hearing my own transmitted signal after it's been right round the world or across to the opposite side of the auroral oval, communication via amateur satellites, TV transmissions, radio data networking, ground loop antennas, balloon-supported and kits-supported antennas, bistatic radar experiments and antenna lenses. That experience has also let me into the work of Deep Space Network receiving, VLF and ULF signals at 8.27 kHz and below, 10 THz long-wave infra-red comms and visible light comms including cloud-bounce and non-line-of-site comms via atmospheric dust and reflection from distant structures. Then into running a Rubidium clock reference and multiple frequency and time references and building up a collection of HP/Agilent/Racal/Fluke test equipment. I have a lot of fun using extremely low power signals, the best performance so far on 10 GHz is 21200 miles per watt and on 20 metres I've managed about 40000 miles per watt. I have equipment for frequency bands on 136/472 kHz, 1.3, 3.5, 7, 10, 14, 17, 21, 24, 28, 50, 70, 144 and 432 MHz and 1.3, 2.3, 3.4, 5.7, 10, 24, 47 and 122 GHz, and I have plans for 76, 135, 241, 248 and 288 GHz. Ham radio helped me get some of the jobs I've had over the years. Some folks think it's about talking to people on the radio, but I've always seen it as a licence to carry out radio transmission experiments that the general public cannot do legally. It's kept me fascinated for more than 50 years.

[@kiowablue2862]

@@MachiningandMicrowaves Experimentation is what hamming is all about. I've had a blast with Amateur Radio since getting my first ticket in '68. 73 OM. Bill - KR6K

[@jnelson4765]

Starting to learn more microwave stuff thanks to CuriousMarc's Apollo comms hackery, this is a brilliant demo of SDR for experimentation.

[@MachiningandMicrowaves]

I have some upcoming projects making antennas for satellite comms and moonbounce, so I'll try to include more of the radio tech elements once I clear my huge backlog of machining jobs. I need to get the vacuum chuck built, then I can start making Fresnel Sone Plate mmwave lenses and dielectric loading phase correction plates.

[@knickebien1966]

I though this fellow was an Aussie

[@MachiningandMicrowaves]

I've been listening to Kurtis from Cutting Edge Engineering too much I reckon. Picking up his intonation!

[@AdamSpurgin]

Hmmm....yes, I understand some of these words. In all seriousness though, this is crazy fun to peek in on. Radio is not my wheelhouse but I feel a few hobby coming on

[@LongnoseRob]

.. for those listening in monochrome.. great!

[@MachiningandMicrowaves]

Thanks for listenin'....

[@w8bya]

Hiya Neil......no stopping you is there !!!!! 73

[@MachiningandMicrowaves]

Hi Gedas, well, when it's 2am and I'm in the radio room, it's tempting to do silly experiments!

[@brenwyattm0rij909]

I stumbled across your channel for the machining side but...i have used WSJT-X and I'm intrigued by EME even though 70cm is as high as I go, at the moment. I do use Web SDR on HF just to check my signal and check propagation. Keep um coming, especially the machining!

[@MachiningandMicrowaves]

Hi Bren, 70cm is an excellent choice for EME, I made an elevation table and clamp setup for a 70cm EME system last year, and the op I made it for came first worldwide in the ARRL EME 70cm single band contest, so although I don't operate on 70cm, I feel a bit of an affinity for it. I've just been making the Tellurium Copper double-ended spigots to fit into the 2.1 mm bore of some 3.0 mm copper tube to extend the tube through a hole in the wall of a coaxial low pass filter cavity, maintaining 50 ohms, then transitioning to 2.37 mm to fit into the socket on the rear of a Radiall N four-hole panel plug, so there's no discontinuity, but running the Colchester at 1800 rpm after 10.30 pm is a bit antisocial. However, the neighbours had builders in today, laying some paving, and they spent the entire day either using a horrifically-loud Wacker plate or shouting at each other or into their phones. WHILE I WAS TRYING TO FILM SOME QUIET AND DELICATE MACHINING. So I don't feel at all bad about keeping them awake until 3am, but the other neighbours are innocent bystanders, so I'm giving up for the night. Plus I'm at the Day Job tomorrow, which is an awful prospect. I'm just buying a whole crapload of electroplating chemicals, anodes, cleaners, ultrasonic tanks, bubblers and the like to do a load of silver and gold plating and copper electroforming. This is for an upcoming series of videos which I'm naming Project Swordfish. If I can pull it off, this is going to be huge, and I mean properly huge. It's the sort of thing I could do a crowdfunder for, but I'm under a NDA so can't say what it's about at all. Secret Squirrel has nothing on me...

[@MachiningandMicrowaves]

There's also a project going on trying to get some equipment made that will allow EME on 47 GHz, as if 10 and 24 GHz aren't hard enough. A normal Kuhne 47 GHz G2 transverter has a noise figure around 6dB at best, but a team in Spain have developed a 2dB NF VLNA and are working on a multi-watt PA. I have a new 2 metre solid dish that should work well on 47 GHz, probably better thanthe 2.4 offset which has four petals or the 3.0 whcih has eight. When you need surface precision better than half a millimetre, petals just don't work. Trouble is, the 3dB beamwidth is only 0.2 degrees, which is smaller than the moon, so it's really too large, but the theoretical gain is 58 dBi, which is kind of radical. I'm having so much fun they' are bound to make it illegal. Neil

[@Freestylethemovie]

Enjoyable as all your vids! Can somehow relate to the joy of copying ones own weak signal, when no one will reply :) BTW have a look at the LMX2594 PLL chip. This one will give you a proper clean 10 GHz signal. This is my plan for Eshail-2, it can be both LO for RX and a CW transmitter on 10 GHz.

[@MachiningandMicrowaves]

I started laying out a board with a 2594 in 2019 but other projects pushed it to the back of the priority list. I'd already purchased the 0402 chokes and caps for the power supply pins. It is probably better than the ADF5356, but I got sidetracked by trying to build a twin-cavity VCO machined from solid copper with gold plating and using a reverse-DDS design with an AD9912, but the DDS board and other parts are still in boxes. I need to retire from the Day Job so I can spend more time in the lab and machine shop and less doing boring cybersecurity

[@Freestylethemovie]

@@MachiningandMicrowaves I always seems to be lacking time... so I'm about to take an early retirement in 3 weeks time. But then stupid enough to get another sheepdog to keep me even more busy :) Regarding 2594, they are pretty hard to find, extreme lead times, but if one to spend a few more quid, an eval-kit can be bought (have a few at the salt mine). But beware, earlier ones did supply a fairly good 100 MHz ref, but newer kits omits these, but still demads the same price. Shame! I have been trying to find any other PLL's that offers similar phase noise but noting is even close. My mecanical skills are pretty abysmal, but really enjoy watching you play with your toys! Keep up the good work and maybe, just maybe cu on the air :)

[@MachiningandMicrowaves]

have some excellent 100 MHz OCXOs, but I'm only using them as references on some ADF5355 boards. I'm watching the supply situation with the LMX2594 and looking for anything else that looks promising, but for now I'm using crystal LO chains with injection locking, and getting reasonable phase noise results.

[@BuckCalabro]

Now I need to put my ADF4351 on the spectrum analyser and see what the spurs look like. I'd love to get all of those pieces I've collected for 10G on the air. Very enjoyable video, thanks so much! 73 de KC2HIZ FN32

[@MachiningandMicrowaves]

Main issue with the 4351 is phase noise. The shoulders out to +- 100kHz with the standard PLL filter components ends up not much more than 40dB down. That makes you unpopular in a heavily-populated area of 10 GHz users. If you have a sharp dish then it's fine. Noise of a good crystal multiplier chain is hugely better, but then you need to consider how to stabilise the frequency enough. Weak signal work really needs better than 10Hz stability. It's worth looking at the bypassing caps around your 4351 and the quality of the grounding and separation of analog and digital paths, also using super low noise regulators. If you can elevate your dish and it's 3ft or more, you should be able to hear the DL0SHF moon beacon on 10368.025 +- doppler when you have common moon visibility.

[@soranuareane]

Hmm, yes, I understood some of those words.

[@bambukouk]

whoosh .... the sound of stuff going well over my head 🤣

[@johnydl]

As a stokey I can tell you that Mow Cop is pronounced like mouth rather than like a lawn mower ;) but I do agree with you about the hills XD

[@MachiningandMicrowaves]

I shall wash out my mouth with soap and amend my pronunciation forthwith!

[@RichardKCollins]

I understand almost everything he says. But me ever getting to do it myself? Not likely. He sounds an awful lot like my brother, who does things no one else can.

[@torinstorkey]

What do hams do with these high frequencies. Much activity?

[@MachiningandMicrowaves]

At the moment there is probably more interest and action and experimentation on 122 GHz than 10 GHz. Activity tends to follow interesting propagation enhancements like tropospheric ducting events, which happen more in the summer, but there are regular activity sessions across Europe and the UK. Summer thunderstorms raise a lot of activity as rainscatter and hail scatter allow communication over 600-800 km via Rayleigh and Mie scattering from hydrometeors. There are significant number of folks with 10 GHz stations capable of communicating via reflection from the Moon. That's pretty cool, right at the edge of what's possible from a communications system you can build at home. I'll do some vids on the build of my new 3 metre and 2.4 metre solid dish systems over the next few months. Lots of folks are using 10 GHz on receive from the QO-100 geostationary satellite launch by Qatar, which you transmit up to on 2.4 GHz. It can carry amateur digital TV modes as well at all the usual voice, Morse and FSK/MSK digital modes. In recent days I've been doing tests with a friend over in Wales on 10 GHz, using rainscatter, snowscatter, troposcatter and aircraft reflections. He is using just 140 milliwatts to a horn antenna INSIDE HIS HOUSE, 100 miles from here, and there are several 1500 ft hills in the path. I just made a short video about some experiments using digital modes using a remote unattended receive system in Staffordshire, over the hills from here in East Yorkshire kzbin.info/www/bejne/j2rUqp1rgNRppq8

[@joze8722]

Tone all this is gobbledygook!

[@MachiningandMicrowaves]

It's a highly-efficient closed-group idiolect. If an idiolect can be used by a group, that is. It would take three times as long , maybe five times, to explain it all in Human. I might re-make it one day and see if I can make it more accessible.

[@bobvines00]

Neil, since you fool around, urm..., work with SDR(s), have you ever used one as "test equipment" when troubleshooting a piece of (audio) equipment? :) A channel that I used to watch did this once, but, if I remember correctly, he didn't explain how he did it. I honestly can't even spell "SDR" or anything related to radio, but fortunately do understand many of the words/terms used in this and other videos where you get deep into your RF "black magic." ;) I _think_ that I understand enough about what you talk about here and other your "RF videos" and understand toolmaking & machining to definitely appreciate what you design & manufacture. I just finished watching all of your videos, so I eagerly look forward to your next ones!

[@MachiningandMicrowaves]

Hi Bob, I use a pair of ELAD FDM-DUO SDRs which have 16 bit ADCs and are very linear from -135 dBm to about -20 dBm, which is a good range, but then they are fairly expensive devices. They will only go down to about 8 kHz, and the performance is definitely not sparkling below 18 kHz or so. They use 122.88 MHz clocks so are good to 50 MHz, but above that, the sin-squared relationship starts to bite, so they lose sensitivity above 70 MHz and then are using aliasing above 140 MHz, so need a lot of filtering and a preamp. Below 50 MHz, they are excellent. I also have a lot of other SDRs, some of the simple dongles, a genuine Malahit (20 kHz to 2 GHz) and a Hermes-Lite. Below 15 kHz, the performance of many SDRs falls off, mainly because of coupling capacitors or transformers, but I ten to use a PC sound card as an instrument for audio troubleshooting. My spectrum analyser will go down to 1kHz as well, but not to DC. DC-coupled SDRs will work right down to DC so long as you adjust for the DC offset. I think the trick involves removing the DC bias choke and blocking capacitor and using the dongle in direct sampling mode of the Q-branch, but I've never tried. I think I'd go for an upconverter using a very good quality high-level mixer and feed that though filters to the dongle. Or just buy an old Wandel and Goltermann Selective Level Meter, like I ALMOST did in a recent auction

[@lawmand]

Hi Neil, will you be on the next ukac 3cm contest? It will be fun to try from home just south of the Wolds with a rather obstructed path, hopefully we can find a scatterpoint somewhere.

[@MachiningandMicrowaves]

Yep, just spoke to Steve. If I can work Paul on Wales over multiple 1500 ft hills from inside his bedroom with a 25dB horn, we shouldn't have any problems. I used to work Denis UVR every time without fail, and that path goes right over Holme Moss.

[@lawmand]

@@MachiningandMicrowaves looking forward to it Neil. The wolds are 3miles north forming the skyline from JO03BD 5m asl, the rest of the directions are good

[@MachiningandMicrowaves]

Belmont mast might be worth trying for a scatterpoint, but I bet we'll make it over the mud and limestone.

[@lawmand]

@@MachiningandMicrowaves Belmont isn't too far off the direct path so it shouldn't take too much jiggery pokery if it works.

[@daretodreamtofly3288]

Over the hills and far a way for 6 long years he'll stay away.

[@MachiningandMicrowaves]

That would be a very long-delayed echo! I thought it was "ten long years"....

[@2ftg]

Hmm. I might have to do this on QO-100. Or try to get some friends to host 10GHz websdr's in good locations.

[@MachiningandMicrowaves]

I think there is a lot of fun to be had using the satellite, with wideband data for instance, but doing a loopback test with WSJT-X is very simple if you have two radios or a full duplex system. I've never tried it. You need to do the trick to set up a shortcut to run the second WSJT-X instance with a different name. There is a planned project to create a simple kit for a receive package that will connect from a head unit consisting of an omnidirectional antenna and LNB feeding into an SDR dongle and a processor that can connect over wifi or ethernet to an internet connection. The idea is to make something that will allow anyone interested in radio to join in the fun by deploying a receive node in a similar way to the ADS-B receivers that feed into OpenADSB. The challenge is in how to collect the data streams. If we use the WebSDR approach, there is a tight limit on number of concurrent users. If we use local decoding and send only a decoded data stream to a central node, perhaps on an AWS instance, then we face the problem of what decode software to use and how to arrange GPS frequency lock at low enough cost, or how to decode over a wider range like the MAP65 approach. Then you have to ensure centralised software/firmware distribution and it starts to get complex. If we can get it right, there is an opportunity to do some interesting real-time mapping of propagation and if the timing can be set tightly enough, perhaps even time-of-flight analysis, but that needs timing precision to the microsecond level or decoding at an IF frequency of time pulses from a central reference source, perhaps by locking to the phase of the central reference. Lots of possibilities.

[@jakubniemczuk]

Mah god. Give us more details on those tools.

[@MachiningandMicrowaves]

I can feel a follow-up video is going to happen!

[@tacticalrabbit308]

Have you ever seen the secret of skinwalker ranch tv show they have a signal that they have found in 1.6 GHz and it affect people differently, just wondering if you have ever come across that band and found a signal ?

[@MachiningandMicrowaves]

Not heard of that, but it's in the cellphone band over here

[@Keithdib1]

Nice Neil, I understood some of that [I think]

[@MachiningandMicrowaves]

Glad about that, I didn't understand a word of it...

[@aleksandarvasilevski7410]

What is the backround around Emy? English is not my native language and it is difficult to understand. Is it artificial voice generator bot or some popular lady in UK?

[@MachiningandMicrowaves]

AIMEE is like the older sister I never had. She is my conscience. She has perfect hindsight. She can say things I don't feel able to say. Her face was generated by a Generative Adversarial Network. Once computer program creates what it thinks is a good human face. A second program looks at the image of the face and decides if it is human or a fake. If the second program is unable to decide if the face is real or fake, that image is published. The website is thispersondoesnotexist.com Aimee's voice is generated using Google Text-to-speech service directly from the script. I created 300 faces and picked the one that looked like the way the TTS voice sounds. AIMEE's name comes from Artificially Intelligent Machining + Engineering Expert.

[@PE3FS]

that was a rather smart thing to do.

[@MachiningandMicrowaves]

I think "silly" might be a better description, but it was a whole lot of fun

[@PE3FS]

@@MachiningandMicrowaves no no how is it called those days (even in Holland) out of the box thinking... I only used that program for FT4 and 8 and its a way of in "radioamateurism" that I like. I have started to getting interested in high frequencies and you help me with it. nice silly videos

[@Strothy2]

understood maybe 50%, enjoyed 100% :D

[@MachiningandMicrowaves]

About the same as me then, excellent!

[@chrisstephens6673]

You lost me after Staffordshire but as I'm sure its all very clever could we have subtitles in English so i could stand chance of following the narrative. 😕

[@MachiningandMicrowaves]

It is a bit of a breathless nerdstream of consciousness, isn't it! I might do a translation into Human at some point when next some exciting propagation happens, This vid was done to illustrate the method to readers of the Radio Society of Great Britain's "Radcom" GHz Bands column after a short item was published. All done in a bit of a hurry as I'd mixed up the publication date and thought I had another week.

[@northof-62]

chuckle

[@vidasvv]

WOW !!!

[@Dave.Wilson]

Obviously your not married from all that. Only joking. Thanks for sharing, I'me going back to my lathe now!!

[@MachiningandMicrowaves]

My wife of 34 years died recently after a long fight with cancer, so no, I'm not married any more. For the last few years of her life, I was her carer, as she didn't like medics, seeing as she used to be married to one. I found it hard to sleep at times, and messing about on the radio in the early hours was a great source solace and support. Machiing at 2am does tend to wake the entire village, but radio experimentation can be done in near-silence. Before that, Caroline was very accepting of my weird pastimes and often unsocial hours. I'm not sure why folks seem to make the unspoken assumption that their chosen life partners are going to prevent them doing interesting stuff. I certainly never felt constrained in that way