Exactly this. You've got about 20dB dynamic range to play with, 10dB if you want the best performance. So that'll easily take you from 100W to 1500W without changing anything. Below 100W you will want to use a smaller coupling factor.

Be extremely cautious. In the past I purchased a CleanRF variable sampler and, after testing it with a scope, I just put it aside and never touched it again. There was no way I was going to take a chance of damaging a $3000 piece of hardware with it. Consider building a fixed coupler or sampler and tweaking its performance with a NanoVNA so that you know exactly where you stand with it.

That's definitely the place to install the sampler/coupler so that it works with every antenna. However, as discussed above, a fixed 50dB sampler may not provide the necessary dynamic range for less than 100W.

A side note, if you have a Telepost station monitor, Larry will modify a coupler & add a sample port.

Buckhorn?

Final parts showed up last week.

Attached File


Everything laid out. My plan was to start assembly earlier than I did but did not work out. I got the parts in for the control board but I am going to use the board that I bought and the RF sense circuit for now. I am going to build my Display/R&D board first the second board will be for my amp. The nee board does CIV,BCD, Band voltage decoding already so its a better solution.

Lots of things to consider when squeezing this big old heatsink in a 6u case. So I had to make compromises. Also made some mistakes on the fan module but no one is going to see it. And since the 12x24 aluminum sheet was $33 I am not buying another one. I wanted the fans sucking air through the heatsink but do not want to mount the pallet upside down. So I compromise and air will be drawn in from the bottom and sucked out the rear. Placement is tricky due to RF power this thing is going to do. So here is some of the pictures.

Attached File


Fan module and plate the heat sink will sit on. Bottom slide in panel is bolted to the fan. I may add some screen to the bottom plate but for now rolling with it.

Attached File


Bottom plate. Pardon the mess in the background but it my bench next to the electronics desk is holding my big 3d printer and a ton of parts for all my projects.  

Attached File


Initial mock up to adjust and figure out how I am going to mount the heat sink to the side rails. Side rails are also heat sinks so some of the voltage regulators and small boards will be mounted on them.


Attached File


Here is phase 1 complete. Now I will need to find where I am laying everthing out and the wire length and molex connector locations. Mark everything pull back apart and drill and tap. Then assemble and make sure everything is good. I will then have to pull the pallet out and begin the wire process and basic testing to make sure all the boards are working then reassemble.  I will also need to mill the spot for the touch screen  and power switch. Going to be nothing on the front but the power switch and 7” screen.

Attached File


Attached File


Phase 2 will be milling the panel and making brackets and drilling and taping for the stand offs and getting it mounted up.

Phase 3 will be the full wiring.

Note to self, never show photos of toroids I wind on here……. Lol. Beautiful work there, looks awesome.

You and me both. I looked at that picture and my fingers started to ache and all I could think of is how long it'd
take me nudging individual windings to get them even like those.

Got some work on the Amp done but not as much as I would have liked.   Attached File


Waiting on some 8 pin molex connectors for the LOF and have some more machine work to do. Attached File


Still have some more layout and mounting to do. But its getting there. Lost 1 day this weekend as it was my Anniversary so had to spend some time with the wife.

Did start on some SMD stuff for the control boards.  I did get one more amp pallet done for the LDMOS QRP amp.

Attached File


2”x2” and uses a sigle MRF101AN TO-220 package device for around 130w output. Still need to get the TX-RX directional coupler board done and low pass filter. Going to use an Arduino Mega control board and 3.5” Touch screen.

I built this using my usual crude construction techniques to play more with my gutter-tenna and other prototype projects.  I may have gotten carried away with diameter and turns.  I suspect I could tune AM broadcast bands with this coil if I wanted to, 160m only used half the coil.   I was able to tune my rain gutter to 20m, 30m, 80m, and 160m.  I just could not get a good SWR from 40m, maybe some harmonic demons in the counterpoise.   I made contacts on all but 160m.  There was just noise and static there, not sure if that was all the lightning last night of just too compromised of an antenna.  The 80m was not great either, I tried to join a local state net but was only recognized for my momentary contribution to static.  I did get 3 80m FT8 US contacts.  20m and 30m were making Europe and South America no problem.

Attached File

I got a out 10 states on 160 ssb using a 40m coil, 20m ham stick, random bit of wire and a 9:1 unun last year,

I have some ideas for more 160 shenanigans this winter..

Make yourself a 1:1 Balun if you haven't already and put it at the antenna feed point. It will keep the CM off the coax and maybe solve some of your wonky-ness.

Yes, I have one there.  It helps with 20M which gets a bit flakey without it.  I had 40m working without a counterpoise (just a ground), so I should swap that with a different length and try again.  Also the feed line is short 8', I have read that can create more issues than a longer line.

Shelby Hamfest score

Attached File


1296 circular feed horn for my 2.4m dish. Got the hybrid coupler and feelines. Also picked up some linear actuators for Az-El control

This is my Shelby Hamfest score.  So far I am in the hole for $20.  The clean audio and controls work, but as soon as the DSP is mixed in I get pure static.  The PCM54hp DSP chip seems to be a common failed part for a lot of DSP and synth products, so I am rolling the dice on another $5 to replace that.  In a few weeks we will see if I am 25% more foolish than I am now already.

Attached File

Nice - is the hybrid coupler used to generate circular polarization?

I'm building an OK1DFC-style septum feed for 1296 MHz for my 2.4 meter dish - following the build instructions from KL6M, who used a 2-foot section of 6" x 6" x 1/4" square aluminum extrusion instead of bending sheet metal like the OK1DFC plans. My dish has an offset feed, and (I believe) an f/d ratio of 0.7 - and so I'm going to add a pyramidal flared horn to it a la OK2AQ, in an attempt to better illuminate the flatter-than-the-usual-prime-focus dish.

I'm trying to decide whether to make the horn out of sheet aluminum or whether to use copper flashing. There are some details that I'm not sure of, since I don't have much prior microwave experience, but I think that I can get away with screwing on the feedhorn to the open edge of the aluminum extrusion with multiple screws, putting the screw heads inside the feed - thinking that the size of those screwheads would be pretty small compared with the 23cm wavelength (maybe I wouldn't try that at a higher frequency, though). I'm also thinking that the tiny slot that would be formed by the overlap of the horn inside each inner wall of the extrusion would not cause any problems, because it would work as a 'waveguide below cutoff' - but maybe I should cover the joints with copper tape or something, just to be sure. I'm also wondering about conductivity between the extrusion walls and the walls of the horn over time, as it would be exposed to the weather. Again, not a whole lot of prior microwave experience, and my understanding of the waveguide propagation mode(s) for the E-field is rudimentary at best - and how that may be affected by the circular polarization is also a bit of a mystery to me...

The OK1DFC sheet metal feed plans connect both halves of the rectangular feed outside of the feed, by screwing together flanges that are outside of the propagation path. I don't think that I can do something like that to attach the feed horn to the aluminum extrusion, however.  

https://kl6m.com/23cm/23CM-SEPTUM-FEED.pdf

https://www.ok1dfc.com/eme/Technic/septum/23-13draw.pdf

https://www.radio.feec.vutbr.cz/esl/files/EME/Doc/OK2AQ_23cm.pdf

That's a lot of socketed ICs! I wonder if reseating everything (which you've probably already done) would help?

Yes, I re-seated.  Then I removed each one to see what functionality/noise disappeared.  The issue is in the DSP processing.  When the processed audio mix knob is turned up, pure static is the output.  The clean audio is fine and the controls and display is working.  I suspect the DAC chip, and found a replacement.  Others have replaced this chip in similar devices to resolve static/noise issues.  Its a $5 gamble.

Correct it is, there are some newer feeds that have a separate RX and TX port. This one depending how it is fed will change the polarity. So feeding the 2 ports from port 1 and 2, and transmitting into port 3, and receiving with the LNA on port 3 you get RHCP tx and LHCP rx. Basically you use coaxial realys to switch port 4 to a 50ohm load when Txing and switch the LNA into the circuit when RXing. It takes some more equipment to make this work and there is a couple tenths of db more losses then a separate RX/TX septum feed, but it has been done this way for a very long time. And the price was too good to pass up. Paid $90 for the feed, the Hybrid Coupler, and the Hardline jumpers.

ETA: This is a VE4MA Kumar feed that is made from Aluminum and welded together. It's a solid piece.  I am going to machine some Billets that clamp it in that way I can make some adjustments to make sure the dish is illuminated correctly and be able to adjust the SWR. I think I may ultimately use a Transverter and PA mounted at the base of the Dish and pipe the 144mhz IF back to the shack with 1-5/8 hardline. So that will save me from having to use too many relays. This will get me on the air and then I can build a Septum feed and eliminate most of the relays needed. Just 1 relay to switch the LNA to ground and the AMP powering directly to the feed path with 1/2 inch hard line. Minimal losses that way and simple to configure. But like I said I paid effectively 60 bucks for the feed, he paid more than that to have it welded.

https://www.k4qcd.com/

Interesting portable unun kit, housing a 49:1 and 9:1 in a compact package, claiming to be good for 100W ssb.

Might snag one for some testing.

Still experimenting with mast support systems that work backpack portable.

I am getting close to liking something.

Attached File

Attached File


Doing test fits and basic wiring. Going to replace these stick on board mounts with standoffs eventually but since I will be replacing the main board I will rock these for now. Also ended up changing the coax position to keep the input and output coax away from one another for some isolation to keep it from going into self oscillation.

Attached File


Used Molex connectors to keep everything modular. The only thing that will not be is the Coax. Since SMA connectors will not be able to handle the power when the second pallet goes in.

Attached File


Once I do the power on test and function test everything I will clean the wiring up and cover it all with the abrasion resistant split sheath. Ran out of RG-400 to wire in the LPF to the 2 dual directional couplers. I will also have to tune the length to supress harmonic reflections, seems the board builder suggests 25 or 85cm for that. Have some basic wiring left, and my thermocouples need to be mounted along with my fan relay to bypass one of the 2 50 ohm resistors for kicking the fans on high. Need to cur the back plate and mount all the I/O parts and of course cut the front plate for the screen then debug it all.

She is heavy though. Bet its got a couple of pounds on my SB-200 but even with this config it should do legal limit without being in compression even at 50V. But I will probably run it around a KW max with the single pallet. Its a tight fit with the massive heatsink but should stay cool.

I am using a W6PQL mosfet switch to switch the 50V supply and a noise free converter for the 12v control and bias voltage. Protection board will kick the bias off if any issue arise. I will have to go to West Marine tomorrow to get a 50amp breaker for the back panel. Eventually each pallet will have its own supply with breaker and switch. Lots of ferrites everywhere to keep RF off the circuits and on the Coax. Eventually I want to switch the 13db worth if attenuators with a relay system for the low drive radios but for now using the W6PQL method of bringing them out to the back panel and using a jumper.

This site needs a like button!  That is an impressive project!

It really is. Stanprophet09's project makes me feel like this guy:

And on the other end of "same effort as building the pyramids" scale from Stanprophet...  

Browsing through an antique shop in Toledo a month or so ago, came across this beauty.

Attached File
Attached File


The only key I had at the time was something I cobbled together out wood scraps, spare hardware, and a feeler gauge that I needed for some testing.  Saw this and thought, well, for $24, why not?

Six weeks later, here's what it  looks like:

Attached File


Base is cast iron.  Dremeled all the rust off, repainted with some gloss black enamel model paint.  Wire brushed all the metal parts to get the corrosion off, followed by some Flitz polish.  Wood base got a good scrubbing with some Dawn and a toothbrush, then a couple of coats of tung oil.  Knobs are Bakelite, they got a couple of sessions with some Novus scratch remover.  

Two real challenges:  the key itself, and the contact point.  Started wire brushing the key, realized the nickel plating had all flaked off but there's some good steel underneath, sooo, off to learn how to DIY nickel plate.  Wound up replating it entirely, as well as the wire terminals, which had some of the brass showing from wear.

The contact point had some kind of putty, and string wrapped around it to electrically isolate it from the base.

Attached File


I decided that wasn't permanent enough, and besides, it looked kind of janky.  I basically needed a two-step washer to electrically isolate it, raise it high enough for the circuit closer to fit underneath, and center the contact point in the hole in the base.  Enter the Prusa 3D printer:

Attached File


Only thing left to do is replace the leaf spring with a conical spring, that's on the way from Amazon, should get here next week.  

Been a fun little project, kept me out of the wife's hair for the last month-ish, and now I have a nice-looking key to start learning CW on.

I have some experience with waveguide.

Advice: start with a simple horn.

BUILD A BACKYARD ANTENNA RANGE AND TEST EVERY SINGLE PIECE OF MICROWAVE GEAR.

Every piece.

Do simple a/b comparisons, is this "10 db" thing really 10 dB better?

You will find waveguide to be extremely tolerant of "sloppy " construction.

Prove it to yourself by building several horns from different materials,  even one made out of aluminum foil covered cardboard.

You will see just how non critical waveguide is.

You will like this:

W1GHZ microwave book

https://www.w1ghz.org/antbook/contents.htm


Rege

P.S. at the Intelsat earth station I worked at, we used a frequency reuse scheme where we sent right hand circular and left hand circular polarization on the same frequency to each satellite.

I HIGHLY  recommend that you start your backyard antenna range with 2 commercial coax to waveguide transitions and add complexity one bit at a time.

A dual polarity,circular septum offset feed is a   pretty ambitious first attempt

Yep I have been reading all of the W1GHZ materials I can get my hands on. I will know pretty quick if its going to work based on RX signals off the moon. There is a massive beacon with a 500kw ERP that is in Europe. Plans are to tune feed point on the RX signals then move to low power testing. I will be using Waveguides on my 10ghz setup. I may make a dual band feedhorn for this dish but at 1.2ghz I will stick with Coax. If I build the Transverter and LDMOS amp for the back of the dish and run 1-5/8 coax back to the shack it should minimize the losses. I have a 55 element looper for Tropo on 1.2ghz and learned pretty quick that microwave can be a picky to coax terminations and quality.

This project is in the planning stages of course. Once I get through a couple of the Amplifier projects.

I started with a single band, linear polarization horn....

I calculated dish f/d

That told me the angle the dish subtended from the horn.

I built and tested in my backyard range a horn with a 10db Beamwidth corresponding to the above angle.

Then I placed the horn at the feedpoint of the dish and adjusted for prime focus.

In my opinion, trying to build a dual polarity and or dual band feed, without independent testing, and then only optimize feed focus is unlikely to work.

And the whole idea of mounting the electronics package at the antenna is so you don't need "1 5/8"" "hardline" back to the baseband gear. We mostly used rg59 for that sort of stuff in the Intelsat System.

Rege

Have a buddy that is a Microwave Guru with about every bit of test gear known to man. So I will have help on this project for sure. That is the biggest issue when it comes to Microwave is the equipment and people that want top dollar for old equipment. But slowly gathering what I need to build what I need. Always appreciate advice.

Attached File


Attached File


Revised how the LPF was mounted to make clearance for the breaker. Also moved some of the wires on the bottom by the rear panel. So it cleaned it up quite a bit. I have some wiring to do tonight for the power and the rear panel pieces.

Back panel is cut and mounted. Turned out pretty good. Need to find a better way to cut the large diameter circles in the aluminum for the fans. The dremel work but its not the cleanest method. Good thing on the back panels the fans hide it a bit. Tool my time and tried to make it as clean as possible.

Attached File


Ran out if black labels on the last one! But found a cassette just a min ago. So close right now, cannot wait to see if I can melt down my dummy load. When the second pallet gets done it may be able to that. I did buy a Vectronics DL-2500 dummy load just for this project.

On a side note, I seen I was getting a UPS package yesterday in the Wife’s name. Looked and said it Shipped from Shelby NC. Hmm maybe she won something and we did not know. She sure did and an Alinco DJ-MD5 DMR radio showed up yesterday. Nice thing the CPS is almost identical to my Anytone 878. I was able to open my 878 code plug and program it to the Alinco. That was pretty awesome.

If you have a router you could make a circle jig from wood for cutting large openings.

Attached File


Realized I missed the 12v wire to the RF delay relay. So after that I can power it on and smoke test it. Keeping the Bias line and 50v line off the pallet so I can check all the boards and keying circuits. If that passes I will begin the key down test and the calibrations for the power meter and make sure everything is working.

Ended up using RG-316 from the pallet to directional couple to the LPF and from the LPF back to the coupler. Since it may need to be tuned I figure its best to find the length with the RG-316. I will then change that out to RG-400.

Tests will be done with the front panel off they once I a confident its working well and then cut the front panel. Then it is time to get another pallet going and the splitter and combiner circuit.

The HV protection board, Directional Coupler, Low pass filter, and relays are all rated to 2500w. So everything has been built to take the power if 2 pallets. Oh and I have to hook up the second antenna port. Going to redo the antenna connections a bit. Get the load and VNA on the LPF and the 13db worth of attenuation on the amp. Very very close now.


Attached File


Back wiring cleaned up well.

You can do this!

The secret is to test every single piece as you build your station.

My first 5760Mhz rig was a G3JVL waveguide  converter, I built it on my kitchen table.

http://microwave-museum.org/exhibits/mwm-027.htm

RF test equipment was limited to a Diode detector.

And I had nothing other than a ruler, hacksaw,file, and a propane torch for tools.

It, the converter, worked fine.

Don't listen to jagoffs that tell you microwaves are some special place regular people don't go.

P.S. For reasons, please make your first EME contact "random", that is just answer the other stations cq without any pre--arrangement.

It better that way!

Rege

Indeed randoms are the best. I routinely do randoms on Meteor scatter. And most of my Eme contacts on 432 were randoms.

Have some bugs to work out still. PTT line is not working like it should. Some other stuff needs to be dialed in. Not a huge fan of the controller interface. I need to move the power switch by extending the wire. Made a huge mistake when I drilled the holes for the power switch. I was not happy how the screen cutout worked out. Its not bad but if I am forced to look at it every day it will bug me. So one all the bugs are worked out and the new control board it done think I will have a custom panel made and rack mount it with a remote screen added.

But it does work. Once I get the PTT line done I can dial in the directional coupler lcd reading.

2w of drive on 80m equates to 1143 on the LCD and the 1KW bird slug the needle is on the stop peg.

Same drive on 40m shows 1004 on the LCD and about 950 on the bird.


Takes about 8w of drive to make 1KW on 6m. My10db and 3db pads are not working as designed and I damaged one trying to test it. So going to have to replace them. Also my RF delay circuit was not working, and resulted in no power out. Used the Oscilloscope to find that and just bypassed it for the testing. But hey it does work. So about 90% of it worked on the first go as wired.


All readings were through the low pass filter. I do need to check the Harmonics just to be certain, and check the IMD levels. 2w of drive going through the 4db pad circuit on the board is .8w total hitting the device.

In a bit of a side project I'm working towards getting set up to operate terrestrial microwave on the 10 GHz and 24 GHz bands.

As for 10 GHz, last weekend I picked up a used 2-foot dish antenna with a built-in waveguide feed and with a radome cover. I ordered a used WR-90 waveguide to type N transition adapter from eBay, which I received on Friday. Today I temporarily removed the radome in order to visually check the feed, and then rotated the feed by 90 degrees to set its orientation for horizontal polarization. I didn't really have to do that, as I'm in the middle of fabricating a different method for mounting the dish than what it previously used, as the old mounting method involved a very heavy steel bracket, which I'm going to replace with lighter-weight aluminum - I'm not going to be tower mounting this dish, rather I will be mounting it on a surveyor's tripod for portable use, and I'm also planning to affix the DEMI 144 MHz to 10 GHz transverter to the back of the dish.

A photo of the feed and another one showing the back of the dish after removing the old bracket (blue painter's tape over the waveguide - I have not yet attached the type N transition):

Attached File


Attached File


As for 24 GHz, I just bought a 144 MHz to 24 GHz transverter complete with 2-foot dish that a local ham had assembled from surplus as well as new parts. I have not yet fired it up - but next weekend is the 2nd weekend of the ARRL 10 GHz and Up contest, and I'm hoping that I'll be ready with both the 10 GHz and 24 GHz setups to take to a hilltop and try and make some QSOs.

I will either be using an Icom IC-705 to drive both of them, or I may use the Icom on one band and a Yaesu FT-818 on the other one. I need about 1 watt of drive on 144 MHz for each transverter, and will plan to try both CW and SSB on both bands. I'm about a dozen miles from another local who is set up with similar rigs on both bands, and hope to be able to try things out with him ahead of the contest. This weekend I attached a Manfrotto camera quick-release base plate to a reasonably heavy Slik photographic tripod, and attached a matching quick-release fitting to the threaded tripod screw hole on the bottom of the transverter.

The 24 GHz transverter (attached to the back of the 2-foot dish, mounted on the tripod):

Attached File

okay, it's commercial...
old gear out


new gear in at new site.  Added digital capability, battery backup, and a lot more monitoring/remote control.


On the leftmost leg in this photo... VHF dipole on the top, gold 900MHz grid antenna, and the small gray dish+odu all came off, along with feedlines for all 3.
Tower in the background, on the top of the mountain, is where the VHF antenna is moving to.


Old shit landed.


VHF antenna coming down.  The new kids (this was their first LMR and tower job, ever) were rockstars on the ground.


"Spider-man, spider-man..."


Aspens just starting to turn.


Over to the other site, and all the way to the top I go.  That's the lightning rod i'm looking down at


mounted and leveled


Working, but needs some work.  I'll be back on this site to move some other equipment and antenna over.

Keep the posts coming on this stuff. This is relevant to me since I have the Dish for 10ghz just need to get the other parts.

Took me a while to dial this in but man am I super happy and pumped with this!   Attached File


So here is the input power.

Attached File


And the output

Attached File


Bit scared to take it higher but this pallet is good for about 8 watts of drive. That will put me near compression though. I did a brief key down at 1645 into the load. Need to get the new attenuator in and that will make it easier to dial the power in.

Had a long conversation at a KW earlier and the 22 ambient temp only went to 25c. So cooling even with the fans on low is not going to be an issue.

On 6m its performs ok. Takes about 8w of drive to get 775w. Betting if I play with the coax length between the pallet and LPF I could dial it in. But not a real concern with the KW amp under my desk.

Did have to move the switches to get the LCD mated correctly after my oppsie. But I kind of like it. I may not rack mount this, so I may have a custom panel made that will make the width the same as my JW miller auto tuner. Anyways its working well so far. So much so with the ability to hit Legal limit without going into compression I am questioning the thought of doing the second pallet. Would make it much harder to run lower power. Time will tell.

ETA Just had a long QSO at 1200w with my Hermes driving the amp, no pure signal. Other station on an Anan, great audio reports and they said even with that power level there was no splatter signal looked like it was on rails. Temp got up to 33c and stayed there for the duration. Super happy with that.