No one would put a station here!

A desert arroyo with regular 60+ MPH wind gusts. Tall ridges rising on both sides. HOA to boot. It's a crazy notion.

Inside, a shack on a typewriter table. IBM Selectric gone decades ago. Must keep it lean. The hobby here must not overtake the hobbyist. My wife said it first, but I'm subscribing to that. Many hams don't. More power to 'em!

First thing you do is put up an antenna. Yeah, that's a fern hook at the apex. Very HOA. More at the Outdoors tab.

The pic above is from my balcony looking NNE. Shack is in my den through a sliding glass door about 20 feet away. The antenna is that shot was the old 40/20/12 coupled resonantor W1ZR special. Not extremely stealth.

Those are high mountains. Mt. Lemmon is at 9,600' to the left off the pic (far behind the hills you can see).

I did the May, 2021 FCC-mandated radiation exposure calculations, and in the 10m band, I just pass worst worst case! I have 100 W out of the rig's SO-239. Can't be more at the antenna! 10m checked as OK at 8' standoff for long bursts at 40% DF CW. All lower bands are lower exposure. I urge all ops to do this easy thing. Just go to ARRL's Calculator, fill in the numbers and hit the Calculate button. Here is my completed calculation.

 

 

Electronics

Yaesu FTDX10 with a couple haptic sensors. The Yaesu MD-100 dynamic mic was once their flagship, now only seen on eBay. It is, however, very articulate might I put it. The N3ZN iambic double-lever paddle weighs 3 1/2 lb and doesn't go anywhere courtesy of that sticky neoprene pad I made for it. The Mac, out of sight, connects to the DX10 by USB for digital modes, hardly used. The Astron 35A linear power supply sits on the floor under the desk. And recently added, a manual tuner sitting atop the Astron, which barely gets warm.

Yaesu's design team thought of just about everything you might do as a run-of-the-mill HF radio amateur. It is but a single receiver rig, but the noise level here is typically so low that a receive antenna applique (now, there's a word) is unnecessary.

That's about all indoors.

 

 

Antennae

Electrons dance. Then nothing all the way out. Except fields, whatever they are.

Physicists say fields or photons. Engineers say far (1/r^2) field. Even as an EE steeped in EM, I say magic.

Yes, an infinite series of r^-n parts of something, i.e. all the field components, n > 2, going down so fast that n > 2 becomes simply negligible. The near field is just the sum of everything in that becoming-negligible range of n > 3 parts. We don't often bother to think of it, it having almost nothing to do with communication*, unless we're really close in, like that near field communication (NFC) thing at 13 MHz used on your credit card supermarket checkout.

 

Current HF Antenna, as of 26 May 2024

The Zepp got retired, just to try something different from the very same apex of the painter's pole rigging.

This new thing is a 40m "fundamental" off center fed, or OCF. (Notice I don't call it an OCF dipole. OCF's are not resonant, at least in most bands.) And what an experiment it turned out to be! Receiver noise floor came up more than an S unit (averaged over weeks, from near S0 before to virtually always above S1 after on 40m mornings and evenings). Those consistently noisier band conditions were good, proving one thing: the RF gain of the antenna itself is now an S unit-plus better than before. And reciprocity strikes! Receive reports on my signal have also come up an S unit-plus. Hooray. This must be due to current at elevation, but details are not pinned down.

My OCF is a bent (slightly dog-legged, seen from above) inverted V of two dissimilar wire lengths: 57' to the N, and 11' to the SSW. It tunes 40, 30, 20 and 12m bands. It is almost resonant on 30m. It is still at 21' apex, where the transmission line runs through a 6:1 balun and is then split immediately into the two radiating wires. Dimensions and explanations are due to L.B. Cibek (SK, once W4RNL, which call sign is now recycled but should have been retired!).

The wire is still 26 AWG insulated, silver-plated strands. Many strands. MIL surplus! Whether this is too thin for good radiation efficiency is still unclear. The 57' leg measures less than 0.5 ohm DC resistance. Because it's stranded, the effective surface area is larger than solid. At 100W fed into the nominal 300 ohm real part of its impedance at 7155 MHz, this is less than 1.5W lost to DC resistive dissipation. But I readily admit that this could be multiplied due to skin effect at RF. Still, I find it unlikely that it would, say, double the loss at 7 MHz. It would be worse than that at 30 MHz, of course. More research needed. Longterm plan is to go to 20-22 AWG, but not lower gauge due to weight.

There might be a slight improvement in bandwidth roll-off at any tuning point for lower gauge, thicker wires . But I haven't seen much VSWR effect (maybe 1:1 -> 1.1:1) over even 200 kHz at 40m. So why worry!

 

Previous HF Antenna, as of around about Feb 2024

20m Extended Double Zepp, hung from an eyebolt at the end of a painter's pole extended from balcony

Now replaced, this was the same 20m extended double Zepp as before, but with its apex removed from the fern hook under the balcony ceiling to an airborne location above the back yard at the end of a 31' telescoping painter's pole at about a 45 deg take-off angle from the balcony floor. The pole was extended only to about 27' for strength. The result was an apex about 7' out and 21' off the ground, sloping on either side down to about 7' above ground.

Improved performance. Hard to judge from signal reports, but generally better, at least on 40m.

 

Previous Previous HF Antenna, as of 22 Dec 2023

20m Extended Double Zepp, hung from a fern hook on the balcony ceiling ... everything below is obsolete

Now long replaced, this was a 20m Extended Double Zepp in an inverted V geometry. Apex at a big 18 feet AGL, under a corner of my roof. Designed for 20m, where it worked great here right off. This design type often uses 5/8ths wavelength on each of the two legs, which is what mine was. It is thus quite long compared with a dipole (typ. 1/2 wavelength all across). Mine is cut to 43' per leg, 86' overall. It did tune up to 40m, but not so well at 80m.

I am using 26 AWG (correction from original article here) insulated stranded wire for the legs. It is moderately stealthy, more or less "splitting the sky" a la a Star Trek Cloaking Device to a viewer a hundred feet or so away. Birds had no problem with it, even as hawks dive bomb our backyard every now and them - quite exciting. 30 AWG would have be really stealthy, but shrinking bandwidth and strength at tension would have both gotten really significant. Questions about wire resistance actually make it into the FCC Tech or General licensing exams. But using ordinary AWG tables on the web can be misleading, as RF travels mostly right near the surface or skin of any wire, thereby boosting the loss per unit length into the radio realm.

Putting up antennas involves simple mechanics. Here are a few pics of this installation.

 

 

Here are the two boxes visible in the above feed point installation.

 

Previous^3 Antenna, replaced as of 22 Dec 2023

I used it for so long (two years) and had so much fun with it that it deserves an archival presence here!

It was a W1ZR oddly-named "skeleton sleeve." It served well on 40, 20 and 12m with only the rig's internal tuner for matching (Up to about 3.2:1 SWR.) ("Skeleton" must refer to atrophying bones hanging inside an old Hawaiian shirt on a very OM, no?) A better description is a driven dipole, with a second, closely-spaced parasitic dipole. The easy version is a two-bander. And the occasional, lucky third band for free!

My earliest goal, really my only goal at that point, was an antenna that my rig's internal "fractional" automatic tuner could handle. No tuning success, no transmit. That simple. It has been a huge compromise. Mostly because of the "folded" architecture I chose for the basic design. More on that below.

Here was the old installation.

 

 

An aside: Joel Hallas, W1ZR was a well-known tech-savvy guy at ARRL's service among others, who left us (SK) way too young. He gamely answered my email on his antenna with a simple challenge: it's wire - just build it. I think a very likeable guy. And a clever call sign. That ZR: Z the impedance, R a resistance (or real part of impedance). So, ZR => the impedance is resistive, i.e. the reactive part of the antenna impedance (at a resonant frequency) is zero, and so the antenna impedance is (100%) real, and the antenna is then all radiation resistance plus any (small) dissipative resistance. Just perfect! (Well, except for that fold, more below.)

Any comparison with a true sleeve dipole is a glut of words, and (a la Supertramp) "not quite right!" If you're interested in actual sleeve dipoles, go to a textbook like Stutzman & Thiele. I took a course from them in the late '90s - great guys. Note that neither the textbook sleeve dipole nor the "skeleton sleeve" implements a sleeve balun - that's another creature altogether! The balun in this design is just a brute force RF choke (the ferrite ferrules below) around the RG-58X coax (check out Palomar EngineersTM) to esure that the antenna is the dipole thing exclusive of the feed line. Impedance matching is a cut-and-try affair, but one can get well under 1:2 VSWR in both bands under ideal circumstances (mine far less than ideal, so a bit above 1:2). The FTDX10 internally tunes up to 1:3 across almost all of both bands. The shorter, parasitic-current, coupled "sleeve", resonates by itself in the higher band. The longer, direct driven current resonates across its length that includes a folded-back end length to physically shorten the overall length. See below. Pretty darn clever.

How that "fold-back" at the ends of the driven dipole affects things is much less clear. It - and the gap to the ends of the parasite "sleeve" - affects impedance and is thus VSWR important, but has little to no effect on the lower band resonant frequency. I keep thinking, there must be a trap hiding somewhere in this design! That fold, obvious in the pic below, does physically shorten the whole assembly, a plus if you're short on space and a main feature in Joel's 2011 QST article (join ARRL for free access to just about every QST ever published). And, perhaps, the fold is similar, here a 2D analogy, to the "capacitive hat" used on "short verticals" to both increase bandwidth (decrease Q) and offset inductive reactance. (But after thinking in terms of Poynting's vector, that analogy falls by the wayside.)

Well, here's a major, perhaps the major compromise, that of the folded length of the main, driven (40m) element. You will recall that the current in a dipole keeps going out (in either direction from the central feed) until it hits a dead end. If this dead end is at a current null, then we have resonance. And so it is for the "full length" of the folded element(s). But the folded backward part of the 40m element is carrying current in the opposite direction from that of the other part. And this backward current is traversing almost half the total length of each leg of the antenna! From the Ampere-Maxwell Law (of Maxwell's equations), the curl of the B field vector follows the current density vector in the conductor giving rise to it. When the current "turns around" at the end of the W1ZR folded geometry, the B field turns around as well and destructively interferes with the B field from the driven element. In other words the B x H power vector is substantially reduced by the mere fact of folding the wire 180 degrees back on itself. It is a major loss. I had always wondered about the foreshorted, folded character of Joel's "space restricted" antenna, and rightfully so. Joel did provide an unfolded option for the same two-band design, much longer of course. Size matters! And so we give up something for any compact antenna. The next antenna element I build will not have any 180 degree folds!

I built my antenna from ARRL's great little 2019 pub, HF Dipole Antennas for Amateur Radio (go to the ARRL store, you don't have to be a member). Many interesting designs there.

There are "skeleton sleeve dipole" EZNEC calculations for many pairs of bands down to 10/6m, and probably further.

 

 

Below are my (new) 40/20m dipole VSWRs at the rig. The yellow trace is VSWR. The rest is a Smith Chart of Z (green) and phase (violet). The two minimums, one (marker "1") near 7.200 MHz and the other near 14.300 MHz. The 40m is just great. The 20m not so great, but useable. That 40m dip minimum is 1:1.3 native to rig's UHF connector, 39 ohms real with 40 nH to be tuned out with a rig tuner capacitance. Full band coverage (CW+SSB) is afforded on both bands, to which I must attribute the very capable built-in ATU of the FTDX10!

 

 

Now, about the main modification and the earlier very bad -18 dB relative performance (to a decent dipole) inferred from "dueling" WSPR tests against a reference dipole (up about 1/2 lambda) a dozen miles away at another local ham's QTH.

The mod: what I did was simple and cheap. Instead of literally wrapping it around the balcony, I pulled the two legs of the antenna away from the balcony and downwards toward my backyard wall. It becomes, in effect, a low-vertex inverted V. And it works much better. But I'll have to redo that WSPR comparison again to know just how much better!

I will add this. The "dry" VSWRs of this antenna are a bit higher than shown above. Instead of 1:1.3 with "wet" window line (after an early morning of rain), it's more like 1:2.3, but still easily tuneable. So weather plays a significant role in this nominal 450 ohm window line antenna. Probably not nearly as much in a conventional dipole, but be on the look out. Do not assume if you've tuned it once on a hot summer day it will be the same on a cold, wet winter day.

Antennas can be complex in response and provide (even pleasing) surprises. When I looked broader than just up to 15m (already knowing that 10m would not be possible), I found I could tune the WARC 12m band, all of it, with no change at all to the 40/20m dual band system. And so it was, revealed in a wider scan on the nanoVNA.

 

 

Suddenly, "something for nothing." The Tri-Band (40/20/12m) Coupled Resonators dipole system. Three bands in one string, no traps! Too good to be true, but true enough. The "extra" (third in this sweep) resonance around 18.5 MHz is, alas, not a ham band, so no bonus there. But look at that (yellow) marker at 24.960 MHz!

And how did it (that last 12m resonance) perform? Well, it secured my first JA QSO (into Hokkaido, Japan) one afternoon in SSB, no less. Nearly 10,000 km. So the signal obviously radiates pretty well on this "freebee" band.

 

* Near-field communications (NFC) in your smartphone is exactly that! We've evolved to the point where we don't even call the most common EM exchange in First World consumer experience "radio."

 

 

Second thing you do is buy a receiver. The FTDX10 is a superlative receiver. Many experienced hams have called it the best ever made. Well, that's bound for debate, but it's indicative of just how well-executed its receive design is.

I say buy the best receiver you can afford. Transmit is a simpler thing that most all rigs, except really cheapo designs that fall way short of linearity, do adequately.

I will say one more thing. If you work SSB, cheap hand mics (that may come packaged with a new rig) are just that. Buy a good mic, and people will better understand what you say into it.

If you think you can stand it, additional rig blather awaits you here!

 

 

This is a tiny blog on the FTDX10. There is a very big blog space on this rig here.

1) Yaesu receivers have long had a design feature for CW called "CW Reverse." Other manufacturers have it as well, but may call it something else. Helps with close-in QRM.

2) I had an initial-turn-ON reliability problem that plagued me for the first several months. But ever since hard-wiring the rig's GND post to the power supply's AC POWER GROUND (the UL Green wire in a 3-prong AC connector), the problem's basically gone. The problem was an instant "hang" to an error message on the screen directing me to insert the SD card, no other operational response at all (so the end of the line!), but taking that action didn't solve anything! The rig would acknowlege but just ignore the SD card contents. Awful! But I'd now say far fewer than 1-in-100 starts by now results in this "hang" requiring a full power-down, then back on reset by using a special restart* procedure (hold down the FINE and LOCK buttons as you hit 0/1 button to turn the rig ON). BTW, the "early" Operation Manual that came with the rig omitted this item completely, but the updated Op Man FTDX10_OM_ENG_EH076H200_2102K-CS-6.pdf, has it on page 114. I suspect that this "hang fault" resulted from static discharge from my finger (which spark you can definitely feel) to the rig's chassis, usually the power ON/OFF (0/1) button, resetting the entire internal processor. Yes, I have carpeting in my shack. Yaesu first advised returning the rig to them in the south LA area, which I did. They apparently did nothing with it except execute that special restart (see above), which I knew nothing of due to the missing note in that first Op Man. For a while after that, I just did the special restart with subsequent SD reloads. But upon installing the GND connection from the rig to the Astron PS, the problem's now history. I haven't seen it for more than a year now.

3) The rig's parametric equalizer for receive works pretty well. However, the equalizer for transmit cannot save you from a lousy mic! That freebee hand mic Yaesu put in the box just never satisfied listeners. Buying a Yaesu MD-100 A8X mic and base, respectively, (on eBayTM, since it's no longer made) solved everything - now a "Top 40 AM Superstation" audio report from former broadcast engineer I chat with on 40m. The mic is the key! Just to note, the MD-100 is a (D for) dynamic mic (powered by the sound pressure level of your voice!), yet the DX10 has enough mic gain to handle it easily. The A8X base provides a ham signal profile that rolls off the bottom and the very high end of the communications-grade voice audio spectrum. Sadly, Yaesu no longer makes the MD-100 A8X. The current Yeasu offering, their M-70, is a condenser (electret) mic that works very well (I frequently listen to them on 40m) and uses the mic cable power available from the DX10, thereby requiring very little mic gain in the rig. But the DX10 has great plenty of mic gain for, probably, any dynamic mic.

The main complaint I hear about the DX10 is the arrangement of buttons and knobs around the main tuning knob (and its outer function/tuning "ring). People say it's awkward to use. I don't find it that at all, but maybe I'm just used to it by now. The layout and the touch menu system is imperfect, of course, but it's totally good enough in my view.

Overall, more than two years into using it, I'm still glad I bought this rig. Just a CW jewel and great SSB to boot.

 

* Caution: Doing this restart RESETs all your FTDX10 memories and parameters to Factory Default! So this is IMO just for the desperate, say a Field Day op gone bad in the middle of nowhere. A true Hail Mary. If you have an SD card with your memories and parameters SAVED, a good recovery, so then it's mostly a one-stop return to "as before." But an unnecessary pain if you can avoid it with effective grounding of the rig.

 

 

I'm planning a homebrew DSP QRP XMTR and auto TR switch. I did a design based on an Analog Devices DDS. More bucks than a clock generator chip, but fabulous output specularity and controls for amplitude, as well as phase (and frequency, of course), on-the-fly shifts without phase discontinuities. If you're interested in software DSP-based design and helping jumpstart a group to create a modular QRP rig, check out the GlobalSDR idea.

The knowledgeable homebrewer may ask 'where's the TX output band filter?'. This design may not need one at all! This is because the DDS chip has an inherently clean output. Harmonics and spurs (other than those due to the sampling clock, which are suppressed right at the DDS chip output by the "Interp(olation) Filters" shown in the block diagram below.

The DSP QRP TX idea looks like this (exclusive of considerable software) in block diagram form: