LW-MW-SW Relay Tuned 15' Noise Reducing Vertical Antenna

[vk7jj]

Hi Thomas,

The 40m vertical on the roof is a very good proxy for the TCM ... on 40m.

In consideration of your earlier remarks about ground rods the photo was included to show you that the ground plane was really good one.

My antenna isn't worth keeping in the air because it's too noisy even though I live in a lower noise regional area, and that should give you some insight into whether I think it's worth building a TCM, which is what you asked of me.

But, if you'd like to hear noise on all the other bands too then go right ahead and build one

The comparison with the delta loop was not meant to be a comparison of size but of principle.

The point is that when it comes to choosing an antenna that's likely to provide you with a better SNR then LOOPS WIN large or small, and there are plenty of small ones around that use negligible realestate that work really well down to VLF.

I also think that a lot of people don't actually get off their bums and look to see if they can manage a wire loop, people are inherently lazy and stick up any old end fed wire simply because everyone else on this forum does. Deltas require the same number of mounting points as a dipole yet they are better than dipoles.

A 20m horizontal delta is tiny, if people have a back yard of any sort then maybe they can find two mounting points within 6m of their house.

A 20m delta with a zip cord feed would provide superlative performance* from 20m all the way up to UHF and given today's noisy environment provide immeasurably better performance than Mr Lankfort's vertical whip, on any band.

There is nothing special about a 20m loop, it's just a random example to make a general point that, just like whips or long wires, they don't have to be a particular size to be useful.

Cheers, Phil

--

*very happy to prove that by modelling or pointing out facts such as a wire loop works on both odd and even harmonics and does not present the extreme values of reactance that ground planes or dipoles or end fed wires always must, loops tend to provide a feed point impedance that matches more easily to inexpensive parallel feed lines thus requiring no balun, and parallel feeds are really good at avoiding the plague of common mode noise, they are not as lossy as coax as you go up in frequency, etc etc.

--

PS. Talk of "extreme values of reactance" is probably meaningless to many people so perhaps some numbers might help.

For the 15 foot vertical that is the subject of this thread, here are the radiation resistance, reactance and SWR figures at the two frequencies it claims to work between as calculated by MMANA

30MHz
radiation resistance = 2056 ohms
reactance (jX) = +11343 ohms
SWR = 1291 : 1

150KHz
radiation resistance = 0.002 ohms
reactance (jX) = -32715 ohms
SWR = 1999999 : 1

radiation resistance: for an antenna to work well with a radio receiver, it's radiation resistance has to be "matched" to the feed line impedance which in the case of coax is 50 ohms

reactance: the lower the reactance the better and zero is an achievable goal; matching systems and antenna tuners try to get rid of all reactance in order to match the antenna to the feed line. High reactance is hard to get rid of and causes matching systems to become so inefficient that much of your signal is lost.

SWR: the lower the better, eg. 1:1 is best, it is a measure of how much antenna power is able to be transferred to your radio. That applies both on receive and transmit. (I'm happy to argue the case for Rx Tx reciprocity)

[vk7jj]

By coincidence, an excerpt from an article dated June 2017 I stumbled on during breakfast reading this morning:

A simple random/long wire antenna used with a 9:1 unun is already quite wideband, although it will certainly be more resonant at some particular frequencies depending on the length.

Electric field mini-whip antennas can usually cover the entire HF band, but they are usually too noisy in today’s urban environments as they very easily pick up noise generated by modern electronics.

On sdr.hu you’ll see that most setups using miniwhip antennas seem to have poor reception. In our opinion, the best choice is a wideband magnetic loop antenna like a Wellbrook Loop as these don’t pick up local electrical noise that easily, and can be easily rotated to null out noise sources.

However these antennas can be very costly. An alternative is to build one yourself. If you have a large backyard, you might consider a delta loop, large dipole or TC2M antenna.

How about that, this thread is well behind the times. Though I'd still get into a fight with the author regarding his first sentence, "simplistic handwaving" is what I'd call it

Ref. https://www.rtl-sdr.com/a-review-of-the ... twork-sdr/

[vk7jj]

The W6LVP loop has been mentioned quite a few times on the Forum as an alternative to the Wellbrook.

I've never even seen a Wellbrook or a W6LVP but I have had some really good fun correspondence with Larry W6LVP, the originator of that loop, which started via mutual interest in using WSPR to measure noise.

Larry is a really interesting person whose passion for measuring and reducing noise is amazing, the lengths he has gone to would take your breath away and he has extensive experience. One time he sent me a pic of a computer screen hanging on his wall that redraws a chart of one of his noise test projects in real time, rather cute.

larry.jpg (129.41 KiB) Viewed 84387 times

It's nice to know there are people out there who really care about that last dB of noise and it was my honour to have created the chart software at his request so I was rather chuffed.

Phil

[vk7jj]

It's a rainy morning, so...

The previous figures given were just begging for further calculations. Here are the figures again

30MHz
radiation resistance = 2056 ohms
reactance (jX) = +11343 ohms
SWR = 1291 : 1

150KHz
radiation resistance = 0.002 ohms
reactance (jX) = -32715 ohms
SWR = 1999999 : 1

--

1) When the reactance is -ve (150KHz) then the antenna has capacitive reactance and in order to be matched that needs to be cancelled out, usually by a loading coil as mentioned previously.

At 150KHz MMANA was unable to calculate the inductance as it caused a calculation overflow which terminated the optimisation routine when the inductance required reached 9999.9 micro Henries and at that point the capacitive reactance was still -23346 ohms

Thus it is not realistically possible to match the antenna anywhere near the claimed 150KHz. The Noise Reducing Vertical Antenna diagram shows a very modest coil/inductance that could not conceivably be effective anywhere near the low end claimed.

You can see the problem when MMANA says that the too small 9999 micro Henry coil needs 2061 turns of wire on a 10 centimetre former. And the resistance of that coil would be in series with the 0.002 ohms of antenna radiation resistance. So if the coil had a resistance of say 10 ohms then the received signal is lost according to the ratio of those two resistances. Not good.

2) When the reactance is +ve then the antenna has inductive reactance and that needs to be matched using a series capacitor.

At 30Mhz MMANA calculated the required capacitor was 24pF, lowering the -32715 ohms to 0.86 ohms. The relay switching and coil arrangement in the diagram does not have the option to switch in a series capacitor, it keep a series inductor in circuit making things worse not better.

In both cases, even with the right inductor and capacitor an antenna tuner would still be required to match the 2056 ohms or 0.002 ohms to 50 ohms.

Worse than that, at every frequency in between the two extremes different values of inductance and capacitance would be require and the antenna tuner would have to be adjusted.

--

Looking at antenna tuner losses:

Rob Sherwood is famous for his radio receiver shootouts also has a page on antenna tuners, here:

http://www.dj0ip.de/antenna-matchboxes/ ... shoot-out/

The spreadsheet he links to shows that even a small difference between the antenna's feed point impedance and 50 ohm coax creates substantial losses. eg. here is just a tiny snippet from the spreadsheet showing a few tuners. At the top of the page the "antenna impedance" values are varied between 6ohms and 160 ohms while testing on the 160m ham band (1.8MHz) which is as low a frequency as he tested.

We can immediately see that even with such a modest miss-match there are ATUs that show losses of worse than 50%.

One can only imagine the losses when trying to match the 0.002 ohms or 2056 ohms provided by the 15 foot whip.

--

OK, I'm not trying to flog a dead horse here. I'm trying to provide solid real world evidence of what is really going on with antennas and how easy it is for people to make outrageous claims that others believe because the don't know any better.

The bottom line:

Does it matter what antenna you use? Not a scrap if you can get to listen to what you want to and state of the art receivers like the RSP series are absolutely staggeringly good and can hear vast quantities of signals on just a meter of wire hanging off their antenna ports except of course where noise is a problem.

But it is good to know a bit more when you decide to go a bit deeper, hope the above helps.

Phil

[Mike2459]

Phil:

Sounds likes your modeling software might not be up to the task. Is it the freeware version? Or maybe a case of GIGO.

I you get a chance take a look at this:
Optimizing the Receiving Properties of Electrically Small HF Antennas
by Steven R Best

Link to pdf at IEEE Explore: https://www.google.com/url?sa=t&rct=j&q ... 61iTjKrQAn

Quoting the author:

In Section 4 of the paper, we discuss impedance matching considerations for the receiving antenna. We illustrate that impedance matching is a secondary consideration, as a matching network designed from the transmitting perspective may not improve the antenna's performance for receiving. In Section 5 of the paper, we discuss external noise, noise figure, and signal-to-noise ratio (SNR) for a typical receiving system. We present formulas and design guidelines that aid in optimizing system performance from a signal-to-noise ratio perspective. Finally, we present design examples for electrically small dipole and loop antennas, comparing their performance in a typical receiving system.

I brought this topic up earlier this year https://www.sdrplay.com/community/viewt ... f=8&t=4416 , needless to say it wasn't received well. My only point at the time was to illustrate that much more can be gained by reducing local noise from RFI, IMO.

Best Regards,
Mike

[vk7jj]

Hi Mike.

It's not up to the task, it couldn't do it. It's a freeware version of MMANA running on Windows XP and I dusted it off for this thread so I could throw some representative figures out there for people to get an appreciation of and to use as the basis for the commentary.

Not only that but as Ive said I've focused wholly on the behaviour of a perfectly conducting vertical on a perfect ground plane so it's a theory exercise to show what I think is really really important, eg.

Antenna characteristics change radically with frequency so much that an end-fed antenna's feed point impedance can change from just a few ohms to effective infinity as you change frequency, that people should understand there is no magic solution, and that if you can hear with a given antenna then that's just fine.

--

In the real world the circuit would not work as a ground plane at all, as sdrom33 pointed out it doesn't even have one.

Earlier I said the antenna as illustrated would likely work as an off centre fed vertically polarised doublet with the ground wire and earthing system being the non-obvious second part of the doublet.

It follows that if a 15 foot whip has all the complications that I've been rabbiting on about, how does the earth wire in the diagram behave?

The author said nothing about it so we don't know how long it is but we do know for sure that it's not going to work as some sort of pipe that sucks unwanted noise and interference into a black hole at the end of an earth stake!

It could even be 15 feet long and thus the other half of a vertical centre fed dipole at whatever frequency a 30 foot dipole resonates.

So... it is not really an earth wire, it's an element, a part of the antenna.

--

There is also the argument that people make that things are different on receive. But at the interface between the antenna and feedline, all the same rules apply to Tx and Rx in terms of power loss and power transfer. At the interface between the feedline and radio things might be different, the real-world impedance of the input stage is unknown, eg. in my old Atlas 210x the first device is a diode ring mixer after the band pass filter. But does the band pass filter require a specific 50 ohms at the input? Who knows. So that debate become supposition and no side can make an absolute claim, so surely it's best to try to ensure a match to 50 ohms and then expect optimum performance.

Where do we go with this discussion and how does it end?

I'm half ashamed at indulging myself, normally never rave on like this but I stuck my neck out and here we are, it could become addictive!

[vk7jj]

Hi Mike, have read your referenced article and forum through, really good fun, thanks for the ref.

Clearly I wrote the bit about receive above before having the chance to absorb your ref. which I still really haven't had time to appreciate, but in saying what I did about Rx/Tx comparisons I'm not unhappy with my response, though it doesn't have the requisite gravitas.

As an aside, my current 40m WSPR receiver is an old Kenwood 120S with a frequency agile GPS-DO feeding the VFO input. On WSPr that receiver outperforms everything else I've got.

To take your point, when I initially set it up I just dropped a bit of carrier to stimulate auto-tune an SGC-230 on the 40m WSPR frequency.

Later, when I got serious, I substituted an old MFJ manual tuner, one with a large roller inductor inside, and laboriously peaked it by hand using a remote antenna fed from a signal generator. That did make a noticeable difference in receiver performance, indicating that the Tx/Rx matching were not coincident.

Since then, with the preselector build, I substituted a preselector for the MFJ and then retuned that using the remote sig gen and obtained quite a nice improvement. I've no way of knowing any absolute values but the fact of needing to do it was interesting in that a) not only were Tx and Rx not coincident, but b) matching the Rx does definitely made a meaningful difference.

Phil

[Mike2459]

Phil are you sure about this:

n the real world the circuit would not work as a ground plane at all, as sdrom33 pointed out it doesn't even have one.

I can see in Lankfort's schematic a very good ground plane - that 4 to 8 foot long copper coated earth ground - the best ground plane you can find with the right soil conditions. Where Lankfort is from (Southern Louisiana), I can garoontee (as they say down there) they have the best soil conditions you could hope for. Looks really simple. It work's the same way as a car radio antenna, you know the kind the losing team always breaks off after the football game. Only difference is the RF transformer is built into the radio and goes to ground in the radio's chassis. Same result. The received RF flows through the primary is stepped down to the Z of the feedline and goes it's merry way to the RX stages of the receiver.

Mike

[glovisol]

Hi guys,

You can take a liter of water, beat it in a mortar for as long as you like, but in the end, what have you got ? Only water (in reality my professor of physics at the university was using another material for this example). This is the impression the article on "small antennas" Mike referenced made on me. A lot of formulas, taken from other references, many without a dimensional definition of parameters, in order to impress the reader and make him lose the fact that the end results are vey meagre and already known conclusions, e.g. that the only small antennas worth using are (perhps) ferrite loaded loops and that mismatch and attenuation are acceptable only as long as they remove signal and noise ABOVE the noise sensitivity of the receiver (Mike, are you listening?). Small, pathetic result for beating so long and learnedly into the bush. Just by carefully reading the Recommendation ITU-R P.372-13, ITU 2016 on propagation and noise (also cited by the Author, but without any meaningful or useful conclusion) one would learn much more on the subject.

And Mike, it is always a matter of surprise to me to see intelligent, experienced and knowledgeable guys always falling to the delusion that one can get all for nothing, or finger sized antennas for serious reception of radio waves, be them 15' whips with impossible impedance ratios invented by quack doctors now lost in the arctic ices and touted by near paranoids, or small pieces of PC board cum FET amplifiers sticking on top of fibreglass (for some) or (God forbid!) on steel masts for others. It is the same old story like the philosophal stone that would transmute iron into gold. Yes Mike, we can speculate and discuss for a lifetime, but science is based on the principle that the experiment, to be valid, must provide the same result when experenced by any operator, at any time and in any part of the Universe (this is Galileo, the father of Modern Science, so uncoungrously cited in this very thread!). And science, serious science, I am afraid, cannot contradict the rules of physics.

And Phil,thank you for entusiastically bringing daily your drops of scientific knowledge and honest and true common sense.

Thanks to all, this is a really interesting debate.

glovisol

[Mike2459]

Glovisol:

You must be a fast reader and learner. To go through all of that material and come to a definitive conclusion so quickly, is simply amazing. Color me impressed. And I thought the IEEE was a respected organization. To top it off they have the gall to claim to have the top 38 out of 40 journals in Electrical Engineering, Electronics Engineering and Telecom. Perhaps you should rebut some of the misinformation in the paper I cited. Let us know when it is published. I'm looking forward to studying it.

Mike