Another TDR Question--Driving antenna

Seems coincidence to me.

Consider: the ground rod doesn't stop being a transmission line, it's still a piece of metal in space. What is its impedance?

What's more -- since it's a transmission line connected in series with your antenna transmission line, what effect would that have on your measurement? This also brings in probe and scope effects (probe cable length as a TL stub, scope as a bulk capacitance more or less), which introduce new considerations to common mode error and ground impedance.

Or, I read that as grounding at the scope end -- but it applies to both ends equally.

And, being an antenna, how have you estimated its radiation loss with respect to frequency, and how would that affect the measurement of apparent impedance? (I wouldn't expect you to actually /know/ this -- well, unless you plugged it into NEC2 or something. But hopefully, mentioning it will lead you to think about it. :) )

All in all, it may well be that your soil acts as a short length transmission line, whose width increases with depth, and consequently, its impedance drops, and it becomes lossy. And then it just so happens to come out to about the same resistance at any frequency.

It's plausible that the soil has a lossy capacitance characteristic, or a network-of-resistors characteristic, and in either case ends up with an impedance that tends to be surprisingly close to the DC figure, despite everything. It's still coincidence, but you get to nudge your nose when you say it. :^)

You'd have to do a more complex measurement to figure this out. Say, grounding a monopole antenna to it -- so you have a source of displacement current "to space", so you don't have to worry about, say, the transmission line characteristics of your wire antenna. Or, well, that would be fine too, as long as you have some means of calculating ground effect along the length of that antenna -- that is, treat it as a one-end-shorted stub, and figure out how lossy it is at various frequencies, subtracting the contribution from radiation, which then must be due to lossy ground.

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Design Website:

Backup and decide what you're actually trying to accomplish.

If your objective is maximum directivity, don't you want the wave to travel down the wire and not reflect? Doesn't no reflection step in the TDR makes that happen? Doesn't matter where it goes as long as it doesn't go back up the wire, or radiate back into the wire.

Put the TDR back at the receiver and look for a step created by mismatch at the junction of the feedline and the wire. Match that too.

You can obsess over the math, but optimizing the symptom you're trying to optimize should be the best you can do.

A TDR like that is not going to be very useful for matching an antenna anyway, as an antenna is usually tuned to a specific frequency and you want to measure its response at that frequency, while a pulse as used by a TDR of course is composed of many different frequencies. The TDR works fine when measuring open/shorted cable or resistively terminated cable (wideband termination), but when measuring a load that is not wideband, like an antenna, the response will always be off.

For this task you should use a vector network analyzer. Simple network analyzers specifically made for analyzing antennas are available from sources like ebay as well.

I appreciate you taking time to drag me into thinking.But I'm not sure it got me anywhere. Much of what you reference is beyond my comprehension.

As always, more than one thing. I just want to characterize the antenna for my edification. I have never ran an impedance graph of an antenna so I did. 60kHz to 35MHz.

I want to make the antenna as directional as possible with the length I have at the lowest frequency I can. AM band is my objective, but I listen on 160Metrs and 80Meters. This is the best example that I have of silencing some of the co-channel interference. Using the BOG.>

I can make the antenna act longer by inserting inductance along the line, according to this. > But apparently you do not want to get below .5 as a VF. I have it measured as .78. I forget how I got into measuring the ground resistance, but I found it interesting and more data to add in.

That's what I expect. Limited my the apparent length of the antenna.

I don't have a clue what the impedance would be, but compared to a normal transmission line, the L's and C's or going to be swamped by a low R.

I think it will be minimal since it is mostly resistive. I don't see any impedance bump in my scope display from ground, But maybe I'm seeing reactance when the rising edge hits the termination, it is not a sharp rise, it is rounded. Next round of TDR tests I'll get some scope screen shots.

God, I hope they're minimal, and can be ignored. Here's the device, it plugs directly into the scope.

It's plausible that the soil has a lossy capacitance characteristic, or a network-of-resistors characteristic, and in either case ends up with an impedance that tends to be surprisingly close to the DC figure, despite everything. It's still coincidence, but you get to nudge your nose when you say it. :^)

You'd have to do a more complex measurement to figure this out. Say, grounding a monopole antenna to it -- so you have a source of displacement current "to space", so you don't have to worry about, say, the transmission line characteristics of your wire antenna. Or, well, that would be fine too, as long as you have some means of calculating ground effect along the length of that antenna -- that is, treat it as a one-end-shorted stub, and figure out how lossy it is at various frequencies, subtracting the contribution from radiation, which then must be due to lossy ground.

Tim

I'm stuck using the equipment I have to do what I can. Scope, Cheap TDR an antenna analyzer and a signal generator. My goals at this time are improve the Ground rod resistance. Probably add one 8ft rod at each end and add a liberal of calcium chloride at both ends. And redo the measurement to see if I gained anything. Then get a good measure of the characteristic impedance of the BOG (if I can come to a conclusion) use that number to build a matching transformer. Then start trying to figure out what size inductors to put in the antenna to slow the velocity factor, put those in and start all over finding the characteristic impedance and a new matching transformer. Mikek

We agree on all this.

I have a matching transformer at the feedline end of the antenna.

No obsessing over the math. Just trying to optimize what I have. OH, I think you said that! Mikek

Ya, I'm not tuning to s specific frequency, I'm trying to optimize the directional characteristic. Enlarge my cone of silence.

To all, Let's suppose I don't have a TDR and I want to find the proper Resistance to terminate the wire to prevent reflection. How would you find that resistance?

If you had a TDR would you use that to find the proper resistance?

Mikek

That works if you know the impedances and frequencies involved everywhere and construct the match accordingly.

I have no direct experience with this...I'm just thinking out loud.

I fear that you're assuming equivalence between frequency and distance where it doesn't apply.

Consider a dipole. At the resonant frequency it has a reasonably characterized driving point impedance. It couples well to the ether in the broadside direction. The reason you care about matching at all is so that you can get the maximum amount of energy into your detector.

Shorten the dipole. You reduced it's effectiveness and messed up the driving point impedance. You have less energy captured and lower ability to transport that energy to your detector.

Now, put loading coils to resonate the dipole again. You have restored your ability to transport captured energy to your detector, but have you improved the amount of energy captured at all? That's a question, because I really have no idea.

If I understand your antenna, the preferred direction is along the axis of the wire. I'd expect, for low angle of incidence, that the wave traveling along the wire stays approximately in phase with the wave propagating in the ether. When you put loading coils in the wire to "make it look longer" you might think it's 'resonated' in the sense of broadside angle of incidence, but you're putting a speed-bump in the wire that disturbs the phase so that the on-axis incident wave gets ahead of the wave in the wire and reduces the energy captured. If we assume that the propagation in the wire is slightly slower than propagation in the ether, you may want the wire to look 'shorter'. Stated another way, the preferred direction is slightly off axis by some function of the relative propagation velocities.

I submit that the best you can do is to match the antenna to the feedline to optimize energy transfer to the detector at the frequency of interest.

To optimize directivity, terminate the other end to absorb signals coupled into the line from the opposite direction. I'd expect that is a function of frequency and that you probably won't be able to discern that with a conventional TDR.

Even tho the relationship between time domain and frequency domain is well established, the devil is in the details. For these types of problems, actual frequency measurements at the frequency of interest are much, much easier. Like tweak the termination to minimize a signal from opposite the preferred direction.

I'm not sure that the concepts of impedance as a single complex number and resonance can be of much help in this situation. They exist, but conventional application may lead you astray.

I've been struggling to find a way to explain my understanding of length vs pattern. And that's it-- It's length vs pattern. It's not length vs frequency or length vs energy captured. This chart shows pattern vs wavelength.

Notice as the wavelength gets longer the pattern gets more directional. That is what I'm after by making my antenna seem longer. At 1200kHz my antenna has deep nulls to the East and West, I don't know about the F/B ratio, The backside of the antenna points at the West tip of Cuba. I've looked up a couple stations there, but have not been able to receive them., Just realized, I could open my termination to recieve of the back side.

Here's a graph of the Resistance of the BOG wire, notice particularly the section I marked as the AM band 540kHz to 1700kHz. The resistance stays fairly flat around 400 ohms*.

• my antenna analyzer like most, is not accurate when far from 50 ohms. That 400 ohms is probably closer to 325 ohms.

Yes, But I hope you notice the resistance of my BOG is pretty flat over a 10 to 1 frequency range. My point is a Beverage antenna is not your standard antenna and a Beverage on the Ground also does not fit the usual graph of multiple resonances. Mikek

0

Just want to make sure something is clear. It's not the wavelength that is getting longer in the charts, it is the antenna relative to the wavelength . So you could do the same thing by keeping the antenna the same size and using a shorter wavelength... or am I missing something?

Rick C.

That's right. That antenna is getting longer, thus more wavelengths at the fixed frequency. Mikek

"Notice as the wavelength gets longer the pattern gets more directional." Restated to say what I meant to convey, Notice as the antenna gets longer (more wavelengths) the pattern gets more directional.

Mikek

I don't know how to represent parasitics in the soil around my ground rods, but I do notice that the ground parasitics are isolated by either the termination resistor or the Resistance reflected back from the radio through the transformer. I tried to represent that here.

I'm not sure what I'm arguing here, because I can agree that an RF ground will be different than a DC or 60Hz ground. I guess I'm saying, "ya but" all those parasitics come after a big as resistor, or maybe not so big, but it's there on both ends.

Rechecked my ground resistance today, after a couple days with the calcium Chloride in the ground, the resistance dropped a little more from 73 ohms to 60 ohms. I reconnected everything and ran some, BOG to Long wire comparisons in the upper BC band. 1190kHz is interesting, a different station on each antenna.

Mikek

if you want a low frequency antenna for receiving only and you want it directional..

consider making a resonant loop

they have deep nulls.

you can make it very small and physically sterrable

ie a transistor radio.

mark

I'm in dire need of a highly directional rx antenna for VHF & UHF right now. Any suggestions? Pref something that could be made up at home from wire to save the time wasted waiting for a mail order job to turn up.

--
This message may be freely reproduced without limit or charge only via  
the Usenet protocol. Reproduction in whole or part through other  
protocols, whether for profit or not, is conditional upon a charge of  
GBP10.00 per reproduction. Publication in this manner via non-Usenet  
protocols constitutes acceptance of this condition.

"Highly directional" and "VHF & UHF" (i.e. wide frequency range) is tough to do in a single antenna, at least in any reasonable amount of space. Wide-frequency-range directional antennas tend to be log-periodic or LPDA, and they're only modestly directional.

If you have specific frequencies in mind, individual Yagi antennas are probably a better bet... they have narrow bandwidth but can be quite directional.

Consider the "Cheap Yagi" designs by Kent Britain WA5VJB. You can make them from a stick of wood or some PVC pipe, some stiff wire or brass or aluminum rod, and some coax. Pick one of his example designs that's fairly close to your intended frequency, and scale all of the dimensions.

If you do not need to turn it, look for a rhombic antenna.

--

-TV

What frequency? What bandwidth? How directional? Wideband, highly directional, and 'made from wire at home' don't coexist easily--they're another form of "good, fast, and cheap: pick two."

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

What's your definition of "highly"?

Same antenna for BOTH UHF and VHF?

Numbers are extremely helpful when starting a design.

Just the best I can find.

That would be nice, but like Phil said, you can't have everything in one antenna.

There aren't any specific frequencies. I just want to hook up my main spectrum analyser and checkout the 'RF landscape' up to about 15 miles around me for fixed and mobile stations operating from about 70Mhz to

500Mhz. Something capable of DFing would be nice, but I'm guessing that ain't gonna happen. At least two pluses are that I won't be using it for transmitting - and gain is not important.

--
This message may be freely reproduced without limit or charge only via  
the Usenet protocol. Reproduction in whole or part through other  
protocols, whether for profit or not, is conditional upon a charge of  
GBP10.00 per reproduction. Publication in this manner via non-Usenet  
protocols constitutes acceptance of this condition.