The AM Forum

Hi to all:
Is there an antenna analyzer that is designed for open wire feeders, i.e 300 ohm, 450 ohm, 600 ohm etc., that is accurate?
Chuck

Hi,

 Well I think a good place to start is what volume of signal are you planning and what type of equipment you plan on using Commercial. Homebrew give the group an idea of your station
set up and plans..Then go from there...


73
Jack
KA3ZLR

Hi Jack:
At present I am using a center fed 75 meter doublet test antenna.  There will be different feed-lines ranging from 300 ohms to 600 ohms and possibly a little higher.  The rig in use for testing will be a FT-101FX or a Drake C-Line or a Drake TR6.  The power range 1-120 watts 160 - 6 meters.
Chuck

Hi again,

 Here I found a good list of Analyzers I want you to know they can get quit expensive it's on eham.

http://www.eham.net/reviews/products/31

Here's an antenna coupler for Balanced lines alot of us built and use you can design it for your power situation:

http://amfone.net/ECSound/K1JJ13.htm

Here's a Neat little device you can Build yourself monitor your feedlines man:
(Click on Antenna Current Meter)

http://www.dxzone.com/cgi-bin/dir/jump2.cgi?ID=25435

Hope this Helps a little bit O.M. :)
And remember building is more fun than Buying.. ;D

73
Jack
KA3ZLR


 

There does not seem to be an analyzer for open ladder line. The ones listed in eHam are for 50 ohms.

The best way is pretty simple with just some small indicator lamps with their wires coiled on the actual ladder line and you tune for max brightness. If there is unequal brightness then you change the taps on the tuner components to have equal brightness.
The rigs you are using are unbalanced 50 ohm output. You're kinda stuck in using a tuner to transform the ladder line or open wire feeders to 50 ohms to make the radios happy.
Fred

This was nice info from eHam:
http://www.eham.net/articles/9751

I give up,

Mop you don't read much yea that's what I found looking quickly it's a good bunch to look at
but of No good use...Then......................hard to figure that..

I listed a Coupler that he can build and use very effectively and a Metering System he can build and use effectively he doesn't need an Analyzer...Just an Effective Coupler and a Way to monitor his Feed lines..

Just step on over me man......

Cheeze..........the gift of my post was Building what is needed...it's right there in front of you

Damn..

If you know the true line impedance then you can wind a binocular core transformer to match the 259B or other analyzers.

I did that for my 2 wire Beverages using unknown cables, measured wire size and spacing, and went to an online calculator for the impedance. Use a carbon pot on the transformer to confirm a match where its supposed to be.

Carl

Or better still, thermocouple  rf ammeters, preferably two each, one in series  with each feeder.

But shifting taps on the tuner coil to achieve equal brightness does not guarantee balance in the feeders.  The currents may be equal at the point of measurement, but what about 1/8λ or 1/4λ down the line? The standing wave patterns on the two feeders may be significantly different, and adjusting for equal currents at one isolated point would do nothing but give a false sense of "balance".

Assuming the OWL is working into a  reasonably symmetrical load, unbalance in the feeders is most likely the result of common mode "antenna" currents along the feed line.  Each feeder is biased with an additional current, usually of the same amplitude and phase with an identical standing wave pattern, and it is the sum of the normal (differential) current and common mode current that is indicated by the meter. Grounding the tuner and futzing with taps on the coil is apt to make matters (common mode currents) worse.

The best way to avoid common mode currents is to use a balanced, link coupled tuner, with no direct ground connection anywhere to the main coil. The separate link coil isolates the main coil from ground, allowing it to float. An rf choke attached at the "cold" spot on the coil is OK for taking static charges to ground, without affecting the RF. Even with a floating coil, it is sometimes possible to still have common mode currents, particularly if the length of the feedline is  such that a current node (high rf voltage point) occurs right at the end - IOW, the transmission line may act like an ungrounded halfwave (or other even number of quarterwaves) "no radials necessary" vertical (including its inherent losses).

The only way to check for true balance is to use two sets of lamps or rf ammeters at two separate points along the line. Avoid placing these two points close to a half-wavelength apart at the operating frequency. Although unbalance is often, if not usually, a result of inadvertent, parasitic, common mode currents, it may also be the result of an inherently unbalanced antenna. An example of this would be the half-wave end-fed zepp. The "dead" feeder that is attached to nothing will not have exactly the same current as the active feeder, even right at the resonant frequency of the flat-top. As the transmit frequency is moved away from exact resonance, additional unbalance inherently occurs as the current node on the active feeder/flattop combination slides away from the exact point where the end of the flat-top attaches to the active feeder, moving to a point either up on the flat-top or down the feeder, while the current node on the dead feeder is fixed in position at the unterminated end of the wire, regardless.  This is an unavoidable trade-off with the end-fed zepp, and is usually not a significant problem if precautions are taken to avoid common mode currents via a grounded tuner coil, thus preventing the feeders from serving double duty as a (poor) grounded vertical  radiator.

From what I have seen, the most promising Antenna Analyzer for Open Wire Feeders is the Micro-match, as described in a late 40s QST. It is designed to measure SWR in a 50-300Ω balanced transmission line.  I believe that, with careful construction, it could be made to work up to 600Ω, at least on the lower frequency bands. My General Radio antenna impedance bridge is strictly designed to measure an unbalanced load with one side grounded.  I'm not sure if a commercial unit exists that is designed to accurately measure the impedance of a balanced load to ground.

OK, Now that the Lecture is over did we read his Requirements  anyone..?

A 75 meter Doublet capable of operating 160-6 metrs with a potential
of 100-150 watts of Power and there will be changes made in the feedlines.

Rather simple I'd say I put up there JJ's coupler and a feed line metering system
capable of handling his power requirements...that he can build relatively inexpensively
and  he doesn't need thermal couplers  they'll stall out his signal before it gets to the
friggin radiating elements sheezee... he's operating QRP NOT QRO CCS.

Unbelievable...isn't it better to try an suggest inexpensive ideas...

Best 73 bud...good luck on your system..
Jack
KA3ZLR

Chcuk, since you're obviously going to feed it with coax from the xmitter into a tooner of some sort, there's really no reason for a balanced line antenna analyzer.  Not sure what you'd want it to tell you.

Treat everything upwind of the xmitter as a system, that you want to appear as a 50 ohm resistive match to the xmitter.  Just affix a MFJ or whatever analyzer to the coax, go to town with the tuner, mark your settings, and be done with it.

Eggxactly.. what do we need with the analyzer.....My thoughts also, I think he should build the "SupaTuna" as Described and save money and build the Antenna Current Meter for practically
nothing , all he needs Quote JN "Upwind" respectfully is a VSWR meter...Simple play with all the
line inputs ya want...Keep yer money in your Pocket....

73
Jack
KA3ZLR

P.S. And hey if he really needs to see what's what's Get a decent Dip meter
       they're cheaper and werk just as well...

http://www.ebay.com/itm/EICO-MODEL-710-GRID-DIP-METER-/300599725452?pt=LH_DefaultDomain_0&hash=item45fd23cd8c

Clearly the Grid dip meter would be the best way to me. I have a book by sams on how to use a grid dip meter and one of the sections is many ways to use the meter when building and tuning an Open wire line antenna.  Maybe I can photo copy the pages for you.

As for balance. You can use Xmas Tree bulbs. Not the small ones. But the larger type. Please look up kc6mcw on QRZ.com and see his schematics. 

I use a cheap MFJ balanced line meter.  It shows current up to 3 amps per leg on two meters.  This was instrumental for accurate adjustment of the balanced tuner. Not only for balance but for effeciency.  Mistuning the link coupled tuner can make the transmitter happy but the effiency way way down. The current meters will allow you to tune for the highest output current while keeping a good match to the transmitter.  Simply swapping the line left to right can sometimes fix a huge imbalance. Without the meters, you wont see this.

The last tool that nobody mentioned is a field strength meter. This is a must have to me. You can take baseline readings as you walk around the line,   When balanced, the radiation at the lines will be much lower.  Later, You can leave the meter on a shelf or table and when mistuning, you can see the IN SHack RF come way up as the line in the shack or the tuner starts radiating.

Good luck

Clark

Chuck, what isn't clear to me is what type of tests are you  contemplating?  What are you looking for in changing the feed lines? Simply the 'balance' of current flowing in the two wires? What?

Seems to me we need to know your test goal to determine what test equipment you need to perform it.

Or has my aged mind missed something already stated?

Walt

Jack, Fred, Carl, Don, W3JN, Clark, Walt and all:
I just got back from a cerebration for my Uncle.  Sorry for the delay in replying.

Lots of good information everyone provided.  I will be using a home brew link antenna coupler with series and parallel ability.  I will be able to select series with two output capacitors or series with two output inductors with the line connected in the center of these two coils.  (This is the tank coil mechanically cut in two parts with the line connected between the two coils and in series with the tank capacitor)  Traditional parallel tune will be the third option.

When the entire antenna system length is midway between a current loop and a voltage loop, at the coupler and line connection, I would like to find the reactance required to bring this antenna system to resonance.  I would like to try series inductance (both sides of line) or series capacitance (both sides of line) or shunt inductance or shunt capacitance, if needed, on the line between the coupler and ladder line.  Then, I would like to compare the reactance requirements for this midway point for different line types, i.e. 300 ohm, 450 ohm and 600 ohm.

Today, I would like to be able to calculate the predicted reactance (inductive or capacitive) required to bring the above antenna system to reaonance and compare this calculation to an accurate measurement of actual required reactive components (inductive or capacitive) needed to bring the entire antenna system length to resonance.  

No Walt, I was not clear in stating my goal.

I am not the best writer so if I am not clear please let me know.
Chuck

I had  that exact same problem shortly after I moved back here when I first put up my present antenna system, before I had installed  the ground radials for the 160m vertical. In order  to get on the band, I decided to load the 80m dipole up as a quarter-wave dipole on 160. (I didn't waste my time trying out the quarter-wave vertical with no radial system). Problem was, the feedpoint of the OWL was just that - midway between a current loop and a voltage loop.  I threw together a balanced link coupled tuner, using the split-stator plate tuning capacitor out of a BC-610 , and tried tapping down on the coil. I found a spot that gave a perfect 1:1 SWR looking into the tuner, but I couldn't modulate more than about 100 watts 100% before the 7 KV BC-610 capacitor would arc over. That tuner simply wasn't capable of handling such a highly reactive load very efficiently. Instead of futzing around with series or shunt capacitors and/or inductors to cancel out the reactance, I simply added an additional 1/8λ of OWL that switches in automatically whenever I select that tuner, which is parallel feed using a 300/300 pf 7 KV split stator bread slicer, with an additional 50 pf fixed vacuum cap shunted across the whole thing. It loaded (loads - I still use it) up perfectly, and I could modulate a KW DC input 150% positive, with no sign of arc over. But it is extremely sharp tuning - I can QSY maybe 5 kc/s before having to re-resonate the tuner coil. But most of the time when I operate 160 now, I use the vertical.


Look at the title of the thread to see to see what he was enquiring about. I didn't see any mention that the analyser had to be "inexpensive"; he wanted something accurate. Accurate and inexpensive are not necessarily the same thing.

And, thermocouple rf ammeters will not "stall out" the signal.  A good one will have negligible voltage drop and negligible insertion loss.

Hi Don:
Glad you got your antenna working on 160 with the extra feed line.   For me, I want to try capacitance (variable capacitor) or inductance (tapped coil) to cancel the high reactance midpoint between current loop and voltage loop.

I wonder how you determine if the line is capacitive or inductive reactive at this mid-point?

i.e. An antenna system total length is 150 feet long.  

246/frequency x velocity factor = one electrical quarter wave.  

i.e. 21.0 mHz = 246/21 x .95 = 11.13 feet for one electrical quarter wavelength.  Then, 150'/11.13 feet = equals 13.48 quarter wavelengths and a     mid-point between current loop and voltage loop.  

Is the above 150 foot long antenna system, at 21.0 mHz, capacitive or inductive reactive?
How reactive is it measured in ohms?

Chuck

Are there any more suggestions?
Chuck

I'll take one more shot..........
IF your total length is 150 feet and you are using 6 inch spaced ladder line, you would be good down to 75M. Building the famous K1JJ tuner would be all that you need.
I think you might be complicating things worrying about capacitive and inductive reactance on the actual system. The tuner will make your radio happy. The rest takes care of itself. Get it up a minimum of 40 feet and it will serve you well.

This link will get you started..... by looking in the various pages you will see pictures and hopefully schematic drawings of this very nice tuner.

http://amfone.net/Amforum/index.php?topic=18054.0

If I may add a few comments... in the spirit of trying to be helpful (even if some of the readers of this thread don't consider my comments helpful to them)

I believe that placing lumped circuit elements (capacitors or inductors) at the midpoint of a feedline that is longer than 1/8 wavelength at the operating frequency will be of little or no benefit in improving the performance of the feedline. The standing wave ratio (the ratio of maximum to minimum voltage and the ratio of maximum to minimum current) in the uncut sections of the feedline leading from the antenna to the midpoint of the feedline will be essentially the same. Therefore the loss (if significant) in that section of feedline and the peak wire-to-wire voltage in that section of feedline will be the same as it would be if you just used a tuner at one end.

It would be true that at certain frequencies, the wire-to-wire voltage would be less at the midpoint itself... and the effect of using lumped components at that point could be to reduce the required wire-to-wire voltage rating of the components in the tuner at the far end. But, as Don pointed out, one could achieve that same result by making the balanced feedline a little longer or shorter at that frequency.


In the telephone industry, years ago, loading coils were added periodically along long telephone lines to achieve an improvement in transmission performance... but the frequencies of operation were audio frequencies... and therefore the spacing between the loading coils was very small compered to 1/8 wavelength.

Stu

MOP and Stu:
Thank you for you inputs.  I agree, the tuner you mention is very good.  I have built it and it works well.  By trial and error with the taps and input and tank output capacitor setting changes I am able to load wide sections of the bands.  But, I would be like to have the ability to predict and measure the actual reactances required to bring the total length of the antenna system to resonance.  That is why I was looking for an antenna analyzer for open wire lines.

If you have time look at my earlier comments in this thread.  I was wondering how I would determine if the line, at the tuner, would be capacitive or inductive reactive and how much in ohms (calculated and measured)?

Chuck

A good question, and there must be a simple obvious answer, but it's something I'll have to think about, whether the reactance at the mid point between a voltage loop and a current loop is capacitive or inductive.

If the feedline terminates right exactly at a current loop, the load it presents to the tuner will be a low impedance and purely resistive. If it terminates exactly at a voltage loop, it will be a high impedance, and purely resistive.  The exact resistance will depend on the antenna being fed and the surge impedance Zo of the line.  If the line is slightly too long, but near a voltage or current loop, the tuner will see something resistive plus some inductive reactance, and  conversely, if it is slightly too short, it will be resistive plus some capacitive reactance. But if it is midway between the two, will it see an inductive or a capacitive reactance?

If you take a measurement with an impedance bridge, it will show up as some specific value of R ± jX (resistance in ohms ± some value of reactance in ohms), (+) indicating inductive reactance and (-) indicating capacitive. So if it is 1/8λ longer than what would terminate at a current loop but 1/8λ shorter than what would terminate at a voltage loop, would it be some value of resistance plus jX ohms, or some value of resistance minus jX ohms? If midway between the two, the resistive component would vary along a sine curve to some intermediate value between the adjacent high and low values, but off the top of my head I can't say if the reactance is plus or minus. Something I had never really thought about until now.

Hi Don:
This is very interesting to me.  I am looking at a graph on page 79 in the ARRL Antenna handbook, 1960, chapter on Transmission Lines.  The graph information states, "Reactance at the input terminals as a function of line length in wavelengths open-circuited line (there also is a graph for closed-circuit line).

Anyone please jump in here and let me know if I am thinking correctly.  
(1) Input terminals I assume is the connection point of the line and antenna coupler.  Yes?
(2) Open-circuited line I assume is like a dipole and not a closed loop.  Yes?

The graph shows the first quarter wavelength, 2nd, 3rd and 4th quarter wavelengths.  The zero point starts on the far left side of graph and goes to the right to the 1/4 wavelength mark (on the horizontal line of the graph).  Zero to the 1/4 wavelength mark is shown as capacitive reactive.  From the 1/4 wavelenght mark to the 1/2 wavelength point, going right on the graph, shows the reactance as inductive.

This graph is showing odd quarter wavelengths is the starting point of capacitve reactance and even quarter wavelength as the beginning of inductive reactance.  Thus, I think, (jump in here to correct me) my 13.48 quarter wavelength long 15 meter antenna system would be capacitve reactive.
I am guessing here.  13 quarter wavelengths is an odd number of quarter wavelengths.  The .48 length lets say is .5 or one half way between the current loop and voltage loop.  Another way to thick about this is .5 quarter wavelength could be a 45 degree point or a 1/8 wavelength.  

On page 80 there is a Universal reactance curve graph and two formulas.  I quote: "When a line section is used as a reatance, the amount of reactance obtained is determined by the characteristic impedance and the electrical length of the line.  In the case of a line having no losses, and to a close approximation when the losses are small, the INDUCTIVE REACTANCES of a short-circuited line less than a quarter wave in length is

XL (ohms) = Zo tan L

where L is the length of the line in electrical degrees ( i.e. 45 degrees) and Zo is the characteristic impedance of the line (i.e. 600 ohms).  The CAPACITIVE REACTANCE  of an open-circuited line less than a quarter wave in length is:

Xc (ohms) = Zo cot L

My section of line of interest is 45 degrees long (.5 quarter wavelength) represents 45 degrees.  Using the Xc = Zo cot L formula gives me this:

cot of 45 degrees = 1 x Zo (600 ohms) = 600 ohms capacitve reactive.  When I convert 600 ohms to pf of capacitance at 21 Mhz I get 12.6pf of capacitance.

Do you guys think this is correct, 600 ohms of reactance represents the mid-way point between current loop and voltage loop?  For 450 ohm line this would give us 450 ohms capacitive reactive and 300 ohm line would be 300 ohms capacitive reactive.  

Link antenna couplers seem to have their greatest challenges at mid-way points between current loops and voltage loops.  Series tuning would used when a current loop is at the coupler and line connection point and parallel tuning would be use if a voltage loop was at the coupler and line connectoin point.  But what about these mid-way points at the coupler line connection points?

To counter this 12.6pf worth of capacitve reactance, at 600 ohms, would require and inductive reactance of 600 ohms or 4.5uh.

What would be the best way to add this 4.5uH of inductance to the link coupler when we are at a mid-way point between current loop and voltage loop at the coupler and line connection point?   I have three circuit choice, (1) series tuning with two output capacitors, (2) series tuning with the tank coil split in two parts with each split in series with one side of the line or (3) traditional parallel tuning.

Could I shunt 4.5uH of inductance at the junction of the output tank capacitor and line using parallel tuning?

Could I add 2.25uH of inductance in series with the two output capacitors in series tune?

To me, using series tune with the tank coil cut in two parts would not be the way to go?  Adding 2.25uH of inductance in series with each leg of the two output inductors seems counter productive?

Please jump in here and let me know what you think so far?
Chuck





 

Chuck,  Is your goal?     What are you trying to achieve?   More bandwidth? Effieciency?


c

c
I would like to be able to know the reactance at the coupler and line juction when the total antenna system length is at a mid-way point between a current loop and voltage loop at this junction.  I can then add the correct amount of reactance (shunt or series) to resonant the antenna system.

Chuck

The cheap little Autek analysers let you set the impedance.
Not locked into 50ohm measurements.

They work at LEAST as well as MFJ...

Except no analog meters.

http://www.autekresearch.com/va1.htm

I have an RF1, the older model. I find it very handy for the price. Measuring C and L at the operating frequency is very handy. Many people put more faith in those lab units with lots of digits that only test at 1 Khz too 0.1%. At 7 Mhz that capacitance they measured might be at the high side of resonance and look like an inductor. Here the Autek's shine.

Still we have to measure right at that SO-239 connector. Just a few inches of leads make a big error when the frequency is say > 5 Mhz. There is no provision to calibrate a line extension like some of the more expensive S11 gizmo's where you cal with a 'open', 'short', and 'X-Ohms resistive load'.

The VA1 looks nice to measure SWR at say 450 ohm level. Still, we take a balanced line to an unbalanced connector. Even holding the unit in hand might unbalance the transmission line a bit. Maybe best to sit it on a piece of wood when doing tests like this.

Measuring parallel line SWR at full strap transmit power requires another approach..

Jim
JKO

Jim and all:
I would not have to measure at full 120 watts.  QRP power would be fine if the accuracy is there.

Did my work in the previous replies make any sense to anyone?
Chuck

Nowhere you know why antennas are black art.  Calculate, measure, build and install.  Then fiddle with it until it works.  Another thing, when I started, no one cared what the SWR was.  Just get the florescent buld as bright as possible and operate.

Does knowing the exact impedance help?  In lab circumstance yes.  But when you install, that all changes.

Jim:
You make some excellent points.  Did my calculations make sense?
Chuck

You are right, according to my calculations.  You base the + or - sign of the  reactance  starting at zero  feed line length.  From zero to 1/4λ, the reactance is capacitive. From 1/4/λ to 1/2λ the reactance is inductive, etc.

Your 13.48 quarter wavelengths , ≈ 13.5 quarter λ, has six half wavelengths + another 1.5 quarter wavelengths. When using the chart, you can discount the first 12 quarter λ (six half wavelengths), since the impedance at that point will be the same as the original input impedance. In your case, it is an open circuit. So really, what you  have to deal with is the additional 1.5 quarter wavelengths. The first additional quarter wavelength converts the high-Z apparent load to low-Z, still purely resistive.  So what you are left with is a 1/8λ with a low-Z input, which displays a capacitive reactance as the antenna tuner sees it.  

My 160m dipole has 3/8λ total from the end of the dipole to the end of the feed line at the bottom of the tower.  Each leg of the 160m dipole is 1/8λ, plus the two quarter waves between the feed point of the dipole and the tuner. At the half wavelength point, midway down the tower, the impedance is purely resistive and very high. At the end of the feed line, the additional 1/8λ makes  the apparent impedance capacitive. I could use lumped inductors, but instead simply added another 1/8λ of OWL in my "tennis net".

That clears up the confusion about the odd 1/8λ. Define the starting point as zero; from 0 to 1/4λ, the reactance is capacitive, and alternates each additional quarter λ.

Don and everyone:
I think I may be wrong.  Please look at my 3 attachments.  The crude sign wave graph shows the 13.5 quarter wavelength mark in the inductive reactive side.  Thus, I think my 13.5 quarter wavelength antenna would be inductive reactive and not capacitive as I first thought.  And, because it is at the .5 or 1/8 wavelength or 45 degree point it would be 600 ohms reactive.  Please see the other two text attachments to calculate for 600 ohms or see my earlier replies.

Am I correct in thinking the open circuit lines (top attachment figure 3-31) could be used for a center fed zepp and a short-circuited lines (top attachment figure 3-30) could be used for a closed loop fed with ladder line?

What do you think?
Chuck

So you can bandswitch it?

C

c:
The extra shunt or series reactance (capacitive or inductive) would be mounted on or next to the link coupler.  If needed, either one could be plugged into the circuit.  Band switching would be a snap.

Would anyone care to look at my above work to see if it is correct please?
Thank you,
Chuck

I'm still thinking about it.

From your diagram, from zero all the way to 12/4λ, you have 6 complete half-wavelengths (or 3 full-wavelengths), so that at that point you are still exactly where you started - at a high Z nonreactive point, so you can ignore those 3λ to the left.

The 13th quarter wavelength section converts that hi-Z resistive point to a low-z resistive point. You could replace the first 13 quarter-wavelengths of wire with a low-value non-inductive resistor. Following that, the additional 1/8λ starts out at a low-Z purely resistive point and terminates midway between a low-Z and high-Z resistive point. The additional 1/8λ according to the chart, shows that a less than one quarter-wavelength short-circuited line (and a low Z compared to the Zo impedance of the line has the same reactive sign as a short-circuit) would have an inductive reactance at the input terminals.  ???

I'm going to have to study that a little more, because I see a contradiction, or else I remember wrong. I was comparing the shorted-out OWL section under discussion to a grounded vertical - capacitive at less that 1/4λ, so you use a loading coil to bring the shortened length to resonance.  When I operated in the apartment in Cambridge, I used an end-fed zepp running across the street with a slightly less than 1/4λ feeder, and used a variable inductor in series with the feeders to bring the shortened stub to resonance. But I see that the diagram suggests that a less than 1/4λ shorted stub would be inductive, not capacitive.

We have to be overlooking something simple and obvious here. Confusion over whether the low-Z point is at the generator or at the load end of the stub?

Chuck,

QRP measurements are fine for tune in and adjusting. But If you are trying to troubleshoot a problem, sometimes it wont rear it's ugly head at flea power levels. If something is zorched, or arcing over, you want to make as many of your tests as you can at Full power. If something is intermittant, a lot can change as you QRO

Don:
Do you agree that the 13.5 quarter wavelength long 15 meter antenna systyem is (1) inductively reactive and (2) at 600 ohms or 135 degrees past the high impedance point which is 45 degrees (1/8th wavelength) past the peak current point or low impedance point?   My crude quarter wave chart is showing current.  
Thank you for your help.
Chuck

Slab Bacon:
I agree.  Testing at higher power sometimes provides a different picture.  Very good tip.
Thanks.
Chuck

It still seems like it should be capacitive, but the chart you scanned says inductive. I have several editions of ARRL antenna book, dating from pre-WW2 to just a few years ago, as well as other texts, such as the Laport Antenna book and Radio antenna book. I need to pull out some of those where I can study the chart and read the entire text, and see what I am missing or where the confusion lies.

It would seem to me that if you took an 1/8-wave stub and shorted out one end or placed a low value resistor across that end, that the opposite end would show a capacitive reactance. That's what your 13 quarter waves (or any number of odd quarter-wavelengths for that matter) is the equivalent of: a short across the OWL (if the entire length of wire is a parallel conductor), or a low-value resistor (if the first quarter-wave is pulled apart and stretched out horizontally to form a half-wave dipole, 1/4λ per leg).

600Ω sounds reasonable, but I haven't tried to calculate the actual value of reactance or resistance at that point, or look it up in the charts. Exact figures would depend on the impedance at the feed point of the antenna (which would vary somewhat depending on height above ground, inverted vee versus true horizontal, the presence of nearby objects and antennas, etc).

Regarding the reactance (inductive or capacitive?), either I missing something fundamental, or we are each talking about two different things and don't realise it.  Let me review my ARRL book with those some charts you scanned and maybe some of the other references and see if that helps clear up where the contradiction is.

Interesting discussion in any case.  That's the way we learn things. And sometimes the "books" lack clarity, are inadvertently misleading, and occasionally, just plain wrong.

Don:
The universal reactace chart,I scanned with the two formulas, gave me 600 ohms reactance for the 1/8th wavelength or 45 degree point on the curve.
This is interesting because the feed line impedance, i.e. 600 ohms, would always effectively be the same at the 45 degree point along that crude current curve drawing, i.e. 600 ohms reactance at 45 degrees = 600 ohm impedance feed line.  The same goes for 300 ohm, 450 ohm etc.

I am going to use a variable capacitor or tapped inductor instead of a stub line or extra feed line to counter the reactance found at this 45 degree point.

I too, will dig into some other text material.

Thank you.
Chuck

But what is the resistive component at the 45˚ point? I don't think 600 ohms of reactance at the tuner would mean anything in regards to the line impedance Zo, since the line is running with standing waves, and the 600Ω reactive point exists only at the end of the transmission line where you are feeding it, and at 1/2λ intervals up towards the antenna, and the tuner is cancelling out that reactance anyway. On second thought, the resistive component would be irrelevant for the same reason; is not constant, but varying along the line in sinusoidal fashion, too.

This looks like a situation where conjugate match theory would be helpful. But I would probably just use trial-and-error and a field strength meter to see what works best.  That would most likely be quicker. Nevertheless, it's still nice to know why something works, rather than flying by the seat of your trousers the way Deforest did when he invented the triode tube.

Don:
I agree.
It would like be nice to know how to predict the component values at these 45 degree, 1/8th wavelength points, for open wire lines.
This brings me back to my first question.  I wonder where can I find an antenna analyzer designed and built specifically for open wire lines?
Chuck

Don, if I may horn in, I'd like to comment on the reactances obtained with xmsn-line stubs.

A stub shorted at one end with lengths less than 1/4 wl will be inductive at the opposite end. As length increases from 1/4 wl to 1/2 wl the reactance will be capacitive.

A stub open on  one end with lengths less than 1/4 wl will be capacitive at the opposite end. As length increases from 1/4 wl  to 1/2 wl the reactance will be inductive.

Hope this helps.

PS--I should also have said that at length 1/8 wl, or 45°, the reactance X = Zo of the line.

Walt

Walt:
Thank you for the conformation the 45 degree point is reactive with same Zo of the line, i.e. 600 ohms.

If you have time would take a look at my reply #30 of this post.  The last attachment is a crude current graph of my 15 meter total antenna system length.

It shows at the far right that this system is inductive reactive at the 45 degree point.  

Is this correct (inductive reactive) or do I need to gather further information?

Chuck

Walt,

Thanks for your input.  Maybe you can clear up something for me.

I based my supposition regarding the capacitive reactance of the < 1/4λ stub shorted at one end, on the behaviour of the familiar vertical antenna working against ground.  A 1/4λ vertical will be purely resistive @  approximately 36 ohms.  A vertical shorter than 1/4λ will have capacitive reactance, and you cancel out the C reactance by adding a loading coil in series at the base of the antenna. The inductive reactance of the coil cancels the capacitive reactance of the antenna, and the antenna is tuned to  resonance. Of  course, we know the resistive component is now less than 36 ohms, decreasing rapidly as the wire length decreases, but nevertheless the reactance is  always capacitive and we use a series inductance to tune it.

So, why does a quarter wave or less of parallel line, shorted at one end, behave differently (I'm talking about inductive vs capacitive reactance) from a single wire shorted to earth? Again, we can agree that the single vertical wire radiates while the parallel line, which has a corresponding wire of opposite polarity in close vicinity, doesn't.

One difference in my analogy compared to the vertical antenna is that the loading coil is inserted at the base (the shorted end) of the antenna, while with the stub, the reactance is measured (and tuned out with the ATU) at the opposite (unshorted end).  Could that have something to do with the apparent (to me) paradox? However, the loading coil can be attached to the top of the vertical (the unshorted end), if there is a capacitive top hat or something similar for the loading coil to work into, serving as a terminal to complete the circuit back to earth.

So would a 1/8λ vertical wire shorted to earth, which could be tuned to resonance by inserting a series coil at the base, display inductive reactance at the top (if the measurement is taken with no  loading coil at the base to resonate the wire)?

Don

Good question, Don, but right now it's too late to consider further. I'll definitely get back with you tomorrow.

Walt

I've seen this work out in the field for antennas.

At one time (1987) I had seven 75M  dipoles at 120' high, EACH fed with open wire line. They were all broadside facing NE/SW with 130' of open wire (1/2 wavelength) coming to the ground for each  dipole.  They were positioned just like a Yagi, spaced about 50' apart.

To tune them I had the driven element connected to a receiver and a beacon out in the woods.  I made all parasitic dipoles a little short, like about 120' each. By moving a shorting wire along the bottom of each feedline I was able to reflect back a small amount of inductance to the top junction of these parasitic dipoles and tune the array for maximum f-b or forward gain. It worked really FB using relays to move the shorting stubs for NE or SW coverage.

Earlier I had the same array up at 60', and used 1/4 wave long open wire feeders. By using 250pf variable caps on each end of the openwire, I was able reflect back inductance and tune the elements in the same way.


The drawback was, believe it or not, all that openwire loss added up and I could never get the same gain as using just simple pretuned Yagi elements.  The Yagi elements were as low as 20 ohms at the center once tuned, so the mismatch loss added up over seven elements.  I thought it would be much less, but alas, it was a factor.

T

Addressing reactance in lines and antennas.

Let’s first consider dipoles and verticals over ground. At this point we’ll consider all wires and conductors to be lossless—zero resistance. We’ll begin with a center-fed half-wavelength dipole, each half of which is 90 degrees in length. Current leaving the center encounters 90 degrees in travelling to the far end, sees an open circuit, developing a 180-degree phase change, then encounters an additional 90 degrees on return to the center. Thus, during one round trip from center and return, the current has traveled 360 degrees, returning precisely in phase with the next cycle of current (and voltage) arriving at the center from the source. Consequently, no reactance has been developed, and the dipole is in resonance.

However, because power is lost during return to the center, due to energy radiated by the dipole, the magnitude of the returning (reflected) current is somewhat less than at the beginning. Consequently, the difference in magnitudes of the source and reflected current determine the terminal resistance of the resonant dipole.

However, consider what would happen if there were zero radiation. This condition will occur if we fold the dipole at its center, thus forming a quarter-wavelength open-circuit stub. Radiation is canceled because the radiation from each leg of the stub is equal and opposite.  Consequently, with lossless conductors, the open-circuited stub presents zero resistance at the opposite end, a virtual short circuit. (With real conductors the resistance will be very small.)

We’ll now consider a quarter-wavelength dipole, one-eighth wavelength on each side. One eighth wavelength is 45 degrees.  In this case the current encounters 90 degrees of travel during the round trip, plus the 180-degree phase shift at the open circuit at the far end of the dipole, for a total of 270 degrees of phase change. Thus the returning current arrives at the center 90 degrees earlier than the current (and voltage) arriving from the source. We know that when current leads voltage the result is a capacitive reactance, the condition prevailing in this case.

If we now fold the quarter-wavelength dipole at the center we again get an  open-circuit stub, this time one of one-eighth wavelength, or 45 degrees. Because the electrical lengths have not changed while folding this dipole into a stub, the open-circuit stub less than a quarter-wave in length, the reactance of the stub is capacitive.

 The following are two equations that determine the reactance of a transmission-line stub shorter than a quarter-wavelength :

XL = jZo tan length in degrees.

XC = -jZo cot length in degrees.

For  line lengths between 1/4wl and  1/2wl the signs of the reactances reverse.

I hope this essay helps in understanding the relation concerning reactance appearing on lines and antennas.

Walt

I figured it out!

On several occasions in years past, I have used half wave end-fed zepps with a quarter wave of tuned feeder, and open-wire fed dipoles with a half wave of open wire tuned feeder, as a single band antenna for 75/80m. The transmitter had a link-coupled final.  I was able to use the antenna all the way across the band by cutting the line slightly short for the top end of the band, and placing a rotary inductor in series with one of the feeders. As I moved down the band, I would roll in additional inductance and the antenna would  load up just as it would with a full-fledged link coupled tuner. As I QSYed down, I added in more inductance to tune out the additional capacitance as the frequency was lowered. The same phenomenon exists with the ≤ quarter-wave vertical as discussed in the provious post.

The only way for this to work is, if a shorted stub, less than 1/4λ shows inductive reactance, then an open stub the same length must show capacitive reactance, just as Walt pointed out.  And that's exactly what the charts say.  

The end-fed zepp feed line is like an open stub, since one feeder floats free and the other is attached to a high voltage point. The dipole with the half-wave feeder can be thought of as a quarter wave feeder feeding another quarter wave feeder, which feeds the low-Z dipole.  The high voltage point is right at the middle of the feed line, so that the bottom half of the feed line could be thought of as a 1/4λ open stub, feeding the voltage point.

The charts show that the reactance at the input terminals of on open-circuited line, as a function of line length in wavelengths, is the opposite sign (inductive vs capacitive) from that of a short-circuited line.  So it depends on which end of the line you insert the capacitance or inductance.  In the case of the ≤ quarter-wave vertical and the zepp feeders, the inductance is placed near the shorted end, and at that point, you are looking at an open-circuited line.  But in the case of the antenna/feed line combination with the 13.5 quarter wavelengths, the first 13 quarter-waves could be  replaced by a short-circuit or by a low value resistor, and the  remaining 1/8λ, from the tuner's viewpoint, is a short-circuited line, therefore it shows inductive reactance.

What I have been overlooking is the fact that the same line, with a short circuit or a resistor attached to one end, is an open-circuited line from the resistor's (or shorting bar's) viewpoint, but looking back from the opposite end of that same line, one sees a short-circuited line. I was simply thinking short line = loading coil = capacitive reactance, and long line = series capacitor = inductive reactance.  Of course, this is true only from the perspective of the appropriate end of the line.

The grounded 1/8λ vertical indeed does show capacitive reactance at its base. But if you went up in a hot-air balloon or helicopter and measured the reactance by touching the probe to the tip-top, that same antenna would show inductive reactance.

Senior moment in action.

On my post #46 (yesterday) I forgot to include the minus sign with the jCOT term in the equation for XC. I am correcting that error in #46.

Walt

I am double checking to make sure my graph in reply #30 is correct.  Is it correct?  Don, Walt and everyone your replies were very helpful.
Thank you.

Chuck

This might not be the right topic thread but I think in this one or maybe elsewhere, don posted something to the effect that if you lift your johnson matchbox so it is not bonded to a ground via the lug on the back panel, you achieve a balance of the currents in the open wire line.  I think I misread his suggestion--maybe he meant not grounding the inductor center with a center tap to ground.  In that case the MB is okay as is, for it has no center tap.  Well, I conducted a crude experiment here today.  with the MB cabinet strapped to a rod outside via a 3 or 4 foot run of 3 inch wide copper strap, I tuned it to 52j0 on 3885 and fed it 500 mw.  Then I went outside with a simple diode driven fs meter, one of those CB type things made by the millions with a telescoping antenna on it.  I measured the field about the feedline and noticed it was slightly stronger on one side.  I got away from the feedline and went to each end of the dipole standing on the ground under it and took readings.  The fields at the ends of the dipole were uneven; the one corresponding to the slightly stronger side of the feedline had a slightly stronger reading than the other.  on a scale of zero to 2 it was 1.5; the other end was 1.3

I repeated everything with the MB disconnected from the ground strap.  the feedline field was closer to uniform (not entirely but closer) and the ends of the dipole both measured 1.3 (I would have been happier with 1.5 hi hi). 

I have not yet given the floating match box full power--I gave it a modulated 50 watt test transmission to see if I could hear any signs of trouble like RF in the audio but I didn't notice anything different from the former state of things.  I'm still not sure if I want to try this as I have some sort of innate hesitation to run 300 watts with the grounded sides of the MB capacitors only grounded via the unbalanced feedline shield.  But, it does appear to result in a more balanced dipole and feedline. 

I'd be incomplete in describing the test if I did not also mention that one end of the dipole is held up by a tree; the other end is held up by a 50 foot aluminum tube that also serves as a 75 m. vertical (it has a 15 foot stinger on its top) with a ground system at its base.  It also holds up the 160 m. inverted L.  The ground system radials cover the back yard and extend up to the shack (a.k.a. house) where the MB ground rod is driven in at the open wire line entrance. 

For lack of any explanation I suspect the ground rod is somehow forming part of a counterpoise with the radials and mast and when the MB is connected to it, it works against it, functioning more like a slightly unbalanced off center fed dipole. 

Sorry if this is way off the topic, but now  that I think about it, putting a very weak signal into the antenna system and measuring the fields is a form of open wire analysis.  I wish I had a more sophisticated measuring instrument though.

I'll be surprised if anyone notices any signal strength difference with my matchbox ungrounded but I'll give it a try at night with 50 watts for starters.

Say, if I'm doing anything that might cause the MB to flash over with higher power or if for some other reason this is a dumb thing to do pipe up and warn me--field strenghts don't matter if something zorches.

rob

Rob:
I have read (RSGB Handbook) that leaving the output side of the link antenna coupler is the way to go.  Grounding this side of the coupler (center of tank coil or rotor of split stator capacitor) increases the potential of the Marconi effect.    http://amfone.net/Amforum/index.php?topic=12469.5;wap2

Chuck

Looks like it is.



Too right!  Rob, according to my schematic of the 275 watt matchbox (not sure what size yours is, or if all sizes use exactly the same circuit), the coil is indeed floating, but the rotor of main split-stator capacitor C1 is grounded, and the two inner-most sections of differential capacitor C2 are tied together and grounded. That is effectively the same thing as grounding the mid-point of the coil. Try lifting the connection where the rotor of C1 connects to the sections at C2, from ground (which I assume to be the metal case of the Matchbox). I would also assume the ground terminal simply connects to the case. Leave the connection between C1 and C2.

The connection to C1/C2 and ground is necessary when the Matchbox is being used to feed a single-wire antenna, since that is the only way the circuit has a ground return. But when feeding a balanced two-wire line, the ground is unnecessary, and in fact, may cause unbalance at the feeders precisely for the same reason grounding the mid-point of the coil would: it is providing a ground return for common mode current and allowing the feedline/dipole combination to work like a vertical-Tee against ground in addition to working like a balanced parallel line fed dipole.  

Since the common-mode current is of the same polarity in both feeders, but the differential current is of opposite polarity in each feeder, the sum of the currents will be different in each, since in one feeder I (total) = I (common mode) + I (differential), and in the other feeder, I (total)=  I (common mode) − I (differential). In additional to the unbalance, the common mode current (probably much less than the normal differential current) nevertheless causes some feed line radiation, plus ground losses if the Matchbox is earthed to anything less than a good low-loss radial ground system.

Lifting the ground point at C1/C2 won't blow anything up, since this would not substantially change the voltages across the capacitor sections or at the band-switch connections. If anything, it might actually reduce them, since unwanted common-mode voltages/currents are not being added to to the desirable differential currents at one feeder. Lifting this ground will not affect the ground return at L1, the link. If you were using a link-coupled final and a balanced twisted-pair or parallel-line link instead of coax feed, there would be no need for this ground connection, either.

I would modify the Matchbox to have an external ground strap to C1/C2, that could be removed or added as needed.  It shouldn't be needed with a balanced open-wire transmission line. It is incomprehensible to me why Johnson didn't do it that way to begin with. I would keep the external ground connection to the metal case of the Matchbox, to keep the unit effectively shielded, and to prevent the box from becoming hot with RF, in case something goes wrong to unbalance the system, such as a conductive object contacting one of the feeders, or one leg of the dipole coming down.

Thanks Don.
Chuck

Don, thanks,yes of course I see now.  I looked at the MB schematic again and it is obvious the v. across the caps should be less with them lifted off ground.  it all makes sense.  I have two KW MB so I'll do the cap mod on one and put it into service and perform my field strength measuring again and see what happens.

I started a new post titled "Link Antenna Coupler Circuit Setup".  I have some more questions regarding link antenna coupler circuit selection.  Everyone's help is greatly appreciated!
Chuck