ad: SteppIR-1

AIM 4170C to horizontal loop fed by ladder line?

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by 2E0ILY, May 16, 2012.

Thread Status:
Not open for further replies.
  1. 2E0ILY

    2E0ILY Guest

    On the recommendation of this forum i have bought myself an AIM 4170C antenna analyser. looks a nice bit of kit ;) I am new to amateur radio and have an immediate question. At the moment i have a 25 feet high horizontal quad loop of about 490 feet in length. It feeds off a corner into my bungalow through an upper window, via 450 ohm (I think...) one inch wide ladder line. This goes straight to the two screw terminals on the back of my Yaesu FC-902 ATU, and the ATU then goes to my FT-902DM transceiver. The analyser has a SO-239 socket on it for connection to the DTU. How should I be connecting it to my antenna system? I think I may need something like a 4 ; 1 balun at the house end of the ladder line, then connect the analyser via a short piece of co-ax to the balun, which in turn is connected to either "side" of the loop antenna ladder line? Or have I got this wrong? Is there an easy way to connect the ladder line so it terminates in a PL-259 plug, or is this an incorrect way of measuring things? I wiah to use the loop on HF multi band and in particular see if it can be optimised for either 160 metres or for PSK31 digi modes on either 20 metres or 17 metres. At the moment the AIM 4170C is probably 100 times more complicated than I am comfortable with, but I thought buying something really basic may be false economy in the longer term, as I feel I could get quite into this hobby ;) Thanks.
     
  2. 4L4AGU

    4L4AGU Ham Member

    WOW :)

    I also bought that analyzer recently and I also want to measure ladder line loss using it! But I went a bit far ahead - I've took a piece of ladder line, about 6 meters long, connected it to analyzer via short (70cm) RG58 cable and run several sweeps, with ladder line far end open, closed, terminated with 50 or 450 ohm resistor. I got a lot of nicely drawn curves, which appear to deliver nothing, since AIM refers to 50ohm as base resistance, so it need to changed to 450 ohm somehow.

    Regarding your question, since you're measuring complete antenna system, measure it in way it is connected to transmitter, since you want to determine transmitter-antenna matching conditions, aren't you?
     
  3. K8JD

    K8JD Ham Member

    I don't think your antenna setup is 50 ohms on ANY frequency.
    I have put up several horizontal loops but used a quarterwave of 75 ohm line to transform Z down to 50 ohms. (this ONLY works on one frequency) where the loop is one wavelength in circumference on a single band ! (they worked fine without any tuner).
    You could hook up your analyzer to the radio end of the tuner but that only tells you the tuner is adjusted and nothing about the actual antenna radiator !
    I have resisted using balanced line feeders until I built an experimental DZ/XDZ antenna for some bidirectional gain on 30/20M last fall
    .I Had to buy the new MFJ balanced line tuner for it. The directional wattmeter in this tuner measures the SWR on the 50 ohm unbalanced side so you never actually know or worry about the SWR on the balanced feeders to the antenna radiator !
     
  4. WB2UAQ

    WB2UAQ Ham Member

    I also have the need to measure the impedance of balanced line loads. I made a coil of RG178 on a coil form (neatly wound). With a variable cap connected from the braid at one end of the coil to the braid at the other end, I tune this parallel circuit to resonance at the freq where I am measuring the impedance. This creates a high choking impedance balun where at resonance the Z can be about 75K because the RG178 on the form has a Q of over 200. Knowing the electrical length of the coax on the coil, I can measure the Z at the input end of a balanced line antenna system with an unbalance impedance measuring instrument. In my case the electrical length of the coax is about 4.7 meters so I can rotate around the Smith Chart with this length converted to wavelengths for the freq of measurement. I never had a reason to measure the loss of open wire or other balanced lines. If the line is in good shape, the specs are well known and in the handbooks. 73, Pete
     
  5. AD4J

    AD4J Ham Member

    I do a lot of measurements of balanced line fed antenna with my AIM 4170C. The stock recommendation from Array Solutions and others is to use a 1:1 balun at the analyzer and do a custom calibration using the standard loads (short, open, and a known pure resistance - often 50 ohms). The custom calibration is supposed to factor out any impedance shifts caused by the balun. My tests with a couple of well-regarded baluns (DXEngineering designed by W8JI and Balun Designs) have shown that these baluns often transform the impedance to very low values on the input when the impedance on the output is high. This makes sense because they are just low impedance (~50 ohm) coax wrapped around cores. The length of coax in the balun transforms impedance when the load isn't 50 ohms resistive.

    However, I've found that I can get good readings of balanced line antenna systems without the balun, if I am careful not to ground any of the measuring equipment and use short leads. Specifically, I use a laptop on batteries and power the AIM with a battery pack or 9V battery. I use two adapters on the back of the AIM 4170C - a PL-259 to BNC adapter and a BNC to binding posts. The balanced line attaches to the binding posts. I did a custom calibration of the adapter combination. This setup has worked very well for me in determining the best feed line lengths to use for minimum SWR (50 ohms) on various bands.

    That said, I still use a 1:1 current balun between my balanced line and my antenna tuner after I get my system tuned. I then use the AIM 4170C with the balun to plot the impedances that will be seen by the tuner (or directly by the transceiver if the SWR(50) is low).

    I would recommend that you add a 1:1 current balun between your tuner and the balanced line. This should help minimize the common mode currents that can cause RFI and high noise levels on receive. Don't be fooled by the idea that 450 ohm ladder line needs a 9:1 balun to give you 50 ohms. The impedance at the shack end of the ladder line is almost never 450 ohms or even near it.

    I think you will find the AIM 4170C is a good investment and will serve you well for many years. It just takes a while to learn how to use it to best advantage. There is a forum devoted to the AIM and similar analyzers. I believe Array Solutions has a link to it. If you can't find it, email me at my call sign at arrl.net and I'll send you the link.

    Jim
     
  6. WB2UAQ

    WB2UAQ Ham Member

    Jim
    If the 1:1 coax wound current balun you are using is transforming hi Z's to low Z's its because the electrical length of coax is a quarter wave or odd multiples of quarter waves long. It is not just because the balun is wound with 50 ohm coax. My single frequency parallel resonant balun uses 50 ohm coax and is close to a quarter wave long on 20 meters (elect length is 4.7 M). The Hi Z's I encounter are transformed down so low on 20 meters that my old impedance instruments can not read the Z. If I want to run tests at Hi Z's on 20 meters I will have to reduce the electrical length of my coax coil or go the other way and add some more cable on the impedance instrument side of the coil. I have no easy way to calibrate (O-S-L) my set up so I use 50 ohm cable all the way to the measurement point ( I just have to know the electrical length and rotate around the chart). Even if you can calibrate out the transmission line I bet it is still tough to get an accurate measurement at at Hi Zs thru a typical 1:1 balun when the choking impedance is only 1000 to 2000 ohms plus the network analyzer is designed around 50 ohm directional couplers (internal to the instrument). 73 and have fun tinkering and learning. Pete
     
  7. AD4J

    AD4J Ham Member

    I agree that the most dramatic impedance transformations of mostly resistive impedances are with a quarter wavelength of coax (e.g. 2000 ohms is SWR(50) = 40:1 yielding 50/40 = 1.25 ohms at the other end of the coax). However, anytime the electrical length not an integral multiple of half wavelength, there will be some transformation. When the SWR on the line is high (as it often is for the coax in the balun with a multi-band doublet), the transformations are amplified. I'll try to remember to post some example graphs a little later from the AIM 4170C, with and without the balun, to illustrate this.

    The 4170C does have the ability to calculate the impedance at the load when measuring at the transmitter end of a feed line, if you calibrate the feed line with the analyzer first. I think it will let you do it by just entering the length and velocity factor, too, but I can't remember for sure.

    The Smith Chart works less well for ladder line than coax, in my experience. I think it's because the impedance is not uniform along the ladder line. Some of the dielectric includes air (the windowed part) and some is solid.

    I strive to find balanced line lengths that produce low enough SWR(50) < 10:1 on the bands of interest to allow using my K3's internal tuner. Currently, I have a 67' doublet that achieves this on 40, 30, and 20 meters with one feed line length and on 17, 15, and 10 meters by adding 7.5'. I can add or subtract the 7.5' extension in less than 2 minutes. Similarly, my perpendicular 34' doublet covers all 6 bands from 30 - 10 meters with just two feed line lengths.
     
  8. AD4J

    AD4J Ham Member

    I promised to post some AIM 4170C graphs of the balun impedance transformation. All of the graphs are for a 34' rotatable dipole fed with 60' of 300 ohm ladder line. The first one is without a balun:

    DoubletGraphNoBalun.jpg

    The SWR is in red, the resistive part of the impedance (R) is in orange, and the reactive part of the impedance (X) is in green. Notice that R never drops below 50 ohms (minimum of 50.6 ohms at about 24.6) MHz and hits a maximum of just under 1500 ohms around 11.670 MHz.

    When a high quality balun (DXEngineering BAL050-H10-AT, a W8JI design) is added, the plots look like this:

    DoubletGraphDXEngBalun.jpg

    Notice that R remains below 100 ohms on all frequencies above 10.2 MHz and is below 10 ohms across about half the spectrum (minimum is 2.9 ohms around 28 MHz). The plots for the Balun Designs 1115t are similar:

    DoubletGraphBalunDesigns.jpg

    The 1115t does generally provide higher values of R (minimum of 4.4 ohms at 28.4 MHz, over 200 ohms at 14.75 MHz) than the BAL050-H10-AT, but at the expense of more X yielding slightly higher SWR(50) values over most of the spectrum.

    Tuners tend to be lossy matching low R values. I imagine a 4:1 balun would transform the R part of the impedance even lower, making it very difficult for the tuner.

    In theory, the 50 ohm coax in the baluns should not change the SWR(50), other than lowering it if there are significant losses. However, both baluns actually raise the SWR(50) over most of the spectrum. Here are the SWR plots for the 3 configurations on one graph:

    DoubletSwrCompareBalunNoBalun.jpg

    The lower orange plot is with no balun, the red plot is with the DXEngineering balun, and the upper orange plot is with the Balun Designs balun. Notice that all three show SWR minimums at the same frequencies and their values match closely anytime the SWR is less than 5:1.

    These measurements, and similar ones with other, multi-band doublets have led me to choose feed line lengths that provide relatively low SWR(50) on the balun output. The goal is to avoid very low impedances (R < 10 ohms) to the tuner. Generally, there are lengths that will do this on multiple bands. I have been able to cover 6 bands with just a couple of feed line lengths. For example, the 60' ladder line used in the system plotted above with the Balun Designs 1115t provides SWR < 5:1 on 20, 17, and 12 meters. By subtracting 8', I get low SWR on 30, 15, and 10 meters.

    One other thought: there has been convincing evidence from well-respected sources (e.g. W7EL & W8JI) that putting a balun at the input of a tuner subjects it to no less common mode current stress than putting it on the output. I believe that. However, it does seem baluns on the output of a tuner often transform the impedance from a value the tuner can handle easily to a very low impedance that will strain the tuner. I don't advocate moving the balun to the tuner input. That requires floating the tuner above ground creating problems that VK1OD has documented. I do think it is worthwhile to try to find feed line lengths that avoid a low impedance on tuner output / balun input junction.

    Jim
     
  9. G3TXQ

    G3TXQ Ham Member

    Just to be clear: that is true if the tuner is unbalanced - it's not true if the tuner is balanced.

    73,
    Steve G3TXQ
     
  10. AD4J

    AD4J Ham Member

    Agreed. With a balanced tuner, the only reasonable place for the balun is on the input. Thanks, Steve.

    Jim
     
  11. KA5S

    KA5S Subscriber

    Notice that the effect of line length is reflected (heh) in the variation of SWR every 6 MHz or so; 60 feet is a half wave if the overall VF is 73%.

    WRT to putting a Balun at the tuner input or output, CM current is the same, for both balanced and unbalanced tuners when CM current is present already (ie.: for reasons other than tuner imbalance) if the tuner is floating, except for link coupled tuners -- which break the CM anyway. The advantage to putting a Balun on the input is that it sees a 50 Ohm differential impedance when the tuner is adjusted properly. Mismatch can be considerable on the output. A Balun winding terminated in a reactive mismatch may resonate at some frequencies, in which case, reactive currents may be much higher than planned for in normal operation.

    IMO.

    Cortland
    KA5S
     
  12. WB2UAQ

    WB2UAQ Ham Member

    There is a good reason not to place the balun on the input. I borrowed transmatchs from some good friends for the purpose of testing the balance and common mode impedance. These transmatches represented three configurations.

    Palstar BT1500A has the balun on the input with a pair of roller inductors and caps to from a "balanced" L network. With increased frequency, the output became more and more unbalanced. (400 ohm resistive load). The worst case was a 3.2:1 current imbalance (10 dB more current flowing out one leg versus the other). After measuring the common mode impedance looking into the output terminals with output terminals shorted together and the input terminated in 50 ohms, there is about a 30 pF shunt C to the chassis with very high shunt R. Now, if this transmatch had a balun at the output, the common mode impedance would be much higher and resistive and well balanced. The higher the common mode Z is the balance is almost automatic.

    Another transmatch in the mix was the Johnson KW matchbox. The balance was better than 0.5 dB from 10 thru 80. The common mode Z is very low but this is made up for by the much better balance.
    At lunch again so must cut this short. 73 Pete
     
  13. G3TXQ

    G3TXQ Ham Member

    Cortland,

    I disagree!

    If you compare the CM voltage across a balun at the input of a balanced tuner with the CM voltage across an identical balun at the input of an unbalanced tuner for exactly the same load conditions, you'll find they can be very different. The reason is that in the case of a balanced tuner part of the differential-mode voltage appears in series with the CM voltage, and that can reduce (or with some loads increase) the CM voltage appearing across the balun and therefore the CM current.

    Put simply, the "bottom" series arm of a balanced tuner adds impedance to the CM path. Try a SPICE analysis with tuner loads of varying impedance and varying degrees of imbalance, and you'll see that's true.

    If you look at the tuner in Roy Lewallen's analysis you'll see that it is unbalanced; many folk - me included - have made the mistake of thinking that analysis applies equally to balanced tuners: it doesn't!

    73,
    Steve G3TXQ
     
  14. AD4J

    AD4J Ham Member

    To address the questions in original post by 2E0ILY more succinctly :

    Since your loop is balanced and your feed line is balanced, it would be best to put a balun on the output of your tuner. My suggestions are the DXEngineering BAL050-H10-AT or Balun Designs 1115t. Both have wing nut terminals to connect to your feed line and an SO-239 to connect to your tuner with a short coax jumper. I have both baluns and have found them to be very good.

    If you want to see the impedances the tuner needs to match on each band, disconnect the jumper from the tuner output and plug that into your AIM 4170C. If you are mainly interested in 160, 20, and 17 meters, you can set up the AIM 4170C to just sweep those bands or do 3 independent sweeps (one on each band).

    73,
    Jim
     
  15. AD4J

    AD4J Ham Member

    The AIM 4170C can be set for another "base resistance" (feed line characteristic impedance Zo). To do this, click the Setup menu in the AIM program and choose "Enter Zo Ref." Since 450 ohm ladder line is usually closer to 400 ohms, you may want to enter 400 for the Transmission Line Z.

    Jim
     
  16. AD4J

    AD4J Ham Member

    My own experience with the Palstar BT1500A and the ARRL lab measurements of its predecessor (Palstar AT1500BAL) found the balance to be quite good. The ARRL measurements (Sept 2004 QST Product Review) were across 5 bands (160/80/40/20/10 meters) using 9 loads from 6.25 ohms to 1600 ohms. Of these 45 combinations, there were 5 that the Palstar could not match (6.25 ohms on 160 and 80 meters, plus 1600 ohms on 10 meters). Of the remaining 40, all but one showed the Palstar having 0 imbalance.

    The same ARRL tests included the Johnson Matchboxes (high and low power versions) with the same loads on 80/40/20/15/10. While the balance was good, there 14-15 loads the Matchboxes couldn't handle. They also don't handle 160 meters.

    Jim
     
  17. 2E0ILY

    2E0ILY Guest

    Thanks for the great replies and in depth info, some of which, to be honest, goes a bit beyond my level of understanding at the moment ;) My wife's father is ill at the moment, so life has taken a turn away from the bands, for a while, but I hope to have a play with the 4170C over the coming weekend. In another thread I am discussing helping an American friend recuperating from an accident build a linear HF amp, to take back to the US with him when he returns home. Will my 4170C act as a viable L and C meter to check home wound inductors and check capacitors for his build, or are we better with a dedicated meter? Thanks, I have printed out the replies so far to study in my free moments, much appreciated!!
     
  18. G3TXQ

    G3TXQ Ham Member

    The 4170C will work fine for those measurements; not only will it show you the inductance and capacitance, it will also show you the 'Q' of the components.

    If you use terminal posts to make it easy to connect the components, be sure to run a calibration with the posts in place so that they don't have any effect on your readings.

    73,
    Steve G3TXQ
     
  19. WB2UAQ

    WB2UAQ Ham Member

    Jim,
    Thanks for the QST article. I looked it over and I don't agree with the balance measurements. The reason for testing the BT1500A was because of a severe RFI in the shack issue. An MFJ unit didn't have the issue at all so I borrowed the BT1500A and tested it for balance and common mode Z. The imbalance was discovered and found to be 3.2:1 (3.2 times the current flows out of 1 terminal vs the other terminal into a 400 ohm load). Also the best SWR that could be acheived was 1.2:1 across 10 thru 80 meters. If you have a BT1500A maybe you could check the balance, common mode Z and etc.. The shunt C to the chassis is also what I believe limits the Z matching range. That shunt C is part of the load Z the transmatch has to match:) It works out to about -j200 ohms at 28 MHz. 73 Pete
     
  20. AD4J

    AD4J Ham Member

    Pete,

    It sounds like the BT1500A you tested has problems. Unfortunately, I no longer have my BT1500A, so I can't attempt to replicate your results. These days, I run 100 watts and use balanced line lengths that provide SWR(50) < 5:1. This allows me to just rely on the K3's internal automatic tuner with an external current balun.

    Have you talked to Palstar about the problem. They might be willing to work with you to correct it.

    73,
    Jim
     
Thread Status:
Not open for further replies.

Share This Page

ad: PowerFilm-1