Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Ladder line length and RFI - why did I observe this?  (Read 1525 times)
VK7JB
Member

Posts: 21




Ignore
« on: February 01, 2013, 01:28:39 AM »

Hello Group,

I'm looking to understand something I've observed - can anyone help?

I run an 80m loop as a multiband antenna and feed it with 300ohm ladderline. Outside the shack, I transition the balanced line to a 10 foot run of RG213 via a Balun Designs, 1:1, toroidal choke balun. The coax then comes through the wall into my unbalanced Palstar manual tuner.

Recently, after changing the loop orientation,  I shortened the run of ladderline to make it a more convenient length and to avoid too much excess.  After shortening the line to 52 feet, I noticed that for the first time, I had RF in the shack with distorted transmitted SSB audio on 40m only. I increased the ladderline length to 60 feet and the RF in the audio resolved on 40m, but was now present on 20m. After changing the ladderline length to 54 feet, there is no stray RF on either 20 or 40m and the audio is clean.

Apart from the changes in ladderline length, I made no other changes to the antenna system. During all of this, I used an MFJ854 clamp on current meter to look for common mode current on the coax in the shack, but none was detectable on the most sensitive scale, despite the obvious effect of stray RF on my transmitted signal. I could detect no common mode current on my power supply cabling either.

I'd like to understand why this RF ingress should be so sensitively dependent on the ladderline length? Could this still be due to common mode current, even if my clamp on meter registers none?

Any explanations welcome. I'm keen to learn from this.

73
John
VK7JB
Logged
W5DXP
Member

Posts: 3391


WWW

Ignore
« Reply #1 on: February 01, 2013, 06:53:41 AM »

I'd like to understand why this RF ingress should be so sensitively dependent on the ladderline length?

Since the ladder-line carries an SWR that transforms impedances up and down the line, when you change the ladder-line length, you are changing the conditions to which your choke-balun is exposed. If the balun is located at a current maximum point (low impedance) it will function much better than when it is located at a voltage maximum point (high impedance). A choke-balun located at a voltage maximim point when using ladder-line with a relatively high SWR can actually cause common-mode problems because it is a magnetic field device and doesn't work very well on electric fields.

Maximum current points are usually pretty close to resonance. Finding the antenna system resonant frequencies using tuned feeders is a good idea. I know where my current maximum points are and that's where I locate my choke-balun.
Logged

73, Cecil, www.w5dxp.com
The purpose of an antenna tuner is to increase the current through the radiation resistance at the antenna to the maximum available magnitude resulting in a radiated power of I2(RRAD) from the antenna.
VK7JB
Member

Posts: 21




Ignore
« Reply #2 on: February 01, 2013, 06:13:00 PM »

Hello Cecil,

Thank you.  This makes sense to me - changing the ladderline length is shifting the position of the current minimum/voltage maximum in relation to the balun and, I guess, also in relation to the RF sensitive equipment and connections at my operating position.

 So, out of interest, how do I calculate or measure the position of the current maxima and minima along my feedline?  I've seen your DOS-based calculators, but I understand they relate to dipoles. How do I apply them to my loop feedline?

73,
John
VK7JB
Logged
AE5QB
Member

Posts: 249




Ignore
« Reply #3 on: February 01, 2013, 08:18:29 PM »

This may be of help.

http://static.dxengineering.com/global/images/instructions/dxe-wa-ins-rev2d.pdf

This relates to the recommendations for DX Engineerings dipoles, but I think it applies equally well here.  Scroll down the document until you find the feedline section.  They recommend the ladder line be odd multiples of 1/8 electrical wavelength on the lowest band.  I have used these recommendations to feed a horizontal loop and it tunes OK with no evidence of excessive RF in the shack.

Good read for thought experiments even if you decide not to implement their recommendations.

Good luck.

Tom/AE5QB
Logged
W5DXP
Member

Posts: 3391


WWW

Ignore
« Reply #4 on: February 02, 2013, 10:38:39 AM »

I've seen your DOS-based calculators, but I understand they relate to dipoles. How do I apply them to my loop feedline?

John, I used EZNEC results to generate that calculator. I use EZNEC+, v5.0 for all of these types of problems and I highly recommend it. For instance, EZNEC says that a full-wave horizontal 80m loop fed with 65' of 450 ohm ladder-line presents an impedance of 1370-j700 ohms to the balun/tuner. That's too high for my tastes so adding 45' brings the impedance down to a more reasonable 122-j108 ohms, an impedance easily handled by my balun/tuner. Antenna analysis modeling has saved me hundreds of hours (& $) over the past 25 years. There are numerous other antenna analysis programs but I haven't used anything except W7EL's products for the past 25 years.
Logged

73, Cecil, www.w5dxp.com
The purpose of an antenna tuner is to increase the current through the radiation resistance at the antenna to the maximum available magnitude resulting in a radiated power of I2(RRAD) from the antenna.
K0ZN
Member

Posts: 1501




Ignore
« Reply #5 on: February 02, 2013, 05:45:50 PM »


Hi.
W5DXP's comments/input is spot on.  The cold reality is, that we have so many bands that it is usually the case that on one or two bands, if you are feeding
an antenna with balanced parallel line, some lengths will present an "unfriendly" load impedance to the tuner or balun. Typically, a Tuner is more likely to
be able to handle some odd impedance than a balun because a balun is a fixed impedance device, whereas the Tuner is adjustable.

Keep in mind that a balun is NOT really intended nor functions properly when it sees an impedance
that is a long way from its design input and output impedances. A balun is really just a transformer and functions best when the loads are appropriate or close to
to design input and output impedance. Using a balun in a system where it sees high VSWR and "extreme" impedances is really a long way from an
electrically optimum situation and it is hard to tell what really may be going in and with the balun electrically. Basically, we hams "force" a balun into
an application it is not really fit or optimized for and the results of this "abuse" of the balun can result in some unpredictable results because the balun is
not functioning in an optimum way. Reality seems to be that in most cases, a balun pressed into that kind of service functions pretty well, but not in
all cases or frequencies.....and you may end up with RF on the outside of the coax or other situations.

Bottomline: you likely will have to experiment with line lengths to find some kind of multiband compromise and you may run into a situation where you have
                to splice in a length of ladderline on one or possibly two bands to get the system to work "happily".  I have that situation at my QTH where I feed
                a 130 ft. Center Fed Zepp; it took a fair amount of experimenting till I found an optimum line length, and even with that, I had to add some
                external inductance to the tuner (Johnson KW Matchbox) to get a decent load presented on 30 M. Obviously a good balanced Tuner will tolerate
                "odd" impedances much better than a balun, which, again, is a fixed impedance device being operated well out of its area of optimization.

73,  K0ZN
Logged
VK7JB
Member

Posts: 21




Ignore
« Reply #6 on: February 03, 2013, 02:44:51 AM »

Hi Group,

Thanks to everyone for their input, on and off the list.  I've learnt a lot.

I'd like to clarify a few aspects of my initial posting, because I'm not sure I described my setup very well:
Firstly, my problem related to stray RF in the shack at certain ladder line lengths and not to any inability of my ATU to match the antenna system to 50 ohms.  In fact, at all ladder line lengths, including those where RFI was a problem, the internal ATU in my K3 and my manual, Palstar tuner were both able to find easy matches on all bands. Secondly, the balun I mention at the transition from ladder line to my short run of coax is a 1:1 common mode choke (50 ohm coax wound through a toroid stack)  and I'd expect any impedance transformation is likely to be very small.  Finally, I found my solution empirically - through the incremental addition of one foot lengths of ladder line until I found a length at which the EMC issues resolved on all bands.  I can't see any simple relationship between my ultimately successful length  of ladder line and the published "rules of thumb" for the use of ladder line in multiband antenna systems such as mine.

So what I've learnt is this:  such multi-band antenna systems are complex, with multiple interactive elements. To arrive at a "friendly" feedline length, it's important to know the locations of the differential mode current/voltage maxima and minima and  also the locations of the common mode current maxima along the feedline. To complicate matters, the velocity factor for energy transfer along the line in differential and common modes may well differ and then, placement of any balun in the system will influence the common mode current distribution along the feedline too, depending on the specifics of the design and implementation of the balun in the system.

I didn't really appreciate the complexity before this and all I can say is that I'm pleased that I found my "successful" length of feedline by experimentation. Given the multiple variables, I'm not sure how I would have confidently or accurately modeled or predicted that length prior to taking out my wire cutters and cutting the ladder line to its final length.  Maybe that's a reflection of my very limited modelling skills.  But, I'm an average ham and were I to approach this situation again, I would do it in the same way - by experimentation, a foot at a time.  In my case, the difference between obtrusive RFI on 2 bands and perfect audio and an EMC free shack was just 2 feet of ladderline. 

Knowing what I know now, I can't see where I'd start to be able to predict that "magic length" with confidence. 

Thank you all for your input.  It's been an interesting learning curve.

73,
John
VK7JB

Logged
KQ6Q
Member

Posts: 946




Ignore
« Reply #7 on: February 03, 2013, 03:37:38 AM »

Feedline lengths can cause odd problems, and sometimes rather than over-analyze things, just making a minor change will solve the problem. my recent case - feeding a Hustler mast with four resonators - 20, 17, 15, 10 meters. The 15 meter resonator wouldn't stay tuned after initial loading. Other 3 bands just find. I thought I'd damaged the resonator. replaced it. Same thing. addeded 3 feet of coax to the feeder - end of problem!
Rule ? if an antenna tunes up, but won't stay that way, change the feedline length a bit! Conversely, if things were fine until you changed the length of the feedline, that's the problem!
Logged
W3HKK
Member

Posts: 581




Ignore
« Reply #8 on: February 03, 2013, 06:01:19 AM »

Here's a question for the group regarding "cutting ladder line and owl":

Does it have to be cut to be tested? (and Can we avoid a dozen or so short pieces going into the trash can?)

Couldnt the line be folded back on itself ( and then taped in place) to get an approximate solution, before the dikes are brought to task?

I do this all the time with my field day/contest loops: bending and twisting the wire just enough to find resonance - but not cutting it, since at the next event the height above ground, or even  using it on a different band, will require changes.  So I just untwist and lengthen/shorten the loop.
Logged
W5DXP
Member

Posts: 3391


WWW

Ignore
« Reply #9 on: February 03, 2013, 06:55:57 AM »

Knowing what I know now, I can't see where I'd start to be able to predict that "magic length" with confidence.

One of the things not mentioned is the physical layout of the system which can be best case or worst case for common-mode problems. Common-mode signals are generated in two ways, conductive and inductive. The conductive problem deals mainly with impedances and current paths. The inductive problem deals with the electric and magnetic fields/waves and is primarily dependent on the physical layout. That's why we need to follow good engineering practices as far as system symmetry is concerned, e.g. coming off the feedpoint at right angles to the plane of the center-fed antenna.

Couldnt the line be folded back on itself ( and then taped in place) to get an approximate solution, before the dikes are brought to task?

I found such actions to result in relatively unpredictable conditions, e.g. creation of unwanted stubs. What I did to solve the problem is have 32', 16', 8', 4', 2', and 1' sections of ladder-line available with banana plugs/sockets on the ends. With those six sections (kept in my storage shed) I can easily come up with any additional length of ladder-line between zero length and 63 feet. When I get it right, I use automotive crimp butt connectors to connect the additional section to the rest of the feedline.
« Last Edit: February 03, 2013, 07:09:03 AM by W5DXP » Logged

73, Cecil, www.w5dxp.com
The purpose of an antenna tuner is to increase the current through the radiation resistance at the antenna to the maximum available magnitude resulting in a radiated power of I2(RRAD) from the antenna.
VK7JB
Member

Posts: 21




Ignore
« Reply #10 on: February 03, 2013, 04:09:56 PM »

Hi Cecil,

That's how I solved my problem:  I made 1', 2', 4', 8' sections of ladder line and used them in incremental combinations until the solution was found. 

I wish I'd thought of using automotive crimp connectors though - would have made the job far quicker and simpler.  Excellent idea, thanks.

73,
John
VK7JB
Logged
WB2EOD
Member

Posts: 205




Ignore
« Reply #11 on: February 03, 2013, 08:17:34 PM »

When running balanced feed line, there are certain feed line lengths which depending upon operating frequency must be avoided. I am not sure of the particulars but it should be easy to find the exact numbers.  Failure to do this will result in a variety of malfunctions, including but not limited to the ones you describe.  A friend of mine had similar problems where the antenna tuner could not attain a match.  Turned out that the feed line length was right in that "forbidden zone".  After adjusting the feed line length, the problem went away

Hope this helps
73
WB2EOD
Logged
W5DXP
Member

Posts: 3391


WWW

Ignore
« Reply #12 on: February 04, 2013, 04:26:01 AM »

Turned out that the feed line length was right in that "forbidden zone".

For ordinary center-fed wire dipoles, here is some information on good and bad lengths of parallel feedline.

http://www.w5dxp.com/goodbad.htm

Here's a graph, based on EZNEC results, that may help.

http://www.w5dxp.com/majic.gif

Logged

73, Cecil, www.w5dxp.com
The purpose of an antenna tuner is to increase the current through the radiation resistance at the antenna to the maximum available magnitude resulting in a radiated power of I2(RRAD) from the antenna.
K0ZN
Member

Posts: 1501




Ignore
« Reply #13 on: February 04, 2013, 06:20:24 PM »

To W3HKK:

No, you never want to fold a parallel conductor line back on its self. You never want to leave a "rats nest" of "extra" ladderline. You never want to coil ladderline in a tight coil.

Here is why:  Ladder line, and all other parallel conductor transmission lines operate in a much different way than coaxial cable. With coax, the energy
is INSIDE the line, essentially between the two conductors. With a parallel line, the energy is AROUND the line. BIG difference. In an over simplification, parallel
transmission lines are essentially two antennas with the currents 180 degrees out of phase so that the energy cancels out preventing radiation. If you start
doing "weird things" (coils, fold backs, big pile on the floor, etc.) there is highly unpredictable coupling and interaction occurring. The higher the frequency
(e.g. the shorter the wavelength) the greater the interaction/cancellation/coupling will be. It is very hard to know what is happening but it can significantly
impact the impedance, become an impedance bump thus increasing SWR, or throw the line out of balance, thus reducing cancellation and allowing radiation
from the line or making it difficult for a tuner to find a match. This is also why you want to keep parallel conductor transmission lines a reasonable distance from conductors as the energy AROUND the line can couple into the external conductors. Also, very sharp bends are best avoided if at all possible.

Lastly, with parallel lines you can have have VERY high voltages on the line at various point and the location of those points will change with the band. At legal limit and high SWR, such as feeding a double/center fed Zepp or large loop with ladderline on multiple bands, at legal limit powers there are DANGEROUS voltages present on the line. The hazard is severe RF burns which are very nasty since RF energy only flows only on the very surface of a conductor (skin!). Even at 100 watts I would not want to be grabbing ladderline with a high VSWR on it when in transmit.

Bottomline:  Ladderline (any parallel conductor line) should be cut to the needed length and/or any "extra" must be run in a neat and reasonably separated manner.
It is great stuff and very low loss, but it must be handled with some care and planning.

Translation:  a "ham-haywire" mess of the line can invite problems.

73,  K0ZN
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!