Tuner issue on 80m

W9OY

http://www.hamradio.me/antennas/measuring-quarter-wave-coax-stubs-using-mfj-analyzer.html

This article tells you how to find the length of a 1/4 wave stub or a 1/2 wave stub on a given freq for a given feed line.  To get a 3/8 wave stub you merely multiply the 1/4 length by 1.5.  A 300 or 600 ohm stub behaves exactly like a 50 ohm or 75 ohm stub in terms of transitioning from open to short.  If you have a antenna analyzer you can use it to teach yourself about stubs.  You can also measure the velocity factor of your coax.  

It's kind of pointless to go to a LOT of trouble with this since the antenna impedance is variable wrt freq.  You are relying on the tuner to provide the missing match anyway.  What the 24ft of coax does is simply bring the impedance to be matched into a range the tuner can tune.  Put another way an antenna with a given feedline on a given freq might have a 20:1 SWR at the transmitter end of the coax, well outside the range of a 10:1 tuner.  What the extra coax does is bring the mismatch below 10:1 

Here is an article that uses stubs to match a openwire line multi-band antenna instead of a tuner  

http://www.kk5jy.net/dipole/

W9OY

The balun is not the issue.  He is feeding a 80M dipole with a 40M dipole from the same feedpoint on many bands including 160M.  A balun will only introduce extra loss

73  W9OY

Alan - KA4B

Lee, as I mentioned to Luis, I have a good balun (see above). This a 80/40 multi band dipole with the wires well separated.  I would never attempt to transmit on 160 because 80 is far too short for that.  I only mentioned 160 because even though the SWR is extremely high, the 6500 tuner can easily handle it. 

W9OY

Balun or no balun this is a very complex impedance on the 80M antenna.  I made a model of this antenna.  80M element 124 ft resonant at 3.8 mhz.  40M element 68ft resonant at 7.2 fed at the center in free space at 60ft measured with NEC.  elements were at 90 deg.  I took care to balance the feed

The impedance on 80 at resonance was 1.8:1 at the antenna and the SWR at 3.5 mhz was over 10:1 at the antenna!  So you may think it's 3:1 but it's actually far worse at the antenna end of the coax.   When I disconnected the 40M element the SWR at 3.5 mhz was about 7:1 at the antenna so the interaction between 80 and 40 plays a part.

73

Alan - KA4B

Lee, I measured about 5:1 at 3.5, but my problem was at 3.6.  What do you show there?  Do you have any specific recommendations for an antenna analyser?  Would I need to put it at the feed point of the antenna to get an accurate measurement?

W9OY

An electrical halfwave (taking into account vel factor) piece of coax on the freq of interest repeats whatever impedance to which it is connected, but I wouldn't fool with it except to learn something.

73

W9OY

Also you probably need a pretty good antenna analyzer 

Luis del MOlino

If you have already a balun up on the dipole, then you very likely have a resonant lenght ub 80 meters on your coaxial line and the RF is becoming externaly with the coaxial line acting as a resonant receiving antenna. Then you should destroy this resonancy with a current balun down the line just in your shack placed on the line in any place.

Try MFJ-915 as a current balun with plenty of ferrite rings in the inside..

73 Luis EA3OG

Alan - KA4B

Luis, considering velocity factor my coax length is 0.83 wave lengths.  I would expect that to be a safe length.  It is also perpendicular to the antenna for 35 feet and the rest goes off at about 45 degrees, so I wouldn't expect it to pick up the signal. This happens even when I am tuning with only 2 watts.

 I am looking at getting a high quality Antenna Analyzer.  BTW I shortened the antenna to resonate on 3875 (up from 3800) today and the bad spot moved up the band a corresponding amount. 

Luis del MOlino

No dout there is a bad esonance in some place. You do not mention what kind of balun. T he bad resonance could be in other cables. Anyhow put a current balun as MFJ-915 onyour line snd disconnect any common grounds inside the shack. RF groumds must be outdoors. 73 Luis EA3OG

Alan - KA4B

Luis, I have a Balun Designs 1116d which is optimized for low frequencies.  (I did comment on that earlier, but I am no longer seeing that post.)  I think that is a fairly high quality balun from what I know.  I am looking at buying a Rig Expert antenna analyser, but I haven't settled on which one yet.

Bill W2PKY

I just bought the AA-54 because it can display multiple frequencies.

Alan - KA4B

Bill, with the potential for some kind of coax problem I was debating getting the AA-230Zoom so I could more easily find the location of the problem.  Are you using yours with the AntScope software?

Ken - NM9P

Bill, I picked up an AA-54 used for $200 and am very pleased with it.  

Yes, It is even more impressive when you connect it to a computer via the USB and use AntScope software to display the entire HF spectrum in whatever resolution you wish.  It made tuning my OCF dipole and homemade multiband vertical much easier.  It was really nice to see the SWR pattern on my Log Periodic when I got it up, too.

Alan, the AA-230 is even nicer than the AA-54, if you can swing it.

The only problem I have had with AntScope was getting the COM ports set up.  It wants to default to a channel that is already in use and doesn't like to get along easily with Virtual COM ports.  Once I solved that it was a real winner!

Bill W2PKY

Hey guys, just got mine today and was able to run some fast scans on my multiband vertical. Looks like one band is way out and another just outside the lower band edge. This device is great to get the big picture on an antenna. My lowest SWR is 11 Mhz?? can imagine how that is happening. Have burned out some RG8X on too much power. I take an ohm measurement from each end and locate the short fairly close buy using the ratio of the readings. Plan to try the PC software soon. Very happy with this device so far.

Alan - KA4B

I talked to Jay at Array Solutions about my questions about the AA-54 and the AA-230 Zoom, and he suggested I take a look at the AIM-4300, since it is in the same price range as the AA-230.  He says that the TDR should be much more accurate and it is, in general, a more capable instrument.  Anyone had any experience with this unit?

HCampbell WB4IVF

It’s very capable.  I’ve had the AIM-4300’s immediate predecessor, the AIM-4170, for several years, and am happy with it (the  4300 and 4170 run the same software, I believe).  Among features that I use all the time:

- Continually scanning a user-defined range of frequencies, which, for example, lets you tune a manual antenna tuner without transmitting from the radio.  As you adjust the tuner the analyzer shows you changing SWR, return loss, and complex impedances across an entire band or across several bands, depending on the user-selected upper and lower scan limits, in real time.  I have my AIM connected in parallel with my radios, so I can switch it in any time. 

- You can easily factor out the effect of the transmission line by running a calibration routine, so the readings from the analyzer (located in the shack) are identical to readings you’d get by having the analyzer connected directly to the antenna.   This is useful in building and trimming antennas.  After you initially run the calibration routine, you can save the resulting calibration file and reload the file and use it to rescan any time you want to see SWR, etc., at the antenna feed.  Also, you can save the scan charts, and refer back to them at any time.

- Generate Smith Charts.

It performs lots of other functions, and the user manual is excellent.  You can learn a lot about antenna analyzer utility in general by reading it:

http://kr.basicomm.com/static/files/instructions/AS_VNA/AIM_102615.pdf

There’s also W5BIG’s informative AIM info site that has links to program updates, FAQs, and the AIM user group: 

http://www.w5big.com/

The 4300 is not a stand-alone unit, so you’d need a PC, laptop, tablet, etc., to run it. Not an issue for me, since as I mentioned you can easily "calibrate" the unit so that it thinks it’s connected directly to an antenna and so avoid having to take it outside. 

Array Solutions sells top-notch stuff.   Jay (the owner) and W5BIG are both very knowledgeable and responsive to questions. 

Howard

Luis del MOlino

Alan:

Nice toknow that you have a good balun up on your antenna, but anyhow put another current balun as a choke down at your station. This would destroy any resonance you could have on your coaxial line combined with your earth system.. 

Be carefull with ground earths. They do not have to carry any RF Try to put ferrite beads in any common ground earth system you could have. If any RF circulates through this ground systems, you would have RF been radiated inside your shack not in your antenna. Further to this, this ground system  would be now part of your receiving system, and then would be capturing any noise inside the shack. 

Another possibiity is that you could have a bad contact in some place of your transmission path. This would generate plenty of harmonics and those armonics could change the SWR on the line. Check carefully for bad contacts that could act as diodes and rectify your RF and generate harmonics. The small level of power you do mention of 5 W is astonishing low.

73 Luis EA3OG

Alan - KA4B

Thanks, Howard, WB4IVF.  I have ordered an AIM-4300, in part due to your recommendation.  I also read the reviews on eham.com and played with the software.  When I finish my analysis, and hopefully correct the problem, I will post the results here.  I appreciate the comments of each of you who have shared your insights.

HCampbell WB4IVF

I think you'll find the AIM very useful, like I did.  Let us know the results of your troubleshooting.

Howard

Alan - KA4B

I have been testing my 80 meter dipole with the AIM-4300.  Using the TDR I cannot see any problem with the coax.  The resonant point is 3.875.  The tuner on my Flex 6500 refuses to tune between 3.635 and 3.675.  The SWR at the radio is 4.0 and at the feed point it is 5.0.  My Icom 746 appears to tune the entire band down to below 1.1 SWR.  I do not see any especially high inductance or capacitance.  It is true, as was suggested earlier, that if I add 20 or 30 feet of extra coax it will tune.  With 24 feet of extra coax it does.  If seems to me that to get a tuner to work you should not have to cut and try on coax length.  Something seems amiss to me.

W9OY

So what do you think is amiss?  

At 3.635-75 your antenna's impedance is very far away from the expected 70 j0 ohms of a resonant dipole.  In the frequency range in question, this feedpoint mismatch at the antenna is transformed along the length of the coax to another impedance at the radio end which is outside of the range of the automatic tuning algorithm in the radio.  It probably could be "manually" matched by the components in the internal tuner of the radio, If you had some up/down buttons to choose various component combinations but the automatic tuning algorithm for whatever reason can't find what it considers a safe solution.  The tuner in the 746 which probably has different components or a more granular algorithm does find a safe solution.   

The added  24-30 ft of coax provides a different impedance transformation between the antenna and the radio for which the Flex algorithm does finds a tuning solution.  This is all very normal and is not a mystery.  Automatic tuners are not continuous in their impedance matching.  They do not provide a 100% continuous smooth matching solution.  What they do provide are discrete solutions which represent points on a curve and hopefully one of those points is close enough to provide a reasonable match.  

Not all engineering solutions are perfect.  Many are "just good enough" approximations.  There is nothing wrong with cut and try.  In fact cut and try is how you set your dipole to the resonant freq of 3.875 you wanted.  Why do you then think cut and try does not come into play with feedline length?  The information I gave you to extend your feedline was empirical from my 53 years of experience not just some wild guess.  You need more study on how feedlines, stubs and antennas actually work.  My experience is there a whole bunch of folklore and half truths out there regarding antennas and especially regarding SWR and tuners.  You now own an instrument which will allow you to quantitate your  study.  There are a whole bunch of experiments you can run using that instrument and teach yourself a lot about antenna engineering.

73 W9OY

Alan - KA4B

Lee, please do not take my questions to be disrespectful, but I can calculate my antenna length closely enough to get there on the first try most of the time and I don't cut the wire.  I just fold it through the insulator and twist it over the part I am using.  With coax, cutting and trying is a quite literal experience that I wish to avoid by knowing ahead of time what formula to use for matching the desired wavelength to the desired minimum length as closely as possible so that I do not have to use barrel connectors.

It is my understanding that the autotuner in the Flex 6500 is designed to handle impedance ranging from 16 to 300 Ohms.  Steve Hicks told me it should easily do 10 to 1.  At the point where I am having the problem, the impedance ranges from 125 to 135 Ohms, which I would expect to be well within what the tuner can handle.  The capacitance runs from 275 to 315 pF in that part of the band, but the tuner is handling higher levels on other parts of the band.  It is very likely that I can use some additional electrical engineering knowledge to add to my mechanical engineering background.  What should I be looking at with my AIM-4300 that I am missing?

W9OY

You didn't answer my question.  What do you think is amiss?  I've given you my analysis so what is your analysis of the system presented before you:  

radio > a discontinuous tuner > a length of coax with a mismatch at either end operating at a specific frequency > a non resonant antenna which is generating return loss into  the system.  

That is your system  to  analyse.  Next analyse why 24 ft of added coax completely changes the result.   I do not mean what I say in any pejorative way either.  It seems what you want is a simple formula but the system you have is not simple.  It's literally complex.

73  

Alan - KA4B

Lee, is there any kind of formula for coax length so that you can plan ahead of time how long to cut it?  Are you suggesting the coax should be a multiple of 1/2 wavelength, which it is when I add the 24 feet?  Or are you just suggesting that a person either add remove 1/8 wavelength to alter the tuning effect of the coax significantly.

What I think is amiss is that the tuner is not performing according to spec either because of incorrect assembly, bad solder joint, or poor programing.  Again, when I spoke at length with Steve Hicks, at the Huntsville Hamfest in August, he told me there should be no problem with tuning in this situation.

N9VC

Alan,

Lee is giving you sound advice. An excellent article to read
was in QST 1956. 'My feedline tunes my antenna.' A great article
with sound reasoning as to why things happen with SWR.
A single band antenna is resonant at one frequency. Deviate from
that freq and when you get far enough, things happen.
The bandwidth will depend on the gauge of wire used. I see the
same thing here, the auto tuner will tune some freqs, but refuse others.
 It  does not like seeing some loads with reactances it cannot tune to.
Adding 24 feet to your feedline may help in that part of the band,
but cause problems elsewhere. As was said earlier, the tuner cannot
tune in a continuous manner, hence the combinations will not find a
suitable match.
With my 6500, I use a VL-1000 amp. It will not find a match for my
dipole at 7076 KHz. But if I have the autotuner match at 7105 or so,
it will hold the match at 7076, go figure. And that tuner uses motorized
capacitors.
I complicate matters by using all RG 6 feedline for all my antennas here.
And likewise, my 75 meter double bazooka can be tuned on 160 with the
Flex tuner. I just look at it as a bonus and not wonder why it cannot
find a match in some other areas. 
If you are an ARRL member, find the article on their website. It can
answer some questions.   

73, Jim N9VC

Steve W4YHD

I have had similar problems with the internal tuner in my 6700 not achieving a match on my 80 m dipole.  The tuner has difficulty with a VSWR of 2.5:1 and higher in the 3.6 MHz to 3.9 MHz range.
Often the tuner result is a match that is worse than at the start when trying to match a 2.5 :1 SWR or higher, and/or the tuner display is simply blank after attempting a tune, and often with some strange tuned SWR result achieved.  I have used an LP-100A meter with the coupler connected directly with an adapter to the 6700 ANT 1 connector to verify SWR applied to the radio.  I also notice difficulty with the 6700 internal tuner on the 6 m band as well.  All other bands seem to tune all OK as expected.  Looks like some additional improvement is needed for the internal tuner on 80 m and 6 m.  I did just get my 6700 back from Flex after a repair, and everything was checked and verified to be OK.

Alan - KA4B

Jim, I am not an ARRL member, however the 1956 article is long out of copyright, so if you would be willing to email it to the my email address on record with QRZ.com I would like to read it.  Thanks.

Mike W1BFA in Maine

Just to add a little more confusion to the question, when I saw Alan's first post I had to try my full wave 160 horizontal loop on the frequency Alan mentioned.  I saw a similar result.  It is fed with balanced line for 40 feet then 50 feet of 52 ohm coax to the shack.  I added 25 feet of coax and it now allows the tuner to tune through the "notch".  For the heck of it I subbed a 50 foot for the 25 and it tunes even better.  When doing a SWR sweep the additional feed line moves the apparent resonance up some.  Not what I would have expected.  Also the antenna appears a little flatter on 160.  antennas are fun...   Mike

Alan - KA4B

Thanks to the advice of Lee, the QST article mentioned by Jim and a great deal of experimenting with measurements on the AIM-4300, I see a substantial improvement.  The 6500 has gone from not being able to tune the SWR of 2.5 to 3 at all in the range of 3.635-3.675, and in general doing a poor job of tuning the entire 80 meter band, (with a 2.2 tuned SWR at the high end of the band) to now tuning the entire band down to 1.37 at the low end to 1.16 at the high end.  

My conclusion is that matching the wavelength of the coax (using the velocity factor) to about 3.700 MHz produces the smallest change in the phase angle as seen at the 6500.  Perhaps the autotuner was designed to work properly when connected to the feed point of an 80 meter antenna rather than with a variety of different coax lengths.  In any case, it works well in this situation, and it worked very poorly when the wavelength of the coax was 83% of the wavelength at 3.700 MHz.

From what I have read over the years I would not have expected to have to cut the coax to a special length to make the tuner work with a standard design 80 meter dipole at 35 feet, but in this case it seems to be the only thing that works.  I guess its a case of 'live and learn.'

Thanks everyone for your input.