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: Antenna Matching Harness  (Read 1015 times)
KC9ERZ
Member

Posts: 17




Ignore
« on: August 10, 2009, 01:06:13 PM »

I have built two yagis for 70cm on the same boom (cross-yagis).  They are physically 90deg out of phase for circular-polarization purposes.  They were built using the "Cheap Yagi" design where you solder the coax directly to the driven-element.  Now I need to build a matching harness to feed both antennas.  I understand the formula for calulating the length for each side using 75 ohm cable.  I also understand that everyone recommends using a T-adapter instead of trying to splice two 75ohm cables to a 50ohm cable.  So my question is what do I include in the measurements?  So for example if each leg of my harness is supposed to be 4.5 inches long, do I include the T-adapter in the measurements.  What about the ends that connect directly to the driven element?  Do I measure to the very end of the center-conductor, or just to the point where the outer-jacket was stripped off?

I've tested each antenna seperately for a low SWR and both are under 1.5:1.  But my first attempt at the matching harness failed (SWR greater than 5:1).

Thanks in advance for your help.

Greg
KC9ERZ
Logged
WA5UP
Member

Posts: 5




Ignore
« Reply #1 on: August 10, 2009, 01:26:36 PM »

Hi Greg,

I can't give you an exact length, but I can tell you a technique that my elmer taught me for making 1/4 wave stubs.  When cutting 1/4 wave stubs connect a signal source to a "T" adapter.  Take the second side of the "T" and connect it to a power meter (or spectrum analyzer, etc).  Put the proper connector on one end of the feedline that you are trying to cut to 1/4 wave and connect it to the third port on the "T".  Apply signal to the system (low level of course) and trim the 1/4 wave stub for a dip in signal on the meter or spectrum analyzer.  The added length in the "T" should be a very close the length that the second connector will add on the other end of the stub.  Even if you're off just a little bit you'll still end up with a much closer tuned stub than what a calculation would give you due to variances in velocity factor, etc of the feedline (trust me I've tried it both ways).  If you need a diagram or further details let me know.

73

Andy Knepler - WA5UP
Logged
KC9ERZ
Member

Posts: 17




Ignore
« Reply #2 on: August 10, 2009, 01:34:50 PM »

Andy,

Thanks for the quick response.  I want to make sure I understand what you recommend:

1)  Connect an RF power source (e.g. transmitter) to the T-connector input.

2)  Connect one side of the T-connector to one of my antennas that I am trying to feed using the approximate length of 75ohm cable.

3)  Connect the other side of the T-connector to a meter (SWR meter???).

Is the above correct?  I assume I would need to connect the end to the SWR meter using the same length of 75ohm coax (1/4 wave) as what I have connected to the antenna?

If I am putting 10watts into the T-connector, I should see 5-watts on my meter?

Again, thanks for your help.....any further clarification you could provide would be great!

Thanks,
Greg
KC9ERZ
Logged
N7WS
Member

Posts: 35




Ignore
« Reply #3 on: August 10, 2009, 01:52:04 PM »

I guess that you are trying to use two 1/4 wavelengths of 75-ohm cable to transform the 50-ohm feed points to 100-ohm, where they can be paralleled to equal 50-ohm.

If so, then *everything* that's coaxial between the junction to the DEs needs to be included in the length calculation.

The problem with using a standard coax tee is that it will probably be 50-ohm, but this is mostly unavoidable and pragmatically speaking, doesn't matter anyway.

I've never done this, but I would be tempted to use "F" connectors on a homemade tee.

All this said, this is still pretty forgiving.  I ran an analysis where the 75-ohm lines were between 80 and 100 degrees long and with perfect loads (antenna end) the 50-ohm SWR at the junction is < 1.2:1

If you are seeing 5:1 something else is wrong. What are you measuring with?
Logged
KC9ERZ
Member

Posts: 17




Ignore
« Reply #4 on: August 10, 2009, 02:07:44 PM »

Your first sentence is exactly correct and what I am trying to do.  I am using an SWR meter to meaure the the entire system.  In other words I was measuring once I had the 50ohm cable connected to the two 75ohm legs, which were then soldered to the DEs of each antenna.

As I mentioned in my first post, the yagis are based on the "Cheap Yagi" design, where the DE is a folded element.  The coax is soldered directly to the DE.  So in my case, I have two antennas on the same boom.  I tested each antenna seperately by temporarily hooking a 50ohm cable to the DE through an SWR meter and measuring.  Each antenna was less than 1.5:1.  

Then I calculated the length of the two matching sections of 75 ohm coax.  I soldered these directly together along with the 50 ohm coax (kind of a home-made T-adapter).  Then finally I soldered each of the 75 ohm legs to the DEs.  This is the point I measured the SWR and found it 5:1+.

Upon further reading it was suggested to not solder the three coaxes together, but rather to use a T-connector.  Also, I read where you include tip-to-tip when calculating length.  So in other words if the length comes out to be 4.5 inches for 1/4wave, then that is the total length, including the PL259 connectors end to end.  In my case there would only be one PL259 connector per leg and I the other end is soldered directly to the DE.

I have not tried using the T-connector or including the connectors in the overall lengths.  So possibly by matching stubs were to long?Huh

Not sure if any of the above makes sense.....any help you could provide would be greatly appreciated.  

Thanks,
Greg
KC9ERZ
Logged
WA5UP
Member

Posts: 5




Ignore
« Reply #5 on: August 10, 2009, 04:19:52 PM »

Greg,

I made a quick diagram of the test setup: http://www.wa5up.com/images/QuarterWaveStubTuning.jpg

Basically on one side you have an RF source tuned to the frequency you are trying to cut the 1/4 wave stub to.  On the other end you have a power meter.  You should use the formula to get the rough length of the coax, and then cut it an inch or two long.  Put a connector on one end and connect it to the "T".  Then, trim the coax until the power meter dips.  This is when you have a true 1/4 wave stub.  You can calculate the length by taking the wavelength X velocity factor of coax, but I've found that manually measuring / tuning is a much more accurate way to go.

Also, I agree with the previous post.  If you're seeing 5:1 something is bad wrong, probably more than just the length of the matching harness.

Hope this helps,

Andy Knepler - WA5UP
Logged
WA5UP
Member

Posts: 5




Ignore
« Reply #6 on: August 10, 2009, 04:22:42 PM »

Also, should note that the feedline between the RF signal generator and the "T", as well as between the RF power meter and the "T" is 50 ohm.  The stub that you are trying to tune should be 75 ohm.  You should perform this process twice to get two tuned 1/4 wave pieces of 75 ohm cable.  Then use those to pieces in your matching harness.  They should go from the "T" to the antennas, and the "T" should be fed with 50 ohm coax.  The setup pictured is for testing / tuning only, and won't be used in the actual harness.
Logged
KC9ERZ
Member

Posts: 17




Ignore
« Reply #7 on: August 10, 2009, 07:23:29 PM »

Thanks Andy for all the pointers.  I will try this tomorrow after work.  Your testing method makes alot of sense, especially with the drawing!

Thanks again!

Greg
KC9ERZ
Logged
KC9ERZ
Member

Posts: 17




Ignore
« Reply #8 on: August 11, 2009, 05:56:48 AM »

Andy, after studying your diagram further I have another question.  As I trim my 75ohm stub, I assume I am looking for a low SWR on the power meter?  Having the open stub seems strange to me?  Also if I am looking for a low SWR on the meter, what percentage of the power am I looking for?  In other words if I am putting 1 watt from my TX should I see 1/2 watt on the meter?

Thanks again,
Greg
KC9ERZ
Logged
N3OX
Member

Posts: 8854


WWW

Ignore
« Reply #9 on: August 11, 2009, 06:20:15 AM »

"In other words if I am putting 1 watt from my TX should I see 1/2 watt on the meter?
"

An *open* stub 1/4 wavelength long appears to be a *short circuit* at the other end.

So if the cable from the tee is connected to the usual input of the meter, you're looking for lowest "forward power".  If you don't have anything on the output of the meter, it's not really reading "forward power" anymore, so don't expect it to make sense in watts.  No power can flow into the resulting open circuit.   Just make it go as close to zero as possible.

I would recommend using the lowest power setting on your radio and/or putting a dummy load on the "signal generator" side of the tee if you're transmitting a fair amount into this setup.  Keep in mind that you're transmitting into a very nasty load that gets nastier as you get closer to the right length :-)

73
Dan
Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
WA5UP
Member

Posts: 5




Ignore
« Reply #10 on: August 11, 2009, 09:17:42 AM »

Right, in my drawing I was actually assuming that you were going to be using test equipment.  If you're going to use the radio as a source and an inline power meter instead of a spectrum analyzer do the following:

1. Set your radio to the lowest power level that you can will still make the meter move at least to half scale.

2. Set the meter to forward power.

3. Briefly key the mic and take a reading.

4. Be sure to key down before trimming the stub.

5. Make your cut on the stub.

6. Briefly key the mic and make your measurement to see how things have changed.

Make sure to key your radio as little as possible to minimize the risk of damaging the finals due to the mismatch.  It would be beneficial to make a chart to record your measurements so that you can see the change.  Also, don't worry about the value of the forward power, just tune for the dip.

Andy Knepler - WA5UP
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!