Connecting to your ant with coax

Connecting to your ant with coax.

True story..... (maybe a little too long. You can jump
way down to the paragraph labeled - The Practical Lesson):

I built the KX1 (20/40m) and the ATU per the manual. Then
built the vertical that Wayne suggest in the ATU manual;
about 30ft vertical (some military surplus mast sections
resting on a coke bottle) and 4 16-ft radials laid out on the
ground.

Hooked the KX1 to the antenna at the base of the antenna and
hit the tune up buttons. It tuned beautifully (both bands)! I was doing the 'Happy Dance'! But I didn't want to sit at the base of
the antenna while I was operating, so I found some old coax
with connectors (about 15ft in length) and set myself up on
card table that far away. Hit the tune up buttons again and
found the SWR was terrible! (I was reading the SWR on the KX1
read out.)

Now I am a used-to-be-engineer, but have forgotten all that I
once knew about antennas and coax - OK, OK, I admit it, I've
become an appliance operator! Anyway, retirement does that
to you.

I puzzled over this for a few seconds and decided that the
coax or connectors were bad. So I substituted in another coax
cable (about the same length). Still the SWR was way out.
(You guys who are still active engineers know what's wrong
already, I'm sure. So I am writing for the old geezers like
me or neophytes to the hobby.)

The Practical Lesson:
Well, this coax length is a 'gotcha' - Here's the deal....

Recall from your studying the license manual that coax can
act as an impedance transformer. Every half-wave length, the
coax transforms the load impedance (that's your antenna) back
to its original value.

Coax has an attribute call 'Velocity Factor' (remember that
one?). This VF is different for each type of coax. What that
means is (practically speaking) that the coax really needs to
be adjusted in length by multiplying that physical wave-length
by the VF.

So what I needed for coax length that was some multiple of
half-wave length. Recall again from your earlier studies that
the simple formula for a half-wave antenna length (ft) is:  
468/freq (freq is in mhz) . So for 40m that length is roughly 66 ft.
NOW.... you have to shorten the coax by the VF to get the final physical length. I was using coax type RG-59 with a VF of .68
(from somebody's chart on the web). So, 66 x .68 = 45ft (approx).

I rummaged around in my junque box and found a 42ft length
of coax with BNC connectors on each end. I made up an
additional short length of coax with a BNC on one end and a
couple of Alligator clips on the other (this to connect the
vertical to the 42ft length of coax).Hooked all that up to
the KX1 and hit the tune buttons again!

SHAZAM!!! (an old Gomer Pile expression) - SWR was back to
near normal! (more 'Happy Dancing'.)

Oh did I forget to mention that this same length will work on
20m too!

(OK, no flames now, but if you engineer types want to expand on
this, go right ahead).

73 de Jerry/k6iii
San Jose, CA
KX1 s/n 1101
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Rolf Moberg, OH6KXL]

[hidden email] wrote:

> I rummaged around in my junque box and found a 42ft length
> of coax with BNC connectors on each end. I made up an

Replacing 15ft of coaxial with 42ft typically reduces SWR due to coaxial
line losses. Putting 420 ft instead of 42 ft makes situation better when
looking strictly only SWR meter.

I could comment a lot more about your topic but I don't do it.
Mixing the feets and metric units is not clever idea :) We europeans
cannot clearly see what you US fellows are writing ;)

Using 30 feet (about 10 meter?) vertical in 20m is very poor situtation.
Impedance in the feedpoint is pretty different compared to 50 ohm.

Rolf
oh6kxl
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Mark J. Schreiner]

Pretty good explaination.  I'll only add that the formula of 468/f already has a "correction factor" involved as the real formula is 492/f (for lenth in feet of a half wavelenth with frequency given in MHz).  So, using 492/f and the proper velocity factor depending on type of dielectric you should be able to get pretty close to a physical half wavelength of coax.  

If you have an antenna analyzer (pick your favorite, mine is an MFJ-269), you should be able to tweak it slightly as well.  Actually, 1/4 length stubs are really slick as well, again providing a transformer effect.  This is where it gets really cool.  Cut a length of coax slightly longer than your calculation (based on the proper velocity factor, also divide by 2 if you want a 1/4 wavelenth rather than a 1/2 wavelength) and install a proper RF connector of your choice on one end.  Connect the connectorized end to the analyzer.  You know you are at a quarter wavelength when the open circuit becomes a short circuit (with RF using the analyzer, not at DC).  Then, for fun and to prove to you that you just have a 1/4 wavelength line at that frequency, short the loose end and you should see that the analyzer measures an open circuit.  The reason for this can be explained with some nice formulas, but also graphically what happens is you rotate 1/2 way around the Smith Chart, from either an open or a short circuit, whichever case you have, to the opposite side of the Smith Chart.  If you are doing this with 75 ohm feedline instead of 50 ohm feedline, it will act as a transformer from 100 ohms to 50 ohms also.  So a 100 ohm load if applied directly to your antenna analyzer of choice would show  a 2:1 SWR (assuming Zo = 50 ohms).  Put that 100 ohm load at the end of your 1/2 wavelength 50 ohm feedline and see that it is still a 2:1 SWR.  Now attach it to the end of your 1/4 wavelength 75 ohm feedline and whatch the SWR magically become 1:1!  COOL!  Since Field Day is coming up, use this tip to minimize your SWR (for loads that are slightly high impedance) very easily without raising and lowering the antenna to trim it "just right" and without using an antenna tuner.  BTW, I always keep a couple of lengths of 1/4 wavelength 75 ohm "Q-lines" handy and labelled for what band they are tuned.  I saw an old RF engineer do this at my first Field Day at K8DAC in Saginaw, MI back in about 1983 and I was amazed at the simplicity.  Now I'm an RF engineer and I'm still amazed at the simplicity!

73,

Mark, NK8Q
K2 4786

>From: "[hidden email]" <[hidden email]>
>Date: Fri Jun 17 15:19:36 CDT 2005
>To: [hidden email]
>Subject: [Elecraft] Connecting to your ant with coax

>
>Connecting to your ant with coax.
>
>True story..... (maybe a little too long. You can jump
>way down to the paragraph labeled - The Practical Lesson):
>
>I built the KX1 (20/40m) and the ATU per the manual. Then
>built the vertical that Wayne suggest in the ATU manual;
>about 30ft vertical (some military surplus mast sections
>resting on a coke bottle) and 4 16-ft radials laid out on the
>ground.
>
>Hooked the KX1 to the antenna at the base of the antenna and
>hit the tune up buttons. It tuned beautifully (both bands)! I was doing the 'Happy Dance'! But I didn't want to sit at the base of
>the antenna while I was operating, so I found some old coax
>with connectors (about 15ft in length) and set myself up on
>card table that far away. Hit the tune up buttons again and
>found the SWR was terrible! (I was reading the SWR on the KX1
>read out.)
>
>Now I am a used-to-be-engineer, but have forgotten all that I
>once knew about antennas and coax - OK, OK, I admit it, I've
>become an appliance operator! Anyway, retirement does that
>to you.
>
>I puzzled over this for a few seconds and decided that the
>coax or connectors were bad. So I substituted in another coax
>cable (about the same length). Still the SWR was way out.
>(You guys who are still active engineers know what's wrong
>already, I'm sure. So I am writing for the old geezers like
>me or neophytes to the hobby.)
>
>The Practical Lesson:
>Well, this coax length is a 'gotcha' - Here's the deal....
>
>Recall from your studying the license manual that coax can
>act as an impedance transformer. Every half-wave length, the
>coax transforms the load impedance (that's your antenna) back
>to its original value.
>
>Coax has an attribute call 'Velocity Factor' (remember that
>one?). This VF is different for each type of coax. What that
>means is (practically speaking) that the coax really needs to
>be adjusted in length by multiplying that physical wave-length
>by the VF.
>
>So what I needed for coax length that was some multiple of
>half-wave length. Recall again from your earlier studies that
>the simple formula for a half-wave antenna length (ft) is:  
>468/freq (freq is in mhz) . So for 40m that length is roughly 66 ft.
>NOW.... you have to shorten the coax by the VF to get the final physical length. I was using coax type RG-59 with a VF of .68
>(from somebody's chart on the web). So, 66 x .68 = 45ft (approx).
>
>I rummaged around in my junque box and found a 42ft length
>of coax with BNC connectors on each end. I made up an
>additional short length of coax with a BNC on one end and a
>couple of Alligator clips on the other (this to connect the
>vertical to the 42ft length of coax).Hooked all that up to
>the KX1 and hit the tune buttons again!
>
>SHAZAM!!! (an old Gomer Pile expression) - SWR was back to
>near normal! (more 'Happy Dancing'.)
>
>Oh did I forget to mention that this same length will work on
>20m too!
>
>(OK, no flames now, but if you engineer types want to expand on
>this, go right ahead).
>
>73 de Jerry/k6iii
>San Jose, CA
>KX1 s/n 1101
>_______________________________________________
>Elecraft mailing list
>Post to: [hidden email]
>You must be a subscriber to post to the list.
>Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft   
>
>Help: http://mailman.qth.net/subscribers.htm
>Elecraft web page: http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Trevor Day]

I believe the problem here is that the vertical will not perform
adequately when fed with 50 ohm coax unless it is an odd multiple of a
quarter wave in length (to show something approaching 50 ohms at the
feedpoint of the vertical)  If a tuner is used at the radio end of the
coax to successfully reduce the SWR at non resonant frequencies,
performance will be compromised as the coax lying along the ground now
becomes a radiating part of the aerial system.

Adjusting the length of coax feeder may well allow the radio to 'see' a
reasonable impedance, but is this not just achieving the same effect as
in my paragraph above with the associated problems?  I would argue that
a tuning unit directly at the base of the vertical converting the feed
impedance to 50 ohm unbalanced for the coax is the best solution.

Trev G3ZYY

In message <[hidden email]>,
[hidden email] writes

>Pretty good explaination.  I'll only add that the formula of 468/f
>already has a "correction factor" involved as the real formula is 492/f
>(for lenth in feet of a half wavelenth with frequency given in MHz).
>So, using 492/f and the proper velocity factor depending on type of
>dielectric you should be able to get pretty close to a physical half
>wavelength of coax.
>
>If you have an antenna analyzer (pick your favorite, mine is an
>MFJ-269), you should be able to tweak it slightly as well.  Actually,
>1/4 length stubs are really slick as well, again providing a
>transformer effect.  This is where it gets really cool.  Cut a length
>of coax slightly longer than your calculation (based on the proper
>velocity factor, also divide by 2 if you want a 1/4 wavelenth rather
>than a 1/2 wavelength) and install a proper RF connector of your choice
>on one end.  Connect the connectorized end to the analyzer.  You know
>you are at a quarter wavelength when the open circuit becomes a short
>circuit (with RF using the analyzer, not at DC).  Then, for fun and to
>prove to you that you just have a 1/4 wavelength line at that
>frequency, short the loose end and you should see that the analyzer
>measures an open circuit.  The reason for this can be explained with
>some nice formulas, but also graphically what happens is you rotate 1/2
>way around the Smith Chart, from either an open or a short circuit,
>whichever case you have, to the opposite side of the Smith Chart. If
>you are doing this with 75 ohm feedline instead of 50 ohm feedline, it
>will act as a transformer from 100 ohms to 50 ohms also.  So a 100 ohm
>load if applied directly to your antenna analyzer of choice would show
>a 2:1 SWR (assuming Zo = 50 ohms).  Put that 100 ohm load at the end of
>your 1/2 wavelength 50 ohm feedline and see that it is still a 2:1 SWR.
>Now attach it to the end of your 1/4 wavelength 75 ohm feedline and
>whatch the SWR magically become 1:1!  COOL!  Since Field Day is coming
>up, use this tip to minimize your SWR (for loads that are slightly high
>impedance) very easily without raising and lowering the antenna to trim
>it "just right" and without using an antenna tuner.  BTW, I always keep
>a couple of lengths of 1/4 wavelength 75 ohm "Q-lines" handy and
>labelled for what band they are tuned.  I saw an old RF engineer do
>this at my first Field Day at K8DAC in Saginaw, MI back in about 1983
>and I was amazed at the simplicity.  Now I'm an RF engineer and I'm still amazed at the simplicity!
>
>73,
>
>Mark, NK8Q
>K2 4786

--
Trevor Day
UKSMG #217
www.uksmg.org

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com