Elecraft

Using XG1 to measure feedline losses

Classic

List

Threaded

16 messages Options

Vic Dively

Using XG1 to measure feedline losses

Right now I'm using an end fire array antenna that I also can use as a 52
ft. circumference loop. Given the very high SWR on the feedline,
especially when running the loop on 40M and 80M, I recently swapped out
the 18 feet of 450 ohm "window" ladder line between my tuner and the
loop, and replaced it with a homemade ladder line made from 3/8" outside
diameter copper refrigerator tubing. The spacing between centers of the
tubing is 3/4 inch. I immediately noticed much stronger signals after
doing the feedline swap. That had me wondering just how much loss I was
experiencing with the ladder line, in comparison to my current feedline.
I was thinking about how I could measure this. I thought about using
the XG1, attaching it to a 10 ft. dipole made of insulated wire, then
taping the dipole to the center of the loop opposite the feedline, using
the XG1 to generate a weak signal onto the antenna. Then I would measure
the signal strength at the K2 speaker output, going through the
tuner,(after having first tuned up the antenna to 7.040 without the XG1
attached). Then I would repeat the same experiment with the different
feedline and then compare the difference. One of my concerns would be
the impact on the XG1 by hooking it up to a 10 ft. or so dipole, and the
fact that it would not be transmitting into a 50 ohm load.

Is there an easier way to do this? The loop was doing fine with the
ladder line, I used it for the ARCI Spring QSO Party, but I'd like to
improve on it.

Vic
KG4HTT
_______________________________________________
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

Ron D'Eau Claire-2

RE: Using XG1 to measure feedline losses

Vic wrote:
Right now I'm using an end fire array antenna that I also can use as a 52
ft. circumference loop. Given the very high SWR on the feedline, especially
when running the loop on 40M and 80M, I recently swapped out the 18 feet of
450 ohm "window" ladder line between my tuner and the loop, and replaced it
with a homemade ladder line made from 3/8" outside diameter copper
refrigerator tubing... I immediately noticed much stronger signals after
doing the feedline swap. That had me wondering just how much loss I was
experiencing with the ladder line, in comparison to my current feedline.

I was thinking about how I could measure this... using the XG1, attaching
it to a 10 ft. dipole made of insulated wire, then taping the dipole to the
center of the loop opposite the feedline, using the XG1 to generate a weak
signal onto the antenna. Then I would measure the signal strength at the K2
speaker output, going through the tuner,(after having first tuned up the
antenna to 7.040 without the XG1
attached). Then I would repeat the same experiment with the different
feedline and then compare the difference. One of my concerns would be the
impact on the XG1 by hooking it up to a 10 ft. or so dipole, and the fact
that it would not be transmitting into a 50 ohm load.

Is there an easier way to do this?

----------------------------------------

That sounds like a good idea, Vic.

One good preliminary test might be to disconnect the loop at the feedline,
tie the ends of the feeders together and see if you can detect any signal
from the XG1 after it is in place. That's to be sure you aren't picking up
anything by any means other than the loop.

Turn off the K2's AGC and make sure you aren't overloading the K2 at any
point. If you have one, a step attenuator at the antenna connection to the
K2 would give you actual dB measurements, based on comparisons, but if you
aren't overloading anything and the K2's AGC is off, you can calculate the
dB difference by comparing the voltages.

I'm a long-time user of open wire line, and I've slipped and called it
"lossless" more than once. Compared to coax, it seems so, but it really
isn't. And ladder line has a much higher potential for losses than true open
wire line because of its tiny conductors. My open wire line feeding my
doublet is made of #12 copper wire, and I've occasionally thought of
switching to 1/8" or 1/4" copper tubing instead. The issue is, of course,
ohmic losses at the high current points. The currents can get really high -
many, many amperes of current flowing in the wire - and that leads to high
resistive losses in the wires. With true open wire line (mine is spaced with
dog-bone insulators at about 3-foot intervals) the insulation losses are
miniscule, but the resistive losses can still cause a lot of trouble with
high SWR's.

One real advantage of open wire (or ladder) line is the relatively high
impedance. When 50 ohm coax is looking into, say a 2,000 ohm impedance the
SWR is 40:1. A 450 ohm open wire line looking into the same 2000 ohms has an
SWR of only 4.5:1. In the other direction if the antenna presents an
impedance of, say, 10 ohms, the 50 ohm line shows 4:1 while the 450 ohm line
shows 9:1. In either case the 450 ohm line isn't going to show anything like
the SWR, or RF current extremes 50 ohm line can experience.

So, depending upon the SWRs you have on that feedline, I can readily imagine
that you will see an improvement of several dB by switching to a more
efficient line.

Ron AC7AC

_______________________________________________
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

Stuart Rohre

Re: Using XG1 to measure feedline losses

In reply to this post by Vic Dively

Can the XG1 transmit into a 50 ohm resistor with short stiff wire leads to
couple, near the loop?

-Stuart
K5KVH

_______________________________________________
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

wayne burdick

Re: Using XG1 to measure feedline losses

Administrator

The XG1 can be connected to virtually any load without being damaged.
However, if the load doesn't look like 50 ohms at 7.040 MHz, what it
puts out may not be exactly 1 or 50 uV.

Whether you'll hear even the 50-uV signal at any distance remains to be
seen. Let us know how far away you can get and still hear it!

73,
Wayne
N6KR

On Apr 19, 2005, at 10:23 PM, Stuart Rohre wrote:

> Can the XG1 transmit into a 50 ohm resistor with short stiff wire
> leads to
> couple, near the loop?
>
> -Stuart

---

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

Vic Dively

Re: Using XG1 to measure feedline losses

In reply to this post by Vic Dively

I tested the idea tonight, and got some surprising results. I attached
the XG1 to a 12 foot dipole, which I taped to the center of my 52 ft.
loop (16 X 10 ft.) opposite the feedpoint. I used insulated wire, so
there was not an electrical connection from the XG1 to the antenna, just
wanted the dipole to couple with the loop. I could distinctly hear the 1
microvolt tone at 7.040 MHz. Weak but clearly there. So I turned it up
to 50 microVolt to get a better signal to noise ratio, and hopefully
highlight the differences between the two feedlines. I cranked the RF on
my K2 all the way up, but left the audio at 9 o'clock to keep the signal
from being too loud. I measured 9 millivolts AC at the speaker output.
I then swapped out the 3/8 inch copper ladder line and attached 450 ohm
"window" ladder line. I re-tuned the tuner to get a perfect match. I
then re-attached the XG-1 (I left the two wires taped onto the loop, so
there isn't a variable there) and turned it on. I was surprised to get a
higher reading, 23 millivolts AC. That wasn't what I expected. I turned
the XG1 down to the 1 microvolt setting and listened, and it was clearly
louder and with a better signal to noise ratio than what I heard with the
fat ladder line. So I swapped lines out again and tested the fat ladder
line again. It was back down to the 9 millivolt AC level again...

So, what did I conclude? Well, first of all, my fat ladder line wasn't
run the same way as the original "window" ladder line was run. So when I
swapped out feedlines, and re-routed the feedline, I changed two
variables, and I assumed that my noticeably louder signal was because of
the new fat feedline. When I did the test tonight, I routed the window
line the same general way as the fat ladder line. Maybe my homemade
ladder line, which I'm sure isn't 450 ohm, does not provide the tuner as
efficient a match, since I'm measuring both feedline losses as well as
the associated tuner losses when matching the antenna. And the total
system is what really matters.

So, I'll ponder this, and re-install the window ladder line, but with the
new routing. My wife will be happy, as she wasn't thrilled to see the
bigger hole in the (guest) bedroom ceiling, with two copper pipes running
out down to my tuner. Over the past few years she has come to accept
black window ladder line dangling from a hole in the ceiling. And I
learned that the XG1 can be a useful tool to test a feedline/tuner setup.

Vic
KG4HTT

On Tue, 19 Apr 2005 22:47:44 -0700 wayne burdick <[hidden email]>
writes:
> The XG1 can be connected to virtually any load without being damaged.
> However, if the load doesn't look like 50 ohms at 7.040 MHz, what it
> puts out may not be exactly 1 or 50 uV.
> Whether you'll hear even the 50-uV signal at any distance remains to
> be seen. Let us know how far away you can get and still hear it!
>
> 73,
> Wayne
> N6KR
_______________________________________________
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

Ron D'Eau Claire-2

RE: Using XG1 to measure feedline losses

Vic, KG4HTT wrote:
I tested the idea tonight, and got some surprising results. I attached the
XG1 to a 12 foot dipole, which I taped to the center of my 52 ft. loop (16 X
10 ft.) opposite the feedpoint... I cranked the RF on my K2 all the way up,
but left the audio at 9 o'clock to keep the signal from being too loud. I
measured 9 millivolts AC at the speaker output.
I then swapped out the 3/8 inch copper ladder line and attached 450 ohm
"window" ladder line...I was surprised to get a higher reading, 23
millivolts AC. That wasn't what I expected. I turned the XG1 down to the 1
microvolt setting and listened, and it was clearly louder and with a better
signal to noise ratio than what I heard with the fat ladder line. So I
swapped lines out again and tested the fat ladder line again. It was back
down to the 9 millivolt AC level again...

--------------------

You haven't described the tuner setup. You are absolutely right that you are
measuring the combined system losses, not just the feedline losses.

One BIG factor is whether you are using a balun in the system. I've yet to
find a balun that isn't unpredictable and frequently lossy when used in a
feedline with high reactance, that is any balanced feedline with an SWR much
above 1:1, no matter the type.

For just that reason I still use a link-coupled fully-balanced tuner when
feeding my doublet with open-wire line. It takes me 30 seconds to change
bands, but I've never seen any automatic system/balun combination that is
better and many that I've tried are much worse.

The losses in a balun all depend on just how high the SWR on the line is -
that is, how much reactance is present.

Thanks for sharing your results Vic!

Ron AC7AC

_______________________________________________
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

brianboschma

Re: Using XG1 to measure feedline losses

In reply to this post by Vic Dively

I have been using my K2 for antenna measurements but the approach I have
taken is to measure the AGC voltage and calibrate it against known
sources in 10 db increments. The chip used for AGC has a linear
region, or I should say linear with respect the db scale, and becomes
non linear on the low and high signal level ends. In any case I would
think this would be much better with regards to repeatability than
measuring the speaker output.

Regards,

Brian n6iz

[hidden email] wrote:

>I tested the idea tonight, and got some surprising results. I attached
>the XG1 to a 12 foot dipole, which I taped to the center of my 52 ft.
>loop (16 X 10 ft.) opposite the feedpoint. I used insulated wire, so
>there was not an electrical connection from the XG1 to the antenna, just
>wanted the dipole to couple with the loop. I could distinctly hear the 1
>microvolt tone at 7.040 MHz. Weak but clearly there. So I turned it up
>to 50 microVolt to get a better signal to noise ratio, and hopefully
>highlight the differences between the two feedlines. I cranked the RF on
>my K2 all the way up, but left the audio at 9 o'clock to keep the signal
>from being too loud. I measured 9 millivolts AC at the speaker output.
>I then swapped out the 3/8 inch copper ladder line and attached 450 ohm
>"window" ladder line. I re-tuned the tuner to get a perfect match. I
>then re-attached the XG-1 (I left the two wires taped onto the loop, so
>there isn't a variable there) and turned it on. I was surprised to get a
>higher reading, 23 millivolts AC. That wasn't what I expected. I turned
>the XG1 down to the 1 microvolt setting and listened, and it was clearly
>louder and with a better signal to noise ratio than what I heard with the
>fat ladder line. So I swapped lines out again and tested the fat ladder
>line again. It was back down to the 9 millivolt AC level again...
>
>So, what did I conclude? Well, first of all, my fat ladder line wasn't
>run the same way as the original "window" ladder line was run. So when I
>swapped out feedlines, and re-routed the feedline, I changed two
>variables, and I assumed that my noticeably louder signal was because of
>the new fat feedline. When I did the test tonight, I routed the window
>line the same general way as the fat ladder line. Maybe my homemade
>ladder line, which I'm sure isn't 450 ohm, does not provide the tuner as
>efficient a match, since I'm measuring both feedline losses as well as
>the associated tuner losses when matching the antenna. And the total
>system is what really matters.
>
>So, I'll ponder this, and re-install the window ladder line, but with the
>new routing. My wife will be happy, as she wasn't thrilled to see the
>bigger hole in the (guest) bedroom ceiling, with two copper pipes running
>out down to my tuner. Over the past few years she has come to accept
>black window ladder line dangling from a hole in the ceiling. And I
>learned that the XG1 can be a useful tool to test a feedline/tuner setup.
>
>Vic
>KG4HTT
>
>On Tue, 19 Apr 2005 22:47:44 -0700 wayne burdick <[hidden email]>
>writes:
>
>
>>The XG1 can be connected to virtually any load without being damaged.
>>However, if the load doesn't look like 50 ohms at 7.040 MHz, what it
>>puts out may not be exactly 1 or 50 uV.
>>Whether you'll hear even the 50-uV signal at any distance remains to
>>be seen. Let us know how far away you can get and still hear it!
>>
>>73,
>>Wayne
>>N6KR
>>
>>
>_______________________________________________
>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

Vic K2VCO

Re: Using XG1 to measure feedline losses

In reply to this post by Vic Dively

[hidden email] wrote:

> I tested the idea tonight, and got some surprising results. I attached
> the XG1 to a 12 foot dipole, which I taped to the center of my 52 ft.
> loop (16 X 10 ft.) opposite the feedpoint.

I'm not sure why you got the result that you did, but it's possible that the
feedline picked up some xg1 signal in common mode. If you have an antenna
analyzer like the MFJ-259B there's a simpler way to measure the total system loss:

1) Tune your loop for as close to 1:1 as you can get, using the antenna analyzer
connected to the tuner input.

2) Then either short the end of the feedline or disconnect it from the loop.
The SWR should now be Very High.

3) Switch the antenna analyzer to Coax Loss mode, if it has one! It will tell
you what the loss is in db.

If it doesn't have a loss mode (like the Autek VA-1), note that the measured
impedance of the shorted or open line varies as you change frequency. Measure
the MINIMUM impedance at the impedance dip closest to the frequency at which you
want to measure the loss, and multiply it by 0.17 to get the loss in db.

What happens is that if the feedline and tuner had zero loss, then you would get
100% of the forward power reflected from the short or open. But you don't,
because some of the power is eaten up by losses in the feedline and the tuner.
So the analyzer can compute the loss.

--
73,
Vic, K2VCO
Fresno CA
http://www.qsl.net/k2vco

_______________________________________________
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

Geoffrey Mackenzie-Kennedy-2

!Re: Using XG1 to measure feedline losses

In reply to this post by Ron D'Eau Claire-2

Ron D'Eau Claire wrote:

One BIG factor is whether you are using a balun in the system. I've yet to
find a balun that isn't unpredictable and frequently lossy when used in a
feedline with high reactance, that is any balanced feedline with an SWR much
above 1:1, no matter the type.

For just that reason I still use a link-coupled fully-balanced tuner when
feeding my doublet with open-wire line.

The losses in a balun all depend on just how high the SWR on the line is -
that is, how much reactance is present.

............................................................................
............

AMEN !! Seems to me that baluns of the *transmission line* type
have become *magic boxes* and so often used with high load (feeder) VSWR -
say >3:1. Just because the balun does not get hot or explode at the
transmitter power used, leads folk to believe that the balun is not adding
loss. Perhaps the habit of some manufacturers of Antenna System Couplers who
put a balun at the antenna feeder side of an unbalanced T or L network to
save cost has something to do with this.

4 wire cross connected line is very useful, especially at VHF. The
loss due to radiation is considerably less than the radiation loss found in
2 wire line, the ohmic loss is less since each *side* consists of two wires
in parallel, and low Zo lines can be built e.g. 200 ohms.

Ron, you have made my day !

73,
Geoff,
GM4ESD

_______________________________________________
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

Stuart Rohre

Re: Using XG1 to measure feedline losses

In reply to this post by Vic Dively

Very interesting test and another use for the XG1!

Remember, the fat ladder line may be not canceling the currents, ie is not
as balanced as the more narrow line. Or maybe, the window narrow line is
just the better match to the tuner or it is the routing and the surrounding
influences.

-Stuart
K5KVH

_______________________________________________
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

Stuart Rohre

Re: Using XG1 to measure feedline losses

In reply to this post by brianboschma

That is a good point, as you do not want the AGC to activate giving a wrong
reading on the comparisons.

If it were possible to disable AGC, you would be able to be sure it is out
of affecting measurements. Also, you should use a step attenuator to make
sure no other circuits are overloaded for the testing. Make sure your
indicating point gives 10 dB steps or whatever step for each attenuator
change to verify linearity at the control settings you have picked.

-Stuart
K5KVH

_______________________________________________
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

Stuart Rohre

Re: Using XG1 to measure feedline losses

In reply to this post by Ron D'Eau Claire-2

An external 4:1 balun is typically less lossy than the built in 4:1 baluns.
Better still, use the W5YR idea, a bead ferrite choke on the coax output,
then connect there with a coax to binding post adapter for balanced line,
and you eliminate questions about your balun.

An alternate is to use a coax cable choke on a plastic Coke bottle, (liter
bottle) for form.

-Stuart
K5KVH

_______________________________________________
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

Ron D'Eau Claire-2

RE: Using XG1 to measure feedline losses

In reply to this post by Stuart Rohre

Stuart K5KVH wrote:

An external 4:1 balun is typically less lossy than the built in 4:1 baluns.
Better still, use the W5YR idea, a bead ferrite choke on the coax output,
then connect there with a coax to binding post adapter for balanced line,
and you eliminate questions about your balun.

------------------------
Beads are great if you're using coax, mostly because they are low-loss and
don't change the impedance presented to the tuner.

Sometimes a 4:1 balun is the worst choice. It reduces the impedance the
tuner 'sees' for a load. If it's being used as it was intended - in a
fairly-well matched load such as 300 ohm line to a folded dipole cut to
length - it's FB. In that case it'll translate the 300 ohms to 75 ohms, and
easy match for the tuner.

But, if the 4:1 balun is looking into an antenna or feed line with a high
SWR the impedance it's driving might be very low, perhaps only a few ohms.
If the 4:1 balun can handle it at all, it'll divide that impedance by four
again, making it a very tough match for most tuners.

Ron AC7AC

_______________________________________________
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

Leigh L. Klotz Jr WA5ZNU

Re: Using XG1 to measure feedline losses

Administrator

For shirt verticals with a low impedance I use a step-up balun, but
again it seems to work better when the X is not the problem, but tehe
R. WB6ZQZ makes a kit.
On Thu, 21 Apr 2005 12:53 pm, Ron D'Eau Claire wrote:
> But, if the 4:1 balun is looking into an antenna or feed line with a
> high
> SWR the impedance it's driving might be very low, perhaps only a few
> ohms.
> If the 4:1 balun can handle it at all, it'll divide that impedance by
> four
> again, making it a very tough match for most tuners.
_______________________________________________
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

Stuart Rohre

Re: Using XG1 to measure feedline losses

In reply to this post by Ron D'Eau Claire-2

W5YR, (sk) used the beads for transition to his balanced line adapter to the
coax connection on his tuner. That is the beauty of the cable choke and
Beads, they are operating on the OUTSIDE of the coax section, and thus the
transformation ratio is taken care of inside the tuner, but RF is blocked
from the tuner case.

-Stuart
K5KVH

_______________________________________________
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

Amanda25650

Re: Using XG1 to measure feedline losses

In reply to this post by Vic Dively