Problem matching real Log Periodic to NEC
 

Just thought I'd throw this out there to see if anybody had any
ideas... I was recently playing with an 800 to 1000 MHz Log design. It
came out quite well in the model. It was showing that it should do
about 1 to 1.5 average SWR, with a couple little spikes of 2.0 or so.
So I went ahead and built it.

Now I know from past experience, that testing logs is very tricky.
Unless you have a phase stable cable right up to the feed point, you
can get all kinds of crazy readings. But still, you can usually get
the gist of what's going on. So I go to test the actual finished
antenna, and across the intended range, it's showing SWR as high as 7
or 8, and the best it goes down to is maybe 3. Totally different than
the model. As a test, I went to the NEC model and took off all the UHF
elements, and just added one VHF element to the low end... It showed
around 155 MHz in the model. I did the same thing on the actual
antenna and got 1.1 around 153 MHz. So at that range, it's right on.
But I am baffled as to why it is so terrible on the UHF. I tried
shorting the back end and not shorting. No real difference.

Any ideas?

Thanks,

Problem matching real Log Periodic to NEC
 

On Wed, 29 Sep 2010 16:36:06 -0700 (PDT), Dave D
wrote:

Any ideas?

Inadequate decoupling of the transmission line at the feedpoint.

73's
Richard Clark, KB7QHC

Problem matching real Log Periodic to NEC
 

Dear Dave D. (no call sign):

I have found it highly desirable with large and smaller HF LPDAs either to
decouple at the feed-point or to choke common mode right as the coax leaves
the boom to go down the mast. Of course, one could do both. Once you have
cleaned the transmission line, you need to attend to the termination at the
end of the boom. Your BW is small for a LPDA so design should be straight
forward.

Mac N8TT
--
J. McLaughlin; Michigan, USA
Home:

"Richard Clark" wrote in message
...
On Wed, 29 Sep 2010 16:36:06 -0700 (PDT), Dave D
wrote:

Any ideas?

Inadequate decoupling of the transmission line at the feedpoint.

73's
Richard Clark, KB7QHC

Problem matching real Log Periodic to NEC
 

Thanks guys... I've never had that problem before. Any tips for what
clamp-on ferrites might work best, or how many? Or the best way to
configure them?

Problem matching real Log Periodic to NEC
 

On 30/09/10 15:23, Dave D wrote:
Thanks guys... I've never had that problem before. Any tips for what
clamp-on ferrites might work best, or how many? Or the best way to
configure them?

Distributed capacitance is the enemy of any multi turn balun, whether in
solenoid style (eg Ugly Balun) or toroidal winding. You could however
try to exploit the self resonance by placing it in the band you wish to
use, then the issue is whether the resonance peak in Zcm is sufficiently
wide.

The W2DU style ought have sufficiently low self capacitance to operate
below resonance at 800 to 1000MHz. I would look to #61 material
suppression sleeves, but it may be that a lossier material works for
your application.

Owen

Problem matching real Log Periodic to NEC
 

On 30/09/10 15:23, Dave D wrote:
Thanks guys... I've never had that problem before. Any tips for what
clamp-on ferrites might work best, or how many? Or the best way to
configure them?

I meant also to ask how you dressed the coax away from the feedpoint.

I often see pics of coax tie wrapped to one leg of the transmission line
spine of the LP. Guaranteed to create common mode current.

Owen

Problem matching real Log Periodic to NEC
 

On Sep 30, 12:28*am, Owen wrote:
On 30/09/10 15:23, Dave D wrote:

Thanks guys... I've never had that problem before. Any tips for what
clamp-on ferrites might work best, or how many? Or the best way to
configure them?

I meant also to ask how you dressed the coax away from the feedpoint.

I often see pics of coax tie wrapped to one leg of the transmission line
spine of the LP. Guaranteed to create common mode current.


Thanks Owen,

In this case I had it down the middle between the haves, through clips.

Problem matching real Log Periodic to NEC
 

On 01/10/10 05:06, Dave D wrote:
On Sep 30, 12:28 am, wrote:
On 30/09/10 15:23, Dave D wrote:

Thanks guys... I've never had that problem before. Any tips for what
clamp-on ferrites might work best, or how many? Or the best way to
configure them?

I meant also to ask how you dressed the coax away from the feedpoint.

I often see pics of coax tie wrapped to one leg of the transmission line
spine of the LP. Guaranteed to create common mode current.


Thanks Owen,

In this case I had it down the middle between the haves, through clips.

Does that disturb the characteristic impedance of the spine transmission
line? If so, does that contribute to the problems you observe?

If this LP is mounted to a mast somewhere forward of the rear element, I
would consider dressing the feedline away from the feed point at right
angles to the spine, installing a common mode choke near the feedpoint,
and bringing the feedline to the mast some distance below the spine.

If you were to perform a test where the LP is mounted a wavelength or
more above ground, pointing to the zenith, feedline falling downwards
from the feedpoint, common mode choke near the feedpoint, it would
reveal whether the array is working as intended. If your actual mounting
causes it to work differently, you know where the problem is.

Owen

Problem matching real Log Periodic to NEC
 

On Sep 30, 12:38*pm, Owen wrote:
On 01/10/10 05:06, Dave D wrote:



On Sep 30, 12:28 am, *wrote:
On 30/09/10 15:23, Dave D wrote:

Thanks guys... I've never had that problem before. Any tips for what
clamp-on ferrites might work best, or how many? Or the best way to
configure them?

I meant also to ask how you dressed the coax away from the feedpoint.

I often see pics of coax tie wrapped to one leg of the transmission line
spine of the LP. Guaranteed to create common mode current.

Thanks Owen,

In this case I had it down the middle between the haves, through clips.

Does that disturb the characteristic impedance of the spine transmission
line? If so, does that contribute to the problems you observe?

If this LP is mounted to a mast somewhere forward of the rear element, I
would consider dressing the feedline away from the feed point at right
angles to the spine, installing a common mode choke near the feedpoint,
and bringing the feedline to the mast some distance below the spine.

If you were to perform a test where the LP is mounted a wavelength or
more above ground, pointing to the zenith, feedline falling downwards
from the feedpoint, common mode choke near the feedpoint, it would
reveal whether the array is working as intended. If your actual mounting
causes it to work differently, you know where the problem is.

Owen

Well, it does different things based on how the cable is routed, but
none are really better than the other... just different. I tried
coming off the back, and also at the mid point... It changed, but not
better or worse overall... The points of good and bad just changed
position across the range.

I'm going to try the ferrites. What do you think about these?
http://www.mouser.com/ProductDetail/...PQp4FXwkz9g%3d

Problem matching real Log Periodic to NEC
 

On 9/30/2010 4:56 PM, Dave D wrote:
On Sep 30, 12:38 pm, wrote:
On 01/10/10 05:06, Dave D wrote:



On Sep 30, 12:28 am, wrote:
On 30/09/10 15:23, Dave D wrote:

Thanks guys... I've never had that problem before. Any tips for what
clamp-on ferrites might work best, or how many? Or the best way to
configure them?

I meant also to ask how you dressed the coax away from the feedpoint.

I often see pics of coax tie wrapped to one leg of the transmission line
spine of the LP. Guaranteed to create common mode current.

Thanks Owen,

In this case I had it down the middle between the haves, through clips.

Does that disturb the characteristic impedance of the spine transmission
line? If so, does that contribute to the problems you observe?

If this LP is mounted to a mast somewhere forward of the rear element, I
would consider dressing the feedline away from the feed point at right
angles to the spine, installing a common mode choke near the feedpoint,
and bringing the feedline to the mast some distance below the spine.

If you were to perform a test where the LP is mounted a wavelength or
more above ground, pointing to the zenith, feedline falling downwards
from the feedpoint, common mode choke near the feedpoint, it would
reveal whether the array is working as intended. If your actual mounting
causes it to work differently, you know where the problem is.

Owen

Well, it does different things based on how the cable is routed, but
none are really better than the other... just different. I tried
coming off the back, and also at the mid point... It changed, but not
better or worse overall... The points of good and bad just changed
position across the range.

I'm going to try the ferrites. What do you think about these?
http://www.mouser.com/ProductDetail/...PQp4FXwkz9g%3d


The one question I've not seen asked is how close to what you built is
your model? Everything is critical at those frequencies and as little as
1/32" can make a difference. If your model doesn't match exactly the
antenna then it's no wonder it's not working as expected. To me modeling
the phasing lines between the elements be they separate from the boom or
are the boom is one of the trickiest parts. As an example if you used a
1" square boom and modeled it as a 1" diameter wire there would be a
problem. They are not equal.


John W3JXP