On 13 May 2007 16:55:44 -0700, dykesc wrote:

Richard, I tired measurements again with my twin lead directly
terminated to the 259b. I got better, more consistent results after
taking great care to insure the analyzer and line were well isolated
from ground, other conductors, and myself.

Hi Dykes,

It should be comforting that observing standard precautions produces
repeatable results.

I then took the same
measurements with the 4:1 balun between the twin lead and the
analyzer. Unfortunately the results create new concerns. For example
at 7.185 Mhz with the balun in the circuit (tuner in bypass mode) I
got 19 -j48. Again at 7.185 Mhz with the balun out (twin lead directly
terminated to 259b) I got 159 -j443. Doesn't look like 4:1 to me.
Similar spreads in the 80m and 20m bands.

This sounds like you've inserted the entire tuner to obtain the 4:1
BalUn (once you threw the right switches).

If, as you say, education is a principle goal, then build a proper 4:1
current BalUn. It is actually quite simple and requires only two
transmission lines and a several dozen beads. Basically it is two 1:1
current BalUns fed in parallel and loaded in series.

You will be simultaneously checking your system, and testing the
authenticity of the MFJ claim:
The MFJ manual for the 993b tuner says the balun is a 4:1 "current"
balun. Haven't looked inside to confirm this.

You have the means to test the assertion, use your 259 to measure the
isolation of the BalUn. This was the subject of a recent thread.

Thanks. I'll search for the thread. Sounds like fun.

If after a fruitless search (it's easy enough to get slogged down in
the snow drift of useless posts here) you don't find it, ask for help
here. Mentioning you tried the archives will save others from whining
about how much effort they went to answer a stupid question. (I won't
whine, and I never call any question stupid - although I frequently
dope slap some of the denser questioners.)

Fixation on BalUns has clouded a simpler solution: wind a choke in
the line and dump the ferrites of suspect quality.

Would you please elaborate on this? Wind a choke where? In the twin
lead?

Sure, twist it candy cane (or barber shop pole) style and wind it
around a liter bottle with at least its width as separation between
windings.

In the short transmitter to tuner coax line?

Actually for severely unbalanced dipoles (and yours qualifies for
Queen of the May), you may need a choke at the feed point to the
antenna, and then again a quarter wave away from there.

Thought I read somewhere that only coax can be used for simple 8 to 10
turn chokes. Balanced lines (i believe because of mutual conductor
inductances) can't be coiled as chokes.

Even if I'm wrong, it is both cheap and instructive. So few here
actually step up to the bench that I don't take their flabby word that
I'm wrong. You may be the first with authentic achievement to break a
record! You've already lapped the field of these arm-chair analysts.

Many antennas work just fine until the operator discovers a new tool
that proves it doesn't (in spite of a wall full of QSL cards).

Partly the reason I'm trying to learn all I can about the
configuration I've currently got. That and I like the technology
aspects of the hobby as much or more than I do operating.

Where this hobby whose technological demand largely consists of
pushing a credit card across a sales counter, antennas still have the
capacity to stretch the imagination.

Thanks for helping out a Stuggling Crippled Newbie Street Urchin.

Wait until you face the sewer rats of Rio.

OK I'll bite. Who are the Rio rats?

This is an allusion to an SK who compared those who couldn't exercise
their minds as being fodder for the orphans of Rio, who had more will
to succeed than they did. He characterized them as sewer rats gnawing
on our lazy carcasses.

Thanks for your help. Any thoughts on those measurement results
earlier in the post will sure be appreciated.

They will reveal more in comparison to those measurements that follow.

73's
Richard Clark, KB7QHC

On May 14, 1:07 am, Richard Clark wrote:
On 13 May 2007 16:55:44 -0700, dykesc wrote:

Fixation on BalUns has clouded a simpler solution: wind a choke in
the line and dump the ferrites of suspect quality.

Would you please elaborate on this? Wind a choke where? In the twin
lead?

Sure, twist it candy cane (or barber shop pole) style and wind it
around a liter bottle with at least its width as separation between
windings.


Richard,

I wound the liter bottle choke as you suggested and do believe I'm
getting better results as follows:


1:1 Liter Bottle Choke 4:1 MFJ Balun (tuner
bypassed)

3.94Mhz 57 -j169 14 -j39

7.185Mhz 62 -j212 19 -j48

14.29Mhz 70 -j313 12 -j63

Appears to be a much better correlation to the expected 4:1. Meaning I
can "start" trusting my line input values, though I do believe I can
do better with a properly constructed 1:1 toroidal balun for my input
isolation. Got to order some toroids before I can do this however.

Looks like I'd be better off if I could just lower (zero out) the
capacitive reactance on the twin lead input leaving the resistive
terms alone. Sure wish there was a 1:1 current balun in the tuner. Its
tempting but I don't think I can bring myself to modifying the tuner.
My soldering / desoldering skills aren't the greatest. Its a nice
tuner and I'd hate to muck it up.

I found the thread on testing balun isolation. Still digesting it
though.

Thanks again and 73s
Dykes Cupstid AD5VS

On 14 May 2007 19:43:14 -0700, dykesc wrote:

1:1 Liter Bottle Choke 4:1 MFJ Balun (tuner
bypassed)

3.94Mhz 57 -j169 14 -j39

7.185Mhz 62 -j212 19 -j48

14.29Mhz 70 -j313 12 -j63


Hi Dykes,

This is what I meant about measurements made in isolation not being as
informative without other information.

Oddly enough, it looks like the 4:1 is mounted backwards.

However, I reserve making a final judgement pending review of my OCF
models (your results are just too luckily hitting the ham bands).

73's
Richard Clark, KB7QHC

On May 15, 1:21 am, Richard Clark wrote:
On 14 May 2007 19:43:14 -0700, dykesc wrote:

1:1 Liter Bottle Choke 4:1 MFJ Balun (tuner
bypassed)

3.94Mhz 57 -j169 14 -j39

7.185Mhz 62 -j212 19 -j48

14.29Mhz 70 -j313 12 -j63

Hi Dykes,

This is what I meant about measurements made in isolation not being as
informative without other information.

Oddly enough, it looks like the 4:1 is mounted backwards.

However, I reserve making a final judgement pending review of my OCF
models (your results are just too luckily hitting the ham bands).

73's
Richard Clark, KB7QHC

Richard,

I'm not following your statement that the 4:1 appears to be mounted
backwards. While the 1:1 values are not exactly 4 times the 4:1
values, they are at least "ballpark". Just to make sure you understand
my arrangement, when measured, the 1:1 values should "ideally"
represent the line input impedance without transformation. The 4:1
values (line on the high impedance side of the balun, analyzer on the
low impedance side) should be down by roughly a factor of 4, right?

Also I'm not sure what you mean by my results luckily hitting the ham
bands. I cut and fed the antenna to accomplish that. The OCF feedpoint
(at least by design) is just off the peak of a current loop on 80, 40
and 20 meters (even harmonics).

73s
Dykes AD5VS

On 15 May 2007 20:17:18 -0700, dykesc wrote:

I'm not following your statement that the 4:1 appears to be mounted
backwards. While the 1:1 values are not exactly 4 times the 4:1
values, they are at least "ballpark". Just to make sure you understand
my arrangement, when measured, the 1:1 values should "ideally"
represent the line input impedance without transformation. The 4:1
values (line on the high impedance side of the balun, analyzer on the
low impedance side) should be down by roughly a factor of 4, right?

Hi Dykes,

Look at your data comparison again and ask: Should it be 4:1 or 1:4?

Also I'm not sure what you mean by my results luckily hitting the ham
bands. I cut and fed the antenna to accomplish that.

No doubt, as that is everyone's goal.

The OCF feedpoint
(at least by design) is just off the peak of a current loop on 80, 40
and 20 meters (even harmonics).

I'm not arguing results, I am arguing common sense. Moving the feed
along a dipole does not change its resonances and anti-resonances
(some prefer the terms series and parallel resonances); it changes
their impedances (I will ignore slight reactances). A balanced 80M
dipole resonates on 40M and 20M (and 10M, and 5M....) too. However,
the balanced dipole shows low resonance (series) and high
anti-resonance (parallel) impedances. You are showing consistent low
resonant impedances.

Well, in my review of all of your correspondence, you neglect to tell
us just how far off from center the feed is, and how much slope there
is to the overall wire, and even how long the wire is. I originally
offered that these peculiarities will induce oddities and caught grief
(well, not actually, perhaps it was more like guff) from Tom for
noting it.

However, I have worked out OCDs in the past in response to other's
discussion and there are a world of results and I am going to attempt
to read tea leaves here to intuit the missing details:

The feed is roughly 15% to 20% from the end and there should be almost
as good a match somewhere between the 25M and 30M band. A flat OCD
will show a much poorer match in the 80M band (it will resonate
there), but sloping may introduce enough variation to pull it in
(where the ground is soaking up some of your power as sort of a Z
pad).

At 25% from the end, and there should be no 20M operation as you
describe.

There are possibilities at 35% to 40%.

Let me know about matches inbetween 40M and 20M.

73's
Richard Clark, KB7QHC

On May 16, 12:46 am, Richard Clark wrote:


Look at your data comparison again and ask: Should it be 4:1 or 1:4?


Richard, I guess I'm not seeing something obvious. Its a 4:1 balun
transforming the high side impedance down by a factor of 4. The 4:1
impedance measurements (line on the high impedance side of the balun,
analyzer on the low impedance side) should be down by roughly a factor
of 4, right?
I can't see anything wrong with how I presented the data.

Moving the feed along a dipole does not change its resonances and anti-resonances
(some prefer the terms series and parallel resonances); it changes their impedances

I agree

However, the balanced dipole shows low resonance (series) and high
anti-resonance (parallel) impedances.


By feeding off center my sloping dipole is not balanced.


You are showing consistent low resonant impedances.


The OCF feedpoint (at least by design) is near (but not on) the peak
of a current loop on 80, 40 and 20 meters (even harmonics). Don't
current peaks occur at points of low impedance along an antenna? My
EZNEC current traces confirm this. Shouldn't my even harmonics
impedance measurements all be low rather than high?

By the way Richard the frequencies my impedance measurements were
taken at are not "resonant" frequencies. The resonant frequencies
(zero reactance) occur at 3.56 Mhz, 8.05 Mhz, and 15.7 Mhz. A little
too low on 80M. Too high for 40M and 20M. The tuner gets me the match
in the 40 and 20 amateur bands.

I apologize for not stating that my sloping OCF is fed at 35% of its
total length from the high end. High end at a height of about 40 feet
and low end at a height of about 10 feet. Total antenna length is 136
feet.


73s
Dykes Cupstid AD5VS

On May 16, 12:46 am, Richard Clark wrote:

Look at your data comparison again and ask: Should it be 4:1 or 1:4?

I must be ovelooking something obvious Richard. Measuring through the
4:1 balun with my transmission line on the high side of the balun and
my analyzer on the low side, I obtain the values listed under the 4:1
MFJ Balun column in my earlier post. It doesn't look backward to me.
I'm getting lower impedances just as I would expect. What are you
trying to tell me?


Moving the feed along a dipole does not change its resonances and anti-resonances
(some prefer the terms series and parallel resonances); it changes their impedances

I agree

However, the balanced dipole shows low resonance (series) and high
anti-resonance (parallel) impedances. You are showing consistent low
resonant impedances.

My sloping dipole isn't balanced. Its being fed off center. The OCF
dipole (at least by design) is near (but not on) the peak of a current
loop on 80, 40 and 20 meters (even harmonics). Don't the lowest
impedances occur at current loops? Why are you surpised my impedances
are low rather than high on even harmonics?


Well, in my review of all of your correspondence, you neglect to tell us just how far off from center the feed is, and how much slope there
is to the overall wire, and even how long the wire is.

My apologies. Antenna is 136 feet long. Feedpoint is 35% from the end
that is the highest in the air. Height of high end is about 40 feet.
Height of low end is about 10 feet. Feeding with 300 ohm twin lead
that is 95 feet long (actual length, not electrical length),


Let me know about matches in between 40M and 20M.

The antenna is resonant (zero reactance) at 3.56 Mhz, 8.05 Mhz and
15.7 Mhz. My tuner is required for working in the Amateur bands.

73's
Dykes Cupstid AD5VS

On 16 May 2007 20:21:29 -0700, dykesc wrote:

On May 16, 12:46 am, Richard Clark wrote:


Look at your data comparison again and ask: Should it be 4:1 or 1:4?


Richard, I guess I'm not seeing something obvious. Its a 4:1 balun
transforming the high side impedance down by a factor of 4. The 4:1
impedance measurements (line on the high impedance side of the balun,
analyzer on the low impedance side) should be down by roughly a factor
of 4, right?
I can't see anything wrong with how I presented the data.

Hi Dykes,

Well then, your feed point Zs are inordinately low. (Get rid of the
idea of using the 4:1.)

The OCF feedpoint (at least by design) is near (but not on) the peak
of a current loop on 80, 40 and 20 meters (even harmonics). Don't
current peaks occur at points of low impedance along an antenna? My
EZNEC current traces confirm this. Shouldn't my even harmonics
impedance measurements all be low rather than high?

As OCF/OCD antennas have an infinite variety of feed points (where as
the balanced dipole has only one), there is nothing set in stone about
any of this until you define the degree of "off-center."

There are MANY OCDs that exhibit conventional Hi/Lo/Hi/Lo spectrums in
contrast to your Lo/Lo/Lo.... Certainly, they offer no advantage over
the conventional balanced dipole in this respect. Too often OCD
claims are made without corresponding supporting details.

By the way Richard the frequencies my impedance measurements were
taken at are not "resonant" frequencies. The resonant frequencies
(zero reactance) occur at 3.56 Mhz, 8.05 Mhz, and 15.7 Mhz.

You might note, then, that meaning of harmonics has been similarly
distorted in this thread.

A little
too low on 80M. Too high for 40M and 20M. The tuner gets me the match
in the 40 and 20 amateur bands.

I apologize for not stating that my sloping OCF is fed at 35% of its
total length from the high end. High end at a height of about 40 feet
and low end at a height of about 10 feet. Total antenna length is 136
feet.

Modeling against these criteria reveal nothing like your measurements
with:
(zero reactance) occur at 3.56 Mhz, 8.05 Mhz, and 15.7 Mhz.
(that is the short story)

However, modeling does suggest the antenna should perform for 80/40/20
in much the way your second set of data with the reactance driven out.
To say the least, you still have a hodge-podge of results probably
dominated by the contribution of unbalanced currents on the feed line
(a classic condition for this design that demands considerable
choking).

All that remains to confirm this last observation is to someway
determine the lobe patterns. If they exhibit weak nulls, then the
feed line is part of the radiator. Given the choke appeared to work
at the feedpoint (I assume this is what you did); then a second one is
warranted, and possibly a third. The usual placement advice is 1/4
wavelength from the feed point, but given this is a multi-band
antenna, that is out the window and you will have to experiment (I
would start with 1/4 of the shortest wavelength away).

However, if nulls are immaterial to the enjoyment of the antenna, then
theory has been satisfied and you can chalk that one up.

And to confirm my forecast of operation between 40M and 20M, it should
be exhibiting a significant SWR throughout.

73's
Richard Clark, KB7QHC

Richard Clark wrote:
Look at your data comparison again and ask: Should it be 4:1 or 1:4?

Seems to me the convention for a BA-LUN is to put the balanced
number first followed by the unbalanced number.
--
73, Cecil, w5dxp.com