Wes Stewart said:
I find it curious that Andrew cable is seen so often at these events
and yet I've -never- even seen a piece of Davis cable, dispite the
claims that miles of it are in commercial use.


It might be that the BuryFlex is still in use and has not been removed from
service as the Heliax you cite has been. After all Davis BuryFlex 9914 does
have a warranted - 20 year service life -

Is there any other coax with such a warranty?
---

Charlie

Charlie wrote:

-ps How many times do you think Davis has tested their 9914 in the past
10+ years?

Before we go much further, is there an engineering data sheet on
Bury-Flex? I couldn't find one on the Davis RF site.

In particular, what is the rated minimum bending radius?

Coaxial cables with braid-over-foil shielding have a generic problem
that if they're bent too sharply, the foil will tear into separated
segments each about an inch long. The cable then relies on the braid for
overall shield continuity. It will still function, especially at low
frequencies, but there isn't much contact pressure to maintain the
continuity between the braid and the foil.

If a cable had been treated in this way, it's not hard to imagine that a
precision measurement at 450MHz would reveal small jumps in the loss and
SWR when the cable is flexed.

But this is NOT something you'd ever notice in a normal amateur station
operation. Even when I was using braid-over-foil coax with 1.5kW at
432MHz, and monitoring the SWR continuously, I never noticed any major
jumps when rotating the antenna. The break-up of the foil only came to
light after the cable had failed for an unrelated reason, and was slit
open for a post-mortem.



--
73 from Ian GM3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek

Here are my personal experiences and observations with "BuryFlex".

Back in the 90's, I went through a period of station upgrade,
followed by a period of antenna experimentation, mainly with lower
band receiving antennas. To support all of that work I purchased a few
thousand feet of coaxial transmission line, in a wide range of types
and brands.

For my "main" lines, which needed to be buried to make it
from the house, under the lawn, to the "antenna fields", I chose
BuryFlex. I also used it for my single tower, and around the
rotator. The buried runs are approximately 150 and 175 feet long.

For other purposes, such as test antennas that I would
put up for a few weeks, to arrays of those antennas, I would use
BuryFlex, RG-213, RG-8, RG-8X, RG-58, all sort of stuff. I even
had some 75 Ohm RG-11 which was used mainly as "phasing lines"
in delta loop arrays, and series matching sections.

Before all the dust had settled, I had probably installed
well over 100 UHF coax connectors on cables of all sorts of
lengths, from 1 foot to about 250 feet.

Mechanically, I found BuryFlex to be very rugged, and
whenever I have had the opportunity to check out cables in the
field, including buried, they are in excellent condition. It
does have a high degree of flex, and a small turning radius.
Around the rotator, it worked very well, and after a few
years of turning, it showed no signs of wear at all.

As far as attaching coax connectors, I did find that the
stranded center conductor was a little too wide for a few
of the various connectors I would try to use. I tended to
pick up packs of 10 to 25 male connectors at hamfests and
from the Internet, and they probably covered the range of
name brand, strange brand, gold pins, silver pins, Teflon
insulator, and on and on.

I found that about 10% of the range of connectors
I had would not fit over the twisted BuryFlex center conductor.
Initially, I cut two of the stranded center conductors to reduce
the diameter, but as I accumulated a range of connectors,
I just learned which connectors to avoid when putting ends
on BuryFlex.


As part of my antenna experimentation, I would often
be out in the field, right at the antenna, with a typical
antenna analyzer, such as the MFJ-259, 259B, 269, AEA HF-CIA,
and the Autek RF-1. I would then come back inside
the house, where I had a computer next to the radio. I had
a serial interface for the AEA HF-CIA, so I could capture
SWR graphs on the computer, for a whole range of purposes.


At some point, I picked up a laptop computer, which
made it easy to record antenna analyzer output data right in the
field, next to the antenna, before I entered the BuryFlex
runs back to the house.

An ideal lossless 50 Ohm transmission line should repeat
the SWR found at its load end at its input. The impedance will
be transformed as a function of the length of the cable, but
the SWR should remain the same. Of course we don't have
ideal lines out in the field, we have real lines, with loss.
The impact of the loss is to reduce the SWR at the input (station side)
of the line. In many typical HF situations the reduction is
small.

I noticed that when I overlaid the SWR graphs in the
field on top of the SWR graphs made inside the house, they
did not follow the expected relationship, which is to say the same
general shape, with the inside values slightly lower due
to loss on the cable between the two points.

In some cases, the SWR would rise - I admit, it was
a small amount, but that made no sense to me.


I then performed a test which I must confess I
had not done before. I put a 50 Ohm load on the end of a
length of BuryFlex, and a typical antenna analyzer on the other
end, and swept the frequency across the HF range, perhaps
1 to 30 MHz. I expected to see a flat SWR of 1.0, since
I had a 50 Ohm load, 50 Ohm cable, and a 50 Ohm analyzer.

Well, with BuryFlex, the SWR would swing, as a
function of frequency, between 1.0 and a high of 1.3.

By this time, most of the coax I had once used out
in the field had been rolled up, and stored within
some metal cabinets in the garage. I went and grabbed an
armful of assorted cables, from high quality RG-213, to
pretty darn cheap RG-58 that I used in receiving antenna
arrays. All other 50 Ohm cable, when terminated with
a 50 Ohm load, had a flat sweep across the HF range.

So, something about that BuryFlex was different.
I found all of the lengths that I had rolled up, and
all exhibited the same behavior. I did wonder if I
got a "bad batch", but I do believe that I was measuring
across a set of cables that I had purchased over
a period of perhaps 3 years, and I assumed that meant
that I was looking at different batches and runs and
seasons.

I wasn't sure what was going on, and I wasn't going
to replace that buried BuryFlex, but I stopped using more of it,
especially if I was trying to make useful measurements.
The cable obviously "worked", since I had been pumping
1500 watts through it for year, from 160 meters to 6 meters.


As a few more years passed, and I ended up
interested in higher quality impedance measurements,
as part of understanding and building phased vertical
arrays. I accumulated even more impedance measuring
devices, including old Gen-Rad and Boonton boat anchors.
I did end up with an N2PK VNA, which I believe has
near professional/lab accuracy.

A year or so ago, I got curious about this whole
area again, and now that I had some good quality
measurement equipment, I grabbed some different cables
from the cabinet, and put on a 50 Ohm load, and swept
them from 1 to 50 MHz. Now, what I was really
measuring was the complex reflection coefficient, which
could be converted, by formula, to complex impedance,
SWR, return loss, and all sorts of useful quantities.

When I displayed the data on a Smith Chart, an
interesting pattern emerged. The other 50 Ohm cables spiraled
around the 50 Ohm load. A 75 Ohm cable (with a 50 Ohm load)
also showed a spiral, but it was centered higher up the resistance
axis. The BuryFlex, however, had a spiral which was
centered under 50 Ohms.

Now I'm no Smith Chart expert, especially when it
comes to combining data from different Zo cables, but especially
when comparing BuryFlex to other 50 Ohm cables, to 75 Ohm
RG-11, the pattern is pretty clear.

My conclusion was that the BuryFlex I had here had
a characteristic impedance lower than 50 Ohms, perhaps around 45 Ohms.



After the messages in recent days about BuryFlex,
I went out to the shop this morning, and again grabbed two
different rolls of BuryFlex, a few typical 50 Ohm cables,
and some 75 Ohm RG-11. I wanted to see if I could
duplicate the measurements. All of these cables are a few
years old, and some have been used quite a bit. In other
words, I am not trying to work with new cable right off
of a roll.

I got the same results that I had seen in the
past. I captured two Smith Charts showing BuryFlex versus
other 50 Ohm cable, and some 75 Ohm RG-11. The test
scenario is a random length of cable (more than a few
feet, less than 100), with a 50 Ohm load, and a sweep
from 1 MHz to 50 MHz with 100 KHz spacing. If I were
willing to terminate the BuryFlex with a variable noninductive
resistor, I could probably nail down the actual cable
impedance.

I put the two Smith Charts on a web page so that
they could be inspected by all interested parties.
The URL is:

http://www.seed-solutions.com/gregor...n/buryflex.htm

I certainly can be interpreting this data incorrectly,
and please feel free to correct me if you believe that
I've come to the wrong conclusion.



I do not have an opinion about the "loss" of BuryFlex.
I could make return loss measurements, but for those to
have meaning, I would need to unroll the cable, and measure the
length, and since it's 10 degrees F outside, with 18 inches
of snow, and more coming down, I'm not quite that motivated.
My VNA also stops at about 50 MHz, and the frequency first
mentioned was quite a bit higher (400 MHz). The initial report
also talked about loss as a function of coiling and bending,
which can complicate the test scenario.

For me, and my test equipment here, BuryFlex does
not have the characteristics of any other "50 Ohm" cable
I have measured. I believe that my conclusion that it is more
like 45 Ohm cable as opposed to 50 Ohm cable is correct -
again, for the samples I have here. I have found it to be
mechanically excellent and very rugged. Some coax connectors
don't quite fit over the center conductor.

I am quite suspect, however. I encourage
anybody with access to the cable and appropriate test
equipment to confirm the loss measurements, since I trust
the source of the original claim. Because I don't believe the
50 Ohm specification, I guess it's easier for me to believe that
the loss specification is incorrect too.

If you want to know if your cable is like mine - that's
easy - terminate a length with a good quality 50 Ohm resistive load,
and sweep the frequency while watching the SWR. My BuryFlex
bounces up and down.


Greg Ordy, W8WWV

On Sun, 11 Dec 2005 17:22:16 GMT, "Greg Ordy"
wrote:

Here are my personal experiences and observations with "BuryFlex".

EXCELLENT REPORT!

Thank you,

Very 73
Danny, K6MHE

Danny, K6MHE

email: k6mheatarrldotnet
http://www.k6mhe.com/

On Sun, 11 Dec 2005 14:56:13 GMT, Cecil Moore wrote:

Owen Duffy wrote:
LMR400: 300' at 14.2MHz with VSWR=1.5, loss~=1.5dB

Open-wire line: 300' at 14.2MHz with VSWR=1.5, loss~=0.225dB
Costs about 16 cents/ft if one rolls one's own.

.... or unrolls it as the case may be!
--

On Sun, 11 Dec 2005 17:22:16 GMT, "Greg Ordy"
wrote:

Here are my personal experiences and observations with "BuryFlex".
....

Very well done Greg.

Hopefully, in the fullness of time, you will roll all this information
into the web page.

I have read articles on issues with control of the foaming of
dielectric during manufacture, it seems more difficult to control than
all of the aspects of conventional solid PE dielectric line, so it
leaves one wondering if that may be the cause of the apparently low Zo
(~43 ohms) that you observed. If so, is it a batch to batch variation,
variable along the line (although your charts don't suggest that),
aging, temperature, migration of the centre conductor on the roll,
etc?

Well done OM.

Owen
--

Charlie wrote:

Wes Stewart said:
I find it curious that Andrew cable is seen so often at these events
and yet I've -never- even seen a piece of Davis cable, dispite the
claims that miles of it are in commercial use.

It might be that the BuryFlex is still in use and has not been removed from
service as the Heliax you cite has been. After all Davis BuryFlex 9914 does
have a warranted - 20 year service life -

Is there any other coax with such a warranty?
---

Charlie

I don't seem to be able to find any mention of a warranty on their site - can
you point it out for me?

Dave

At the bottom of this page the ** footnote
1. http://www.davisrf.com/ham1/coax.htm

"note that Bury-Flex has a 20+ year abrasive resistant jacket of PE"


--

Charlie


"Dave Holford" wrote in message
...


Charlie wrote:

Wes Stewart said:
I find it curious that Andrew cable is seen so often at these events
and yet I've -never- even seen a piece of Davis cable, dispite the
claims that miles of it are in commercial use.

It might be that the BuryFlex is still in use and has not been removed
from
service as the Heliax you cite has been. After all Davis BuryFlex 9914
does
have a warranted - 20 year service life -

Is there any other coax with such a warranty?
---

Charlie

I don't seem to be able to find any mention of a warranty on their site -
can
you point it out for me?

Dave

On Sat, 10 Dec 2005 17:15:44 -0800, Roy Lewallen
wrote:

Charlie wrote:
I have emailed Davis RF and asked them to send me their test results. Yes
it will be interesting.

Roy..doesn't seem a bit odd that NO ONE ELSE in over 10 years of BuryFlex
production has cited these same alarming "test results".

No. My experience is that the vast majority of amateurs don't have the
ability and/or confidence and/or interest and/or equipment to make good
measurements. And lacking the ability to measure it, very few would be
able to discern the difference in loss. If my measurements are typical,
commercial and government users (if there indeed are any for this
particular cable type) would quietly reject the stuff on incoming
inspection and order something else.

I'd ten to agree and most take cable on faith as well.

I do have a high degree of confidence that my measurements are accurate.
I took a lot of care in characterizing the cable before using it for
making remote antenna impedance measurements in the course of a
consulting job. But it's entirely possible that the particular piece of
cable I have is defective. That would just point to a quality control
problem rather than overzealous specsmanship.

There are a number of things that can or possibly happen.

When you test coax connectors and termination effects are part
of the system and need to be considered.

I bought 250ft, then tested the 50ft sections I made up for the tower.
All five tested the same at 440mhz and my confidence factor was
good. The test setup was 0.100 Watt RF source, 3DB attenuator,
RLB, cable section, termaline watmeter. All check on spec
and I was doing to verify connectors installed as well. I only tried
one section for bening effects near the end as I was interested in
how it would behave for rotor loops. I saw now ill efects until I
reduced the bend radius to under 4 inches where it developed
a definitate bump that showed on a TDR. Like most foam cored
cables you can overbend it with bad results. The damaged length
was removed (only 4ft) and the cable placed in service.

Expected 2.9db @ 400mhz. But that does not allow for connectors
or measurment setup. Half that should be 1.45db (50ft).

However my testing was at 440mhz. All loss testing I did was
at 440 because small things look bigger there.

To the limits of the attenuators and meter calibration I'd call the
losses including the connectors right where I'd expect 50ft with
PL259s on at least one end would be. I got at 440, greater
than 1.6db but less than 1.8db. The loss was determined by
removing cable and substituting a known attenuators of .2db
steps as that was as fine as I had. Calculated was 1.66db
based on measurements. Allowing for test error and
connectors the there is some range of error. The RLB was
used to verify there were no visible bumps over the 6m, 2m and
420-450mhz range of the source.

The same setup for 50ft or both RG58A/U, RG8X and RG213
gave me 7db(pl259/ug174), 4.6-4.8(pl259/ug175), 2.4-2.6(pl259).
The RG213 was a suprize as I'd expected losses to be lower.
But the results were consistant. Where I give a range in loss
it is because the lower and higher attenuateors used were
either too low or too high meaning the exact result was somewhere
in between. The loss substitution is likely more accurate than
analog meters. Also I tried to minimize the error from adaptors
and cable transitions where needed though te test method tends
to zero them out by substiution.

I did during installation notice one thing over RG213. The 2M antenna
I'd tuned with a length of RG213 showed a higher VSWR when using the
Buryflex. Further tests reveled the losses of the 213 made the SWR
look better than it was. Antenna was readjusted and all was happy.

Allison
Kb1GMX

On Sun, 11 Dec 2005 17:22:16 GMT, "Greg Ordy"
wrote:

[snip]

Good job Greg. Also thanks for the work on the N2PK VNA software.

By inspection of the loss figures for Buryflex on Davis' web site one
can note that the loss at lower frequencies is higher than other RG-8
type cables.

Importing their data into Dan's (AC6LA) "bestfit.xls" spreadsheet
shows very poor correlation with theoretical k1, k2 coefficients. I'm
observing something similar on some RG-142 that I am measuring with my
N2PK VNA. The '142 has a silver-coated, copper-plated steel center
conductor and I believe that the skin depth encompasses the steel at
lower frequencies.

Anyway, that might be a factor, assuming of course, that their data
are valid.