Wire Lengths
In your post, the comments about short grounding cables have much
merit.
However, the use of TUNED lengths of cables between equipment appears to
be VERY IMPORTANT. Dr Bare specifies 1/2 wave or 18ft cables in his
construction plans.
It appears that frequency tuned lengths of cables are required to avoid
suppresion of harmonics of the fundemental frequencies which essential
to plasma effect . This process is not well understood at this time.
In my pesonal experience, my equipment operated "electronicaly" fine
using short cables, but I found that I did not "feel" the ray beam
effect, EXCEPT WHEN USING TUNED CABLES. I have used full wave, 3/4
wave and half wave connecting cables with success.
It is my opinion that it also may be effective to use smaller fractions,
(such as 1/4, 1/8, 1/16, 1/32, 1/64 th wave) if properly tuned, however
I have not tried this yet. (and tuning becomes much more critical as
you approach a smaller fraction of full wave). Please refer to my
previous posts on cable lengths and math involved. Also errors in
calculating the velocity factor in the wire become significant.
In summary I strongly recomend that until further data is obtained, that
1/2 wave or 18ft cables be used to connect cb to linear and linear to
the antenna tuning device. These are commonly availible from cb radio
sources.
---------------------------
I had it backwards when I stated the length to be increased vs
decreased. I have been using way-off lengths to match wavelength since
I have been using 95%, which is incorrect since looking in catalogs
informs that the propagation velocity is usually around 70%, not
90-95%
For a 70% cable, then, the following chart would be relevant
(rounded to the nearest half inch.)
27.125 MHz
Ft (0.70)xWL (ft) ft, in
1/1 36.276 22.59 22, 7
1/2 18.13 12.69 12, 8
1/4 9.07 6.35 6, 4
1/8 4.53 3.17 3, 2
1/16 2.27 1.59 1, 7
1/32 1.134 0.79 0, 9.5
---------------------------
From: Domenic Spinale
Subject: Re: Propagation Velocity
A while back there was some talk on the List about the use of exact
sub-multiples of 1/4 wavelength transmission lines; and back then I did not
have time to get in my two cents worth.
When a lossless transmission line is terminated in its characteristic
impedance, the voltage measured at all points along the line will be the
same and the line is said to be flat which means the SWR is 1:1. The
opposite extreme is when the transmission line is either open or short
circuited which results in 100% of the signal that went down the line to be
reflected back up the line. The 27 MHz signal from the B/R system is
essentially a sine wave and the reflected signal is also a sine wave. With
two signals going in opposite directions on the same transmission line at
the same time, they algebraically add. The summation of the two waves is
called a standing wave. The highest voltage that can occur is at the point
where the peaks of the two waves are both at maximum with the same polarity
which results in twice the signal voltage. The other extreme is the point
where the two waves are exactly opposite in both polarity and amplitude and
they add to zero. The SWR is the maximum voltage point divided by the
minimum voltage point. When the two waves are of equal amplitude, the SWR
is 2 divided by zero which is infinity. These points of maximum voltage and
minimum voltage repeat every half wave length. A wave length can also be
expressed in degrees. A full wave length would be 360 degrees, similarly a
half wave length would be 180 degrees, and 1/4 wave length would be 90
degrees. Now recalling the sine function from trigonometry, it is zero at
zero degrees and increases in value to its maximum positive value of one at
90 degrees where it reverses direction then passes through zero at 180
degrees, continues in a negative direction to its maximum negative value of
-1 at 270 degrees where it changes in a posit ave direction, and passes
through zero at 360 degrees which is the same as zero degrees. The purpose
of this explanation is to show that the sine wave has a value of zero every
180 degrees (at 0 and 180 degrees) which is every half wave length. Maximum
amplitude points occur at 90 and 270 degrees, 180 degrees apart, also a
half wave length, but displaced from the zero points by 90 degrees or 1/4
wave length.
In conclusion it can be shown that a unique point (a zero or a maximum)
occurs every 90 degrees or 1/4 wave length.
If anyone would like to explain what is critical about 1/8, 1/16, or 1/32
of a wave length, I am all ears.
Domenic
===
===
>> In conclusion it can be shown that a unique point (a zero or a maximum)
>> occurs every 90 degrees or 1/4 wave length.
>>
>> If anyone would like to explain what is critical about 1/8, 1/16, or 1/32
>> of a wave length, I am all ears.
---------------------------------------------------------------------------
>Please explain the implications of this. Does it mean that the best
>size RF cables in this system are (ignoring prop velocity) are 9 ft and
>27 ft (approx 1/4 and 1/4+1/2 wavelengths)?
>
>If so, is there no ideal size if the cables are significantly less than
>these values? If there is no ideal size, then would cables in the,
>say, 1-6 ft range be "shorter the better" (least antenna action)
>or "longer the better (since they would be closer to 9')"?
---------------------------------------------------------------------------
I've been pondering the answer to the above question for a very long time
because sometimes things are not what they appear to be. I currently
believe that the coax cable length is not critical and shorter is better.
When I initially read that the recommended cable length between the CB and
the linear amplifier, as well as between the linear amplifier and the
antenna tuner was 18 feet, I thought: That's a half wave length - except
someone forgot to correct for the velocity factor of the cable. The 18 foot
length of coax is not a half wave length in the signal path, so scratch
that idea.
Another time I thought: Due to the load, the tube and its coupling back to
the tuner, being unbalanced resulted in a ground current flowing through
all of the ground connections; and since there is a voltage null every half
wave length and the velocity factor being closer to one for current along
the braid of the coax, that explains the recommended 18 foot lengths. This
scheme would hold the case of CB, linear amplifier, and antenna tuner close
to the same RF voltage potential. Then the recommendation is to coil the
coaxial cables. This adds series inductance in the ground current path and
defeats the 1/2 wave length grounding scheme. To correct this poor
grounding system, the recommendation is to connect all the equipment
together with short braided wire.
Thus far, I cannot find a good reason to support the idea of using 18 foot
coaxial cables. Provided that the SWR's are reasonably low there is no
magic length of coaxial cable. If the coaxial cables are short, the braided
wire ground connections should not be needed.
With respect to any imbalance in load current resulting in ground current,
it would make sense to null out the ground current at the balun with some
type of gimmick capacitor between the ground connection on the balun and
one of the balanced output wires. Nulling out ground current can solve
three problems. First, minimize RF interference. Second, minimize any
instability in the function generator caused by RF ground current. And
third, minimize any effect on the SWR when touching any of the equipment.
Other opinions are welcome.
Domenic
===
Subject: Re: Propagation Velocity
> I currently
> believe that the coax cable length is not critical and shorter is better.
That is what I have found, empirically, as far as the cable length
itself is concerned. However, using different sized cables can
seemingly affect SWR and power since one size may remove or cause
interference due to proximity with other equipment in the setup. In
the system I am building now shorter is always better so far (this is
where the tuner and balun sit on a shelf above the amp and CB.) In my
other system, where the balun, amp, CB, and tuner are all on the same
shelf, and the amp is 2" from the balun, using a long enough amp to tuner
coax to go around the balun rather than under or over it results in
better SWR and power.
> Thus far, I cannot find a good reason to support the idea of using 18 foot
> coaxial cables.
Bare states in the manual that it helps the tube light easier. It does
not seem to help in my system.
> If the coaxial cables are short, the braided
> wire ground connections should not be needed.
I have also found this as a result of experimentation. I do have one
ground (7") between the CB and freq generator since it
helps keep 120Hz focused for some reason, but grounding all the
equipment together makes it slightly worse.
=====================
I'm using a cheb type tube (He + phanatron type electrodes) and found that
12 gauge rope laid monster speaker wire in thick vinyl jacket of 26" + spade
type connectors works best for me....
The spade connectors are 10-12 ga. crimped on the ends of the wires, and
then silver soldered.... The identical setup w/ flat 1/4" tinned grounding
strap has SWR's of at least .25 to .5 higher (at differant frequencies...
With the spade connectors, the whole cable(s) tip to tip is about 27"
long....
On the cheb tube, there are screw connectors on the ends of the tube that go
to the internal electrodes...
I have not tried an argon tube yet but am saving my pennies for a quartz
bubble tube w/ Argon ir a mix as the Helium seems to work too good...
By that I mean I seem to have created a major kill off of whatever is in me
and was quite toxic for several days including a SVT reaction... and a
blowup (swelling and opening) of a lymph node in my groin.... I wasn't even
supposed to be the patient.... <grin>
It has spooked me however, and the person I was going to use it for, I won't
because of side effects for the moment.... anyways.... I don't want to
cause more pain where there is plenty to begin with.... Need to figure a way
to reduce the tumor in situ w/o more pain....
Hope this helps
Peter
GFoye wrote:
> I am not electronics oriented so this may not be good advice but:
>
> There are two thoughts on tube wire length.
>
> * One is: short as possible.
>
> * Other: wire length to fractional wave lengths.
> Therefore, for 1/4 wave, wire should be about 26" to 30" inches.
> Mine seems to work best about 26 to 28 inches - both equal length.
>
> Any technical input here?
>
> GF
====================
>[snip] I am using 14.5 feet of 9913F coax ( this is the
correct length, not 18 Feet as described by Jim.). The equation
for calculating ½ wave length must take into account the
velocity factor of the coax which can range for 0.66 to 0.84 (
0.80 in my case).
Rick and List:
It would appear that the above comments are based upon the
assumption that the signal path of interest is the RF power out
of the CB and going through the coax to the linear amplifier,
and/or the RF power coming out of the linear and going through
the coax to the tuner. Suppose the signal current of interest
was the result of the unbalanced current on the lines out of
the balun, and that current flows along the ground path. If
the objective was to hold the CB, the linear, and the tuner at
the same RF potential then 1/2 wave length connections between
them would make sense. When the current path is along the
outer conductor of the coax rather than the center conductor
with the outer conductor being the return path, the velocity
factor as expressed above does not apply.
Domenic
======================
I ran a quick, basic test on length of wire from balun to
tube. I believe a quarter wave would be about 28 1/2 "
length. (Don't know formula - got the figure from someone
else.) Just wanted to see what would take place with
varying wire length.
For this test I used the following: #10 solid wire,
attached to balun terminal with copper post connector
CP-4-2B Adamax Inc 495-875 purchased at Home depot. This
terminal made it very easy to remove wire, snip off a piece
and reinsert wire- tighten with a set screw.
Equipment:
Uniden 510XL modified per manual
palomar 225
MFJ 949E tuner with external balun connected with 24" coax
Lodestar AG2603AD
argon bubble tube
#10 wire: (two pieces)- made a coil by forming over broom
handle, slip fit for my tube at large diameter. Just insert
tube into the coils -must be sure coils are aligned properly
so there is no stress on tube. Started with 30" overall.
(Before coil, etc.). Snipped off 1/2" at a time.
(Will have to do this one more time to pin point exact
length for my set up. Think I passed it around 29 ". Also,
my tube started acting up, may have damaged it.) What I
found was: the optimum is going to be close to the
calculated figure of 28 1/2"
As length was shortened the SWR's climbed but not
significantly although at a couple points the SWR's did make
a huge jumps. (This is one reason the test would have to be
done about three times to see if this condition would
repeat.)
The significant change was in wattage reduction as length
was decreased.
At optimum (palomar on med) the wattage output was 200 with
reflected close to one.
As wire length was reduced wattage output dropped, lowest
about 150 and reflected about four. (Except at two
intervals where there were significant jumps: SWR 3:1,
watts 110.)
At a couple points I had to change from inductor L to K.
But, not sure what was happening there- would have to do
this again for repeatablity.
I went down to 26" and saw the wattage dropping steadily.
At that point I snipped off a 6" segment down to 20".
Wattage still low but not much different than at 26".
Results: in order to obtain maximum wattage output with
lowest reflected SWR, wire length is important. (So, for
those that build the compact units, how do you get around
this? Somehow you have to fool the system or do you use
long wire?)
I'm not a person with patience so I went through this
rather crudely. But, am going to do it again, starting with
30" and snip off 1/4" at a time and record results then
utilize the optimum figure.
Note: I believe the coils work nicely and could be used
similarly as the copper sleeves. Make a coil and solder
something else to it if you prefer.
GF
=================
Below is a link to a web site that has some interesting information on
coax cables.
http://members.tripod.com/~cb_antennas/coax_basics.html
Regards, Jason
=================
A note to the group about velocity factors. The output of an R/B is quite
wide band and is not merely fixed at 27.125 MHz. 27.125 Mhz is merely the
center frequency. The primary Lower Side Band (LSB) has harmonics that
extend about 3 MHz below the center frequency.The primary Upper side band
(USB) has harmonics that extend about 3 Mhz above the center frequency .
Most of these harmonics are 40 to 50 Db down, but they still do exist.
Calculation, and then fabrication of cable length based upon velocity
factor and a fixed 27.125 Mhz will result in poor SWR's. There are other
harmonic signals that are generated that extend out to a good 300 to 500
MHz. Again these are very weak. Some harmonics can be picked up below 10
MHz. An optimum coax cable should be able to contain all these harmonics.
The recommended 18' length of coax between the transmitter and the tuner
works just fine. Someone might try and have a 21' length made up to see
what happens to SWR. Might better contain the LSB frequencies.
Jim Bare
============== |