Wednesday, November 22, 2023

A simple VHF notch cavity from scraps of (large) Heliax

In a previous post I discussed how a band-pass "cavity" could be constructed from a chunk of 1-5/8" Heliax (tm) cable (a link to that article is here).  This is the follow-up to that article.

Figure 1:
The dual notch filter assembly - installed at the
repeater.
Click on the image for a larger version.

Notch versus band-pass

As the name implies, a "notch" filter removes only a specific frequency, ideally leaving all others unaffected while a "band pass" filter does the opposite - it passes only a specific frequency.  Being the real world, neither type of filter is perfect - which is to say that the "width" of the effect of the notch or pass response is not infinitely narrow, nor is it perfectly inert at frequencies other than where it is supposed to work:  The notch filter will have some effect away from its frequency of rejection, and a band pass filter will let through off-frequency energy and both will have loss even where it would not be ideal.  These filters may be constructed many ways - from individual coils and capacitors to resonant structures, such as cavities - which are often larger-diameter tubes with smaller tubes inside, the latter being resonant at the frequency of interest.  The cavity-type of filters often have better performance as their operation is closer to that of ideal (perfect) components.

The degree to which a filters is imperfect is significantly determined by the "Q" (quality factor) of the resonating components and in general for a cavity-based device, the bigger the cavity (diameter of conductors and the container surrounding it) the better the performance will be in terms of efficacy - which is "narrowness" in the case of the notch filter and "width" and loss in the case of the band-pass cavity.

While a cavity-based device with a large inside resonator and larger outside container is preferred, one can use pieces of large coaxial cable, instead.  The use of large-ish coaxial cable as compared to smaller cable (like RG-8 or similar) is preferred as it will be "better" at everything that is important - but even a cavity constructed from 1-5/8" coax will be significantly inferior to that of a relatively small 4" (10cm) diameter commercial cavity - but there are many instance where "good" is "good enough.

Case study:  Removal of APRS/packet transmitter energy from a repeater input

As noted in the article about the band-pass cavities linked above, a typical repeater duplexer - even though it may have the words "band" and "pass" on the label and in the literature - RARELY have an actual, true "band-pass" response.  In other words, a true "bandpass" cavity/duplexer would have 10s of dB of attenuation, say, 20 MHz away from its tuned frequency - but most duplexers found on amateur repeaters will actually be down only 6-10 dB or so, meaning that even very far off-frequency signals (FM broadcast, services around 150-174 MHz, TV transmitters) will hit the receiver nearly unimpeded.  When I tell some repeater owners of this fact, I'm often met with skepticism ("The label says 'band-pass'!") but these days - with inexpensive NanoVNAs available for well under $100, they can check it for themselves - and likely be disappointed.

Many clubs have replaced their old Motorola, GE or RCA repeaters from the 70s and 80s with more modern amateur repeaters (I'm thinking of those made by Yaesu and Icom) and found that they were suddenly plagued with overload and IMD (intermod).  The reason for this is simple:  The old gear typically had rather tight helical resonator front-end filters while the modern gear is essentially a modified mobile rig - with a "wideband" receiver - in a box.  In this case, the only real "fix" would be the installation of band-pass cavities on the receive and transmit paths in addition to the existing duplexer.

In the case of APRS sharing a radio site, the problem is different:  Both are in the amateur band and it may be that even a "proper" pass cavity may not be enough to adequately reject the energy if the two frequencies are close to each other.  In this case, the scenario was about as good as it could be:  The repeater input was at 147.82 MHz - almost as far away as it could be from the 144.39 APRS frequency and still be in the amateur band.

What made this situation a bit more complicated was the fact that there was also a packet digipeater on 145.01 MHz - a bit closer to the repeater input,  but since it was about 600 kHz away from the 144.39 APRS frequency, that meant that just one notch wouldn't be quite enough to do the job:  We would need TWO.

Is it the receiver or transmitter?

Atop this was another issue:  Was it our receiver that was being desensed (overloaded) by these packet transmitters, or was it that these packet transmitters were generating broadband noise across the 2 meter band, effectively desensing the repeater's receiver?

We knew that the operators of the packet stations did not have any filtering on their own gear (the only way to address a transmit noise problem if generated by their gear) and were reluctant to spend the time, effort and money to install it unless they had compelling reason to do so.  Rather than just sit at a stalemate, we decided to do due diligence and install notch filtering on the receiver to answer this question - and give the operators of the packet gear a compelling reason to take action if it turned out that their transmitters were the culprit.

A simple notch cavity:

Suitable pass cavities are readily available for purchase new from a number of suppliers and used from auction sites - they are also pretty easy to make from copper and aluminum tubing - if you have the tools.  Because of the rather broad nature of a typical pass or lower-performance (e.g. broader) notch cavity, temperature stability is usually not much of an issue in that its peak could drift a hundred kHz and only affect the desired signal by a fraction of a dB.

As mentioned earlier, another material that could be used to make reasonable-performance pass cavities is larger-diameter hardline or "Heliax" (tm).  Ideally, something on the order of 1-5/8" or larger would be used owing to its relative stiffness and unloaded "Q" and either air or foam dielectric cable may be used, the main difference being that the "Q" of the foam cable will be slightly lower and the cavity itself will be somewhat shorter due to the different velocity factor.

Figure 2:
Cutting the (air core) cable to length
Refer to the calculator on the KF6YB web page, linked
at the end of this article.
Click on the image for a larger version.

The "Heliax notch cavity" described here can be built with simple hand tools, and it uses a NanoVNA for tuning and final adjustment.   While its performance will not be as good as a larger cavity, it will - in many cases - be enough to attenuate signals that are "far enough" away for the somewhat limited "Q" of a notch filter of this construction to be effective without excessively attenuating the desired frequency.

Using 1-5/8" "Heliax":

Note:  For an online calculator to help determine the length of cable to use, see the link to KF6YB's site at the end of this article.

The "cavity" described uses 1-5/8" air-core "Heliax" - and it is necessary for the inner conductor to be hollow to accommodate the coupling capacitors.  Most - but not all - cable of this size and larger has a hollow center conductor.  Cables larger diameter than 1-5/8" should work fine - and are preferred - but smaller than this may not or may note be practical in situations where the notch and desired frequency are closely spaced - this, for reasons of unloaded "Q".  If the center conductor is solid or if its inside diameter cannot accommodate the coupling capacitors (described later on) you will have to improvise their construction, using either a discrete variable capacitor or a small "sleeve" capacitor - external to the piece of cable similar to the coupling capacitors described below.

Preparing the "shorted" end:

For 2 meters, a piece of cable 18" long was cut.  For cables with an air dielectric, it's recommended that one cuts it gently with a hand saw rather than a power tool as the latter can "snag" and damage the center conductor.

Figure 3:
The "shorted" end of the stub with the slits bent to the middle
and soldered to the center conductor.
This end should be covered with electrical tape and/or
RTV/silicone to keep out insects/dirt.
Click on the image for a larger version.

For the "cold" (e.g. shorted) end, carefully (using leather gloves) remove about 3/4" (19mm) of the outer jacket and then clean the exposed copper shield with a wire brush, abrasive pad and/or sand paper.  With this done, use a pair of tin snips cut slots about 1/2" (12mm) deep and 1/4" (6mm) wide around the perimeter.  Once this is done, use a pair of needle nose pliers and remove every other tab, resulting is a "castellated" series of slots.  At this point, using a pair of diagonal pliers or a knife, cut away some of the inner plastic dielectric so that it is about 1/2" (12mm) away from the end of the center conductor.

Now, clean the center conductor so that it is nice and shiny and then bend the tabs that were cut inwards so that they touch the center conductor.  Using a powerful soldering iron (I used a 150 watter) or soldering gun - and, perhaps a bit of flux - solder the shield tabs to the center conductor all of the way around.  It's best to do this with the section of coax laying on its side so that hot solder/metal pieces do not end up inside the coax - particularly if air-core cable is used.  If you used acid-core flux, carefully remove it before proceeding.

With one end of the cable shorted you can trim back any protruding center conductor and file any sharp edges - again taking care to avoid getting bits of metal inside the air-space of the cable or embedded in the foam.  At some point, you should cover the shorted end with RTV (silicone) and/or good-quality electrical tape to prevent contamination by dust or insects.

Preparing the "business" end:

Figure 4:
This shows how the tube for the coupling capacitor is placed.
This photo is from the band-pass version with two tubes.
Click on the image for a larger version.
At this point, the chunk of coax should be trimmed again, measuring from the point where the center conductor is soldered to the shield:  For air-core trim it to 17" (432mm) exactly and for foam core, trim it to 16-1/8" (410mm).  Again, using a sharp knife and gloves, remove about 3/4" (19mm) of the outer jacket and, again, clean the outer conductor so that it is bright and shiny.

Making coupling capacitors:

We now need to make a capacitor to couple the energy from the coaxial cable to the center resonator and for this, we could use either a commercially-made variable capacitor (an air-type up to about 20pF - but much less will likely be required) or we could make our own capacitors:  I chose the latter.

At this point you may be asking yourself, "Self, if I make a coax stub for HF, I connect it directly to a coax Tee - why don't I do that here?"  While you could connect a 1/4" stub directly across the coaxial cable to effect attenuation, this is only practical for notches that are a significant distance away from the desired frequency.  For example, if you find yourself in the situation mentioned above (e.g. you replace your ancient repeater with a modern one with poor front-end filtering) where a nearby FM broadcast transmitter is overloading your receiver, you could reasonably measure/cut an open 1/4 wave stub for its frequency and put it across the coax feed with a "Tee" connector and reduce its energy by 20dB or so.  The problem is that a direct connection like this will have rather poor "Q" and be very wide - possibly suitable for a signal 10s of MHz away, but it won't help you if the signal is just a couple MHz away.

By "lightly" coupling to the resonator with a reactance - typically a capacitor in the 10s of pF range or lower - the "Q" of the resonating element is somewhat preserved and with "critical" coupling (not too much, not too little) one can achieve narrower, deeper notches.

Using RG-8 center for the coupling capacitor

For this, I cut a 3" (100mm) length of solid dielectric RG-8 coax, pulled out the center conductor and dielectric and threw the rest away.  I then fished around in my box of hardware and found a piece of hobby brass tubing into which the center of the RG-8 fit snugly cut to the same length as the center conductor.  If you wish, you can foam dielectric RG-8 center but be aware that it is more fragile - particularly when soldering.

I then soldered to tubing inside the center conductor/resonator as doing so offers good mechanical stability, preventing the piece of coax cable dielectric from moving around and changing its capacitance.

Using RG-6 center for the coupling capacitor:

While RG-8 and brass tubing is nice to use, I have also built these using the center of inexpensive RG-6 foam type "TV" coaxial cable and a small piece of soft copper water tubing that I had laying around - but it can easily be found at a hardware store.  This type of capacitor is fine for receive-only applications, but it is not recommended for more than a few watts:  The aforementioned RG-8 capacitor is better for that.

For this, I cut a 3" (75mm) long piece of RG-6 foam TV coaxial cable and from it, I removed and kept the center conductor and dielectric - removing any foil shield and then stripping about 1/2" (12mm) of foam from one end of each piece.

At this point, you'll need some small copper tubing:  I used some 1/4" O.D. soft-drawn "refrigerator" tubing, cutting a 2" (50mm) length and carefully straightening it out.  To cut this, I used a rotary pipe cutting tool which slightly swedged the ends - but this worked to advantage:  As necessary, I opened up the end cut with the deburring blade of the rotary cutting tool just enough that it allowed the inner dielectric of the RG-6 to slide in and out with a bit of friction to hold it in place.

Figure 5:
The PC Board plate soldered to the end of the coax.  This
is from the band-pass version, but you get the idea!
Click on the image for a larger version.

No matter which type of coax center you are using, using a hot soldering iron or gun, solder the tube for the coupling capacitor inside the Heliax's center conductor, the end flush with the end of the center conductor:  A pair of sharp needle-nose pliers to hold it in place is helpful in this task.  Remember that you are soldering to a large chunk of copper, so you'll need a fair bit of heat to be able to make a proper connection!

Making a box:

On the "business" (non-shorted) end of the piece of cable we need to make a simple box with a solid electrical connection to the outer shield to which we can mount the RF connectors with good mechanical stability.  For the 1-5/8" cable, I cut a piece of 0.062" (1.58mm) thick double sided glass-epoxy circuit board material into a square that was 3" (75mm) square and using a ruler, drew lines on it from the opposite corners to form an "X" to find the center.

Using a drill press, I used a 1-3/4" (45mm) hole saw to cut a hole in the middle of this piece of circuit board material, using a sharp utility knife to de-burr the edges and to enlarge it slightly so that it would snugly fit over the outside of the cable shield:  You will want to carefully pick the size of hole saw to fit the cable that you use - and it's best that it be slightly undersized and enlarged with a blade or file than oversized and loose.

Figure 6:
Bottom side of the solder plate showing the
connection to the coax.
Click on the image for a larger version.

After cleaning the outside of the coaxial cable and both sides of the circuit board material, solder it to the (non-shorted) end on both sides of the board, almost flush with just enough of the shield protruding through the top to solder it.  For this, a bit of flux is recommended, using a high-power soldering iron or gun - and it's suggested that it first be "tacked" into place with small solder joints to make sure that it is positioned properly.

Adding sides and connectors:

With the base of the box in place, cut four sides, each being 1-3/8" (40mm) wide and two of them being 3" (75mm) long and the other two being 2-1/2" (64mm) long.  First, solder the two long pieces to the top, using the shorter pieces inside to space and center them - and then solder the shorter pieces, forming a five-sided (base plus four sides) box atop the piece of cable.

Figure 7:
A look inside the box showing the connection to the center of
the capacitor, the "tuning" strips and ceramic trimmer.
Click on the image for a larger version.
Resonator adjustment capacitor:

You will need to be able to make slight adjustments to the frequency of the center conductor of the Heliax resonator.  If all goes well, you will have cut the coaxial cable to be slightly short - meaning that it will resonate entirely above the 2 meter band.  The installation of the coupling capacitor will lower that frequency significantly - but it should still be above the frequency of interest so a means for "fine tuning" is necessary.

Figure 7 shows two strips of copper:  One soldered to the center conductor (the sleeve of the coupling capacitor, actually) and another soldered to the inside for the Heliax shield.  These to plates are then moved closer/farther away to effect fine-tuning:  Closer = lower frequency, farther = higher frequency.  Depending on how far you need to lower the frequency, you can make these "plates" larger or smaller - or if you can't quite get low enough in frequency with just one set of these "plates", you can install another set.  

NOTE:  It is recommended that you do NOT install the copper strips for tuning just yet:  Go through the steps below before doing so.

If your resonant frequency is too low - don't despair yet:  It's very likely that you'll have to reduce the coupling capacitor a bit (e.g. pull it out of the tubing and/or cut it a bit shorter) and this will raise the frequency as well.

How it's connected:

A single notch cavity is typically connected on a signal path using a "Tee" connector as can be seen in Figure 1:  At the notch's resonant frequency, the signal is literally "shorted out", causing attenuation.  

As can be seen in Figure 7, there is only one connector (BNC type) on our PC board box - but we could have easily installed two BNC connectors - in which case we would run a wire from one connector to the center capacitor as shown and then run another wire from the capacitor to the other connector.

Adjusting it all:

For this, I am presuming that you have a NanoVNA or similar piece of equipment:  Even the cheapest NanoVNA - calibrated according to the instructions - will be more than adequate in allowing proper adjustment and measurement of this device.

Using two cables and whatever adapters you need to get it done, put a "Tee" connector on the notch filter and connect Channel 0 on one side of the Tee and Channel 1 on the other side of the Tee and put your VNA in "through" mode.  (Comment:  There are many, many web pages and videos on how to use the NanoVNA, so I won't go through the exact procedure here.)

Configure the VNA to sweep from 10 MHz below to 10 MHz above the desired frequency and you should see the notch - hopefully near the intended frequency:  If you don't see the notch, expand the sweep farther and if you still don't see the notch, re-check connections and your construction.

At this point, "zoom in" on the notch so that you are sweeping, say, from 2 MHz below to 2 MHz above and carefully note the width and depth of the notch.  Now, pull out the center capacitor (the one made from the guts of RG-8 or RG-6 cable) a slight amount:  The resonant frequency will move UP when you do this.

The idea here is to reduce the coupling capacitance to the point where it is optimal:  If you started out with too much capacitance in the first place, the depth of the notch will be somewhat poor (20dB or so) and it will be wider than desirable.  As the capacitance is reduced, it should get both narrower and deeper.  At some point - if the coupling capacitance is reduced too much - the notch will no longer get narrower, but the depth will start to get shallower. 

Comment:  You may need to "zoom in" with the VNA (e.g. narrow the sweep) to properly measure the depth of the notch.  As the VNA samples only so many points, it may "miss" the true shape and depth of the notch as it gets narrower and narrower.

The "trick" with this step is to pull a bit of the coax center out of the coupling capacitor and check the measurement.  If you need to pull "too much" out (e.g. there's a loop forming where you have excess) then simply unsolder the piece, trim it by 1/4-1/2" (0.5-1cm), reinstall, and then continue on until you find the optimal coupling.

It's recommended that when you do approach the optimal coupling, be sure that you have a little bit of adjustment room - being able to push in/pull out a bit of the capacitor for subsequent fine tuning.

At this point your resonant (notch) frequency will hopefully be right at or higher than your target frequency:  If it is too low, you may need to figure out how to shorten the resonator a bit - something that is rather difficult to do.  If you already added the "capacitor plates" for fine-tuning as mentioned above, you may need to adjust them to reduce the capacitance between the ground and the center conductor and/or reduce their size.

Presuming that the frequency is too high (which is the desirable state) then you will probably need to add the copper capacitor strip plates as describe above, and seen in Figure 7.  You should be able to move the resonant frequency down toward your target by moving the plates together.  Remember:  It is the proximity of the plate connected to the center conductor of the resonator to the ground that is doing the tuning!  If you can't get the frequency low enough, you can add more strips to the center conductor - but you will probably want to remove the coupling capacitor (e.g. the coax center conductor) to prevent melting it when soldering.

Optimizing for "high" or "low" pass:

As described above, the notch will be more or less symmetrical - but in most cases you will want a bit of asymmetry - that is, you'll want the effect of the notch to diminish more on one side than the other.  Doing this allows you to place the notch frequency (the one to block) and the desired frequency (the one that you want) closer together without as much attenuation.

Figure 8:
The simplest form of the "high pass" notch, used during
initial testing of the concept - See the results in Figure 9.
Click on the image for a larger version.

"High-pass" = Parallel capacitor

In our case - with the higher of the two notches as 145.01 MHz and the desired signal at 147.82 MHz, we want the attenuation to be reduced rapidly above the notch frequency to avoid attenuating the 147.82 signal - and this may be done by putting a capacitor in parallel with the center of the coupling capacitor and ground:  A careful look at Figure 7 will reveal a small ceramic trimmer capacitor.

This configuration is more clearly seen in Figure 8:  There, we have the simplest - and kludgiest - possible form of the notch filter where you can see two ceramic trimmer capacitors connected across the center coupling capacitor and the center pin of the BNC connector.  Off the photo (to the upper-left) was the connection to a "tee" connector and the NanoVNA.  If you just want to get a "feel" for how the notch works and tunes, this mechanically simple set-up is fine - but it is far too fragile and unstable for "permanent" use.

For 2 meters, a capacitor that can be varied form 2-35pF or so is usually adequate - the higher the capacitance, the more effect there is on the asymmetry - but at some point (with too much capacitance) losses and filter "shape" will start to degrade - particularly with inexpensive ceramic and plastic trimmer capacitors.  Ideally, an air-type variable capacitor is used, but an inexpensive ceramic trimmer will suffice for receive-only applications - and if the separation is fairly wide, as is the case here.  For transmit applications, the air trimmer - or a high-quality porcelain type is recommended.

"Low-pass" = Parallel inductor

While the parallel capacitor will shift the shape of the notch's "shoulders" for "low notch/high pass" operation, the use of a parallel inductor will cause the response to become "low pass/high notch" where the reduced attenuation is below the notch frequency.  If we'd needed to construct a notch filter to keep the 147.22 repeater's transmit signal out of the 145.01 packet's receiver, we would use a parallel inductor.

It is fortunate that an inductor is trivial to construct and adjust.  For 2 meters, one would start out with 4-5 turns wound on a 3/8" (10mm) drill bit using solid-core wire of about any size that will hold its shape:  12-18 AWG (2-1mm diameter) copper wire will do.  Inductance can be reduced by stretching the coil of wire and/or reducing the number of turns.  As with the capacitor, this adjustment is iterative:  Reducing the inductance will make the asymmetry more pronounced and with lower inductance, the desired frequency and the notch frequency can be placed closer together - but decrease the inductance too much, loss will increase.

Comment:  The asymmetry of the "pass" and "notch" is why some of the common repeater duplexers have the word "pass" in their product description:  It simply means that on one side of the notch or the other the attenuation is lower to favor receive/transmit.

Results:

Figure 9:
VNA sweep of one of the prototype notch filter depicted  in
Figure 8.  This shows the asymmetric nature of the notch and
"pass" response (blue trace) when a parallel capacitor is used.
 The yellow trace shows the low SWR at the pass frequency.
Click on the image for a larger version.

Figure 9 shows a the sweep of the assembly shown in Figure 8 from a NanoVNA screen.

The blue trace shows the attenuation plot:  At the depth of the notch (marker #1) we have over 24dB of attenuation, which is about what one can expect from a notch cavity simply "teed" into the NanoVNA's signal path.

We can also see the asymmetry of the blue trace:  Above the notch frequency we see Marker #2 - which is a few MHz above the notch and how the attenuation decreases rapidly - to less than 0.5dB.  In comparison the blue trace below the notch frequency has higher attenuation near the notch frequency.

If you look carefully you'll also notice that just above the notch frequency, the attenuation (blue trace) is reduced to the lowest value right at the desired pass frequency:  This is our goal - set the notch at the frequency we want to reject and then set the parallel inductor or capacitor to the value that yields the lowest attenuation and lowest VSWR (the yellow trace) at the frequency that we want to pass.

Again, if we'd placed an inductor across the circuit rather than a capacitor, this asymmetry would be reversed and we'd have the lower attenuation below the notch frequency.

Note:  This sweep was done with the configuration depicted in Figure 8 at whatever frequency it happened to resonate "near-ish" 2 meters to test how well everything would work.  Once I was satisfied that this notch filter could be useful, I rebuilt it into the more permanent configuration and tuned it properly, onto frequency.

Putting two notches together:

Because we needed to knock down both 144.39 and 145.01 MHz, we can see from the Figure 9 that we'd need two notch filters cascaded to provide good attenuation and not affect the 147.82 MHz repeater input frequency.  A close look at Figure 1 will reveal that these two filters are, in fact, cascaded - the signal from the antenna (via the receiver branch of the repeater's duplexer) coming in via one of the BNC Tees and going out to the receiver via another.

The cable between the two notches should be an electrical quarter wavelength - or an odd multiple thereof (e.g. 3/4, 5/4) to maximize the effectiveness of the two notches together:  Since we only need a very short cable, we can use 1/4 wavelength here.  A quarter wave transmission line has an interesting property:  Short out one end and the impedance on the other end goes very high - and vice-versa.  To calculate the length of a quarter-wave line we can use some familiar formulas:

300/Frequency (in MHz) = Wavelength in meters

If we plug 145 MHz into the above equation (300/145) we get a length of 2.069 meters (multiply this by 3.28 and we get 6.79 feet).

Since we are using coaxial cable, we need to include its velocity factor.  Since the 1/4 wave jumper is foam-type RG-8X we know that its velocity factor is 0.79 - that is, the RF travels 79% of the speed of light through the cable, meaning that it should be shorter than a wavelength in free space, so:

2.069 * 0.79 = 1.63 meters (5.35')

(Solid dielectric cable - like many types of RG-8 and RG-58 will have a velocity factor of about 0.66, making a 1/4 wave even shorter!  There are online tables showing the velocity factor of many types of cable - refer to one of these if you aren't sure of the velocity factor of your cable.)

Since this is a full wavelength, we divide this length by 4 to get the electrical quarter-wavelength:

1.63 / 4 = 0.408 meters (16.09")

As it turns out, the velocity factor of common coaxial cables can vary by several percent - but the length of a quarter-wave section is pretty forgiving:  It can be as much as 20% off in either direction without causing too much degradation from the ideal (e.g. it will work "Ok") - but it's good to be as precise as possible.  When determining the length of the 1/4 wave jumper, one should include the length to the tips of the connectors, not just the length of the cable itself. 

Figure 10:
The response of the two cascaded notch filters - one tune to
144.39 and the other to 145.01 MHz.
Click on the image for a larger version.

Because we know that the notch filters present a "short" at their tuned frequency, that means that the other end of a 1/4 wave coax at that same point will go high impedance - making the "shorting" of the second cavity even more effective.  In testing - with the two notches tuned to the same frequency, the total depth of the notch was on the order of 60dB - significantly higher than the sum of the two notches individually - their efficacy improved by the 1/4 wave cable between them.

As we needed to "stagger" the two notches to offer best attenuation at the two packet frequencies, the maximum depth was reduced, but as can be seen in Figure 10, the result is quite good:  Markers 1 and 2 show 144.39 and 145.01 MHz, respectively with more than 34 dB of attenuation:  The two notches are close enough to each other than there is some added depth by their interaction.  Marker 3 at the repeater input frequency of 147.82 has an attenuation of just 0.79dB - not to bad for a homebrew filter made from scrap pieces!

Comment:  If you are wondering of the 0.79dB attenuation was excessive, consider the following:  Many repeaters are at shared sites with other users and equipment - in this case, there were two other land-mobile sites very nearby along with a very large cell site.  Because of this, there is excess background noise generated by this other gear that is out of control of the amateur repeater operator - but this also means that the ultimate sensitivity is somewhat limited by this noise floor.  Using an "Iso-Tee", it was determined that the sensitivity of this repeater - even with coax, duplexer and now notch filter losses - was "site noise floor limited" by a couple of dB, so the addition of this filter did not have any effect on its actual sensitivity.

Putting it together:

Looking again at Figure 1, you will noticed that the two notches filters are connected together mechanically:  Short pieces of PVC "wood" (available from the hardware store) were cut and a hole saw was used to make two holes in each piece, slipped over the end and then secured to the notch assemblies with RTV ("Silicone") adhesive.

Rather than leaving the tops of the PC board boxes open where bugs and debris might cause detuning, they were covered with aluminum furnace tape which worked just as well as soldering a metal lid would have - plus it was cheap and easy!  (The boxes were deep enough that the proximity of metal - or not - at the top had negligible effect on the tuning of the resonators.

Did it work?

At the time we installed the filter, the packet stations were down, so we tested the efficacy of the filter by transmitting at high power on the two frequencies alternately from an on-site mobile-mounted transceiver.  Without the filter, a bit of desense from this (very) nearby transmitter was noted in the receiver, but was absent with it inline.

With at least 34dB of attenuation at either packet frequency we were confident that the modest amount of desense (on the order of 10-15dB - enough to mask weak signals, but not strong ones) - IF it was caused by receiver overload - would be completely solved by attenuating those signals by a factor of over 2000.  If it had no effect at all, we would know that it was, in fact, the packet transmitters generating noise.

Some time later the packet stations were again active - but causing a bit of desense, but this was not unexpected:  At the start, we were not sure if the cause of the desense was due to the repeater's receiver being overloaded, or noise from the packet transmitter - but because the amount of desense was the same after adding the notch filter we can conclude that the source of desense was, in fact, noise from the packet transmitters.

Having done due diligence and installed these filters on our receiver, we could then report back to the owner of the packet transmitters what we had done and more authoritatively request that they install appropriate filtering on their transmitters (notch or pass cavities - preferably the latter) in order to be good neighbors, themselves.

* * *

"I have 'xxx' type of cable - will it work?"

The dimensions given in this article are approximate, but should be "close-ish" for most types of air and foam dielectric cable.  While I have not constructed a band-pass filter with much smaller Heliax-like cable such as 1/2" or 3/4", it should work - but one should expect somewhat lower performance (e.g. not-as-narrow band-pass with higher losses) - but it may still be useful.  With these smaller cables you may not be able to put the coupling inside the center conductor, so you'll have to get creative.

Because of the wide availability of tools like the NanoVNA, constructing this sort of device is made much easier and allows one to characterize both its insertion loss and response as well as experimentally determining what is required to use whatever large-ish coaxial cable that you might have on-hand.

"Will this work on (some other band)?"

Yes, it should:  Notch-only filters of this type were constructed for a 6 meter repeater - and depending on your motivation, one could also build such things for 10 meters or even the HF bands!

It is likely that, with due care, that one could use these same techniques on the 222 MHz and 70cm bands provided that one keeps in mind their practical limitations.

 

 * * *

Related articles:

  • A 2-meter band-pass cavity using surplus Heliax - link - This article describes constructing a simple band-pass filter using 1-5/8" Heliax. The techniques used in that article are the same as those applied here.
  • Second Generation Six-Meter Heliax Duplexer by KF6YB - link  - This article describes a notch type duplexer rather than pass cavities, but the concerns and construction techniques are similar.
  • When Band-Pass/Band-Reject (Bp/Br) Duplexers really aren't bandpass - link - This is a longer, more in-depth discussion about the issues with such devices and why pass cavities should be important components in any repeater system.

 

* * *

This page stolen from ka7oei.blogspot.com

[End]


Friday, November 3, 2023

Observations, analysis and field use of the JPC-7 portable "dipole" antenna

Figure 1:
The JPC-7 and its original set of components in the case.  On
the left is a zippered section with the balun, strap, feedpoint
and mounting hardware for the elements.  On the right
can be seen the two telescoping sections, the two loading
coils and the four screw-together mast sections.
Click on the image for a larger version.
The JPC-7 (apparently by BD7JPC) is a portable dipole antenna - somewhat similar to the "Buddipole" - in that it is tripod-mounted, with telescoping elements that can be oriented horizontally.  Both use loading coils to increase the electrical length of the antenna, allowing them to operate down to 40 meters in their standard configuration.

I was able to get mine, shipped, via Ali Express for about US$170, but it is also sold domestically (in the U.S.) from a number of vendors - sometimes under the brand name of "Chelegance".

Notes:

  • Silver-plated coils:  Since this article was posted I have added an article describing the effects of the stainless-steel coils in the JPC-7 and JPC-12 and how to rewind with silver-plated copper, found HERE.  This article also discusses the effects of stainless-steel versus copper loading coils in general.
  • JPC-12 Vertical:  I have also discussed the JPC-12 vertical antenna made by the same folks - you may read about that antenna HERE.

A portable antenna is not the same as a "home" antenna

As you might expect, this antenna is intended for portable use - and easy-to-assemble, quickly-deployable antennas are not likely to offer high performance compared to their "ful-sized, high up in the tree" counterparts that you might have at your home QTH.  Rather, this antenna's height is limited by the tripod on which it is mounted - which, for the lower bands where its height above ground is definitely below 1/4 wavelength - is likely to put it squarely in the "NVIS" (Near Vertical Incident Skywave) category - that is, an antenna with a rather high radiation angle that better-favors nearer stations than being a DX antenna.

Additionally, its total element length as-shipped (with the two screw-in sections and the telescoping whip fully-extended, sans coil on each side) is 125" (3.175 meters) - approximately a quarter-wavelength at 22MHz - near, but above the 15 meter band meaning - that for all HF amateur bands 15 meters and below it requires the addition of the coils' inductance to resonate the two elements.  Being a loaded antenna - and with a small-ish aperture and with coils losses - means that its efficiency IS going to be less than that of its full-sized antenna (e.g. half-wave dipole) counterpart.

Of course, the entire reason for using a "portable" antenna is to enjoy the convenience of an antenna that is quick to deploy and fairly easy to transport - and anyone doing this knows (or should know) that one must often sacrifice performance when doing this!

Having said this, after using the JPC-7 in the field several times I've found that it holds up pretty well against a similar "full size" antenna (e.g. dipole) on the higher bands (20 and up) while on 40 meters, subjective analysis indicates that it's down by "about an S-unit" (e.g. the standard 6 dB IRU S-unit).  For SSB (voice) operation, this is usually tolerable under reasonable conditions and for digital or CW, it may hardly be noticeable.

Figure 2:
The components included with the JPC-7 - except the
strap and the manual.
Click on the image for a larger version.

What is included with the JPC-7:

  • Four aluminum mast sections.  These are hollow tubes with (pressed in?) in screw fittings on the ends - one male and the other female, both with M10-1.5 coarse threads that may be assembled piece-by-piece into a mast/extension.  End-to-end these measure 13-3/16" (33.5cm) each, including the protruding screw - 12-3/4" (32.4cm) from flat to flat.  These are 3/4" (1.9cm) diameter.  There are two of these sections per element to achieve the  125" (3.175 meter) length of each.
  • Telescoping sections.  These are stainless steel telescoping rods that are 13-1/8" (33.4cm) long including the threaded stud (12-7/8" or 32.7cm without) when collapsed and 99-11/16" (8' 3-11/16" or 253.2cm) when fully extended - not including the stud.
As with all stainless-steel telescoping whips, it is strongly recommended that you lubricate the sections as soon as you receive them.  As with about every telescoping whip you will ever see, these sections are "stainless on stainless" and as with many friction surfaces between the same type of metal, they will eventually gall and become increasingly difficult to operate as they scratch each other.  I use PTFE (Teflon) based "Super Lube" for this purpose as it does not dry out and become gummy as normal distillate oils like "3-in-1" or "household" do.  Do not use "lubricants" like "WD-40" as these aren't actually lubricants in the traditional sense in that they tend to evaporate and leave a varnish behind.  If the sections do get stiff over time due to surface abrasion, a buffing with very fine steel wool and/or very fine (1000 or higher) grid sandpaper followed by wiping down and lubricating may help loosen them.
  • Adjustable coils.  These are constructed of what appears to be thermoplastic or possibly nylon with molded grooves for the wire.  This unit is connected to the others via a male threaded stud on the bottom and female threads on the top, both being M10-1.5 like everything else.
The form itself is 4-1/2" (11.4cm) long not including the stud and 1-11/16" (4.3cm) diameter - wound with 34 turns of #18 (1mm) stainless steel wire with an inside diameter of approximately 1.66" (4.21cm) over a length of about 2.725" (6.92cm).  It has a slider with a notched spring that makes contact with the coil and this moves along a stainless steel rod about 0.12" (3mm) diameter that is insulated at the top, meaning that as the slider is moved down, the inductance of the coil is increased.  I suggest that a drop of lubricant (I recommend the PTFE-based "Super Lube" as it doesn't dry and get gummy) be applied to the slider to make it easier to adjust and to minimize the probability of galling.
 
The coils have painted markings indicating "approximate" locations of the tap for both 20 and 40 meters when the telescoping section is adjusted as described in the manual.  These coils are wound with 1mm diameter (approx. 18 AWG) 316 stainless steel wire:  The maximum inductance is a bit over 20uH and the DC resistance of the full coil is about 4 ohms - more on this later.
  • Figure 3:
    A close-up of the feedpoint mount showing the
    brass inserts and index pins.  The holes in the knurled
    knobs are sized to receive the miniature banana plugs
    from the balun.
    Click on the image for a larger version.
    Feedpoint mount.   This is a heavy plastic piece molded about pieces of brass into which the elements/coils are threaded.  There are three 10mm x 1.5mm female threads into the brass inserts plus another female thread of larger size (1/2" NPT) into which the aluminum 5/8" gaffer stud mount is screwed.  On the surfaces with the brass inserts and the 10mm x 1.5mm female threads are a series of index holes into which the element mounts (described below) are seated to allow the elements to be adjusted at various angles.  Electrical connection is made via holes in the brass to receive 2.5mm miniature banana plugs (visible in Figure 3) which contact the adjacent 10mm x 1.5mm female thread bodies.
Element mounts.  These are two heavy-duty nickel-plated brass adapters that are held to the feedpoint mount via 10mm x 1.5mm screws with large handles - both included.  Into the mounting surfaces are holes to receive index pins allow the elements to be rotated to various angles - from a horizontal dipole to a "Vee" configuration - and even to an "L" with one element vertical and the other horizontal.  It can also be configured with just a single element as a plain vertical if one so-chooses - the counterpoise/ground needing to be supplied by the user.  Figure 8, below, offers a better view of how this is used.
  • 5/8" stud (gaffer) mount.  As mentioned earlier, this kit includes a male 5/8" stud mount commonly found on photographic lighting tripods.  The other side of this has 1/2" NPT pipe threads that screw into the feedpoint mount.  This piece is shown in Figure 4.

Figure 4:
5/8 stud mount adapter to be used with
lighting tripods.  The "other" side is a 1/2 inch
NPT pipe thread that screws into the feedpoint mount.
Click on the image for a larger version.

  • 1:1 balun.  This appears to be a "voltage" balun, with DC continuity between the "balanced" and "unbalanced" sections and across the windings themselves.  This is in contrast to a "current" type balun that would typically consist of feedline, twisted pair or two conductors wound as a common-mode choke on a ferrite core. More on this later.
  • Hook-and-loop ("Velcro") strap for the balun.  This is used to attach the balun to the mast to prevent the weight of the coax and balun from pulling on the feedpoint mount.  This strap appears to be generic and doesn't really fit the balun too well unless it is cinched up, so I zip-tied it in place to keep both of them together. 
  • Padded carrying case.  This zippered case is about 14" x 9" (35.5x23cm) with elastic loops to retain the above antenna components and a zippered "net" pocket to contain the components for the antenna mount, balun, and the instructions.  There is ample room in this case to add additional components such as coaxial cable - and enhancements to the antenna, as discussed below.  
  • Instruction manual.  The instructions included with this antenna are only somewhat better than typical "Chinese English" - apparently produced with the help of an online translator rather than someone with intimate knowledge of the English language resulting in a combination of head-scratching, laughter and frustration when trying to make sense of them.  Additionally, the instructions that came with my antenna included those for the JPC-12 vertical as well, printed on the obverse side of the manual.

Construction and build quality

About a year ago I purchased a JPC-12 vertical antenna and it shares many of the same components as this antenna - the only real differences are that this antenna comes with two telescoping whips and loading coils, the center mount for the elements, a 1:1 balun, and the 5/8" stud adapter for the center mount.

Many of these components are the same as supplied with the JPC-12 vertical:  The loading coils, the telescoping whips, and the screw-together antenna sections.  In other words, if you have both antennas, you can mix-match parts to augment the other.  You can, in fact, buy kits of parts for either antenna to supply the missing pieces to convert from one to the other.

Mechanically, this antenna seems to be quite well built:  During use, I have no sense of anything being "about to come apart" or "just barely good enough".  I suspect that the designers of this antenna did so iteratively, and the end product is a result of some refinement over time.  The only really fragile parts are the telescoping whips, but these things are, by definition, fragile - no matter who makes them!

How it is mounted

This antenna does NOT come with any tripod or other support, but it offers three ways of being mounted:

  • 1/2" NPT threads.  The center support, as the primary mounting, has female 1/2" NPT threads.  If you have a piece of pipe with that type of thread on it, you can mount the antenna directly to it.
  • 5/8" male stud mount.  This antenna comes with a machined aluminum mount (seen in Figure 4) that screws into 1/2" NPT threads in the center support that is a 5/8" stud mount - sometimes referred to as a "Gaffer" or "Grip" mount - of the sort found everywhere on tripods used for holding photographic lights.
  • 10mm x 1.5mm thread.  If you want to configure this antenna as a dipole, you also have the option of using a 10mm x 1.5mm thread that is on the side opposite the female threads into which the 5/8" stud mount screws.  While this thread isn't particularly common in the U.S.A., it would seem that this is a common size for portable antennas everywhere else in the world and hardware of this size is available at larger U.S. hardware stores.  As this mounting point may be used as part of the antenna
    Figure 5:
    A homebrew double-female 5/8 stud adapter.  These adapters
    have 3/8" threads and were attached using a thread
    coupler.  This piece was necessary as both the antenna and my
    tripod have male 5/8" stud mounts on them!
    Click on the image for a larger version.
    (when configured in an "L" shape or if configured as a vertical-only)
    so it's the same threads as the screw-in element sections.

For me the 5/8" male stud mount is the most useful as it happens that I have on hand an old gaffer tripod (light stand) of this sort - but there's a catch:  It, too, has a 5/8" male stud mount!  It would seem that these tripods come both ways - with either a male or female 5/8" mount, but for less than US$15 I was able to construct a "double-female" adapter that solved the problem.  From Amazon, I ordered two 5/8 female stud to 3/8"-16 adapters and coupled them together with a 3/8" thread coupler as seen in Figure 5.  The only "trick" with this was that I had to sort through my collection of flat washers to find the combination of thicknesses that resulted in both knobs facing the same direction when the adapters were tightened to the thread coupler.

Element configuration

As with any antenna that you are likely to come across, the only portions of the antenna that actually radiate energy in the far field are those with current flowing through them:  The higher the current, the more energy is radiated.  By extension, the very ends of the wire - or, in this case, the ends of the telescoping section - have essentially zero current and do not radiate.  As the total length of conductors prior to the loading coil (screw-together sections, feedpoint mount, connecting wires) is about 56" (1.42 meters) this represents only about 3.6% of a wavelength at 40 meters.

It is for this reason that the preferred configuration is to have the screw-together sections connected directly to the feedpoint mount, then the loading coil and then the telescoping section, placing the loading coils nearly 30" (75cm) from the feed.  As the total length of the telescoping sections alone put together is about 198" (5 meters) - which is about 12.5% of a wavelength - you might think that they are doing the lions share of radiating - but that's not really the case.

Particularly at lower bands, it is understandable why coil losses are of such importance - and also why even a relatively small amount of lengthening of the antenna can improve performance on the lower frequencies:  Adding two more screw-together sections (one per side) increase the length "before the loading coil" from 56" to 84" (2.13 meters) - or about 5.3% of a wavelength and not only increase the aperture of the antenna, but it will also allow a reduction of the amount of inductance (and coil loss) required to resonate the antenna.

Further improvement can be made by adding a bit of extra length to the telescoping whips by clipping hanging wires to the end of it:  This will further reduce the amount of inductance needed to resonate, but it will also increase the effective portion of the whips that are carrying RF current.  (This is discussed further in the section on 60 meter coverage, below.)

Frequency coverage

This antenna is advertised to cover 40 through 6 meters - and this is certainly true:  When the four supplied mast sections are installed (two per side) the lowest frequency at which it can be resonated with the telescoping rods at full extension and the inductors set at maximum is around 6.7-6.8 MHz - well below the entirety of the 40 meter band.

On 40 meters, the 2:1 VSWR bandwidth was typically around 120 kHz:  A 2:1 VSWR is about the maximum mismatch at which most modern radios will operate at full power before SWR "foldback" occurs, reducing transmit power.  Of course, if your radio has a built-in tuner - even one with a limited range - you will certainly be able to make the radio "happy" across the entire 40 meter band without fussing with the antenna, even if it isn't tuned exactly to your operating frequency.

On the other extreme, with the minimum coil inductance and the two telescoping rods at maximum extension the resonant frequency was about 21.7 MHz:  This means that for all amateur bands 15 meters and lower, you will need the inductors - but for 12 meters and up you can omit them entirely (which is recommended!), bringing the antenna to resonance solely by adjusting the length of the telescoping sections.

Tuning the antenna

This may be where some people have issues.  I am very comfortable using a NanoVNA:  I have several of these as they are both cheap and extremely useful - the only down-side really being that their screens are not easily viewed in direct sunlight - but simply standing with my back to the sun was enough to make it usable as all one is trying to see is the trace on the screen rather than any fine detail.

The biggest advantage of the NanoVNA over a traditional antenna analyzer is that you get the "big picture" of what is going on:  You can instantly see where the antenna is resonant  - and how good the match may be.  More importantly, you can see at a glance if the antenna is tuned high (too little inductance) or too low (too much inductance) and make adjustments accordingly whereas using a conventional antenna analyzer will require you to sweep up and down:  Still do-able, but less convenient.

Tuning is somewhat complicated by two factors:

  • There are two coils to adjust - and they must both be pretty close to each other in terms of adjustment to get the best match.  Simply looking at the coils one can "eyeball" the settings of the slider/contact to get them very close to each other - something that becomes easier with practice.
  • The "resolution" of the inductors' adjustments is limited by the fact that one can make adjustments by one turn at a time with the slider.  At 20 meters and higher, being able to only adjust inductance one turn at a time is likely to result in the best match being just above or below the desired frequency.  At lower frequencies (lots of turns) - say 40 and 30 meters - you can likely get 2:1 or better by adjusting the coil taps alone, but at higher frequencies you will likely need to tune for the best match just below the frequency of interest and then shorten the telescoping rods slightly to bring it right onto frequency.

 Once I'd used the antenna a few times I found that I could change bands in 2-3 minutes as I would:

  • Lower the antenna to shoulder height so that the coils and telescoping rods may be reached.  If you had previously shortened the telescoping elements for fine-tuning a band you should reset them to full length.
  • Set the NanoVNA to cover from the frequency to which it is already tuned and where I want to go:  If I was setting it up for the first time I would set the 'VNA to cover above and below the desired frequency by 5 MHz or so so I could see the resonant point even when it was far off-frequency.  After using it a few times you will remember about where the coil taps need to be set for a particular band.
  • On the NanoVNA I would then set a marker to the desired operating frequency.
  • I would then "walk" both coils up/down to the desired frequency while watching the 'VNA.  As the tuning of the elements interact, you may have to iterate a bit to get the VSWR down.  Again, you may have to tune for best match at a frequency just below the target frequency and then shorten the telescoping sections.
  • I would raise the mast to full height again.  I noticed  a slight increase in resonant frequency (particularly on the lower bands - 40 and 30 meters) by raising the antenna on the order of 50 kHz on 40 meters.  Usually, this doesn't matter, but with a bit of practice/experience you'll be able to compensate for this while tuning.
  • A match of 2:1 or better was easily obtained - but don't expect to get a 1:1 match all of the time as the only adjustments are those of resonating the elements and nothing to take into account the actual feedpoint resistance at resonance.  Practically speaking, there is no performance difference between a 2:1 and 1:1 match unless your radio's power drops back significantly:  An antenna tuner could be used, but this will surely insert more loss than having a modest mismatch!

Figure 6:
As with almost any inductor adjustable using sliders, care
should be taken to assure that only one turn is being touched
by the contact, as shown.
Click on the image for a larger version.
All of that sounds complicated - and it may be, the first time doing it - but I found it to be very quick and easy, particularly after even just a little bit of practice!

Carefully adjusting coil taps

 If you look very carefully at the sliding coil taps you'll notice that if very carefully adjusted that they will contact just one turn of wire - but it is almost easier for the contact spring to bridge two turns of wire, shorting them together.  When this happens the inductance will go down slightly and you may see the resonance go up in frequency unexpectedly.  Additionally, the shorting of two turns can also reduce the "Q" (and efficiency) of the coil slightly.

If you are aware of this situation - which can occur with nearly all tapped inductors adjusted with a slider - you can start to "feel" when the slider bridges two turns of the coil and avoid its happening as you make the adjustments.

* * *

Suggested modifications/additions:

All electrically-short antennas that require series inductance for tuning to resonance - like this one - will lose efficiency due to losses in the coil, but this can be offset - at least somewhat - by increasing the length of the elements themselves.  One of the easiest ways to do this is to purchase a couple of extra screw-on mast sections:  The addition of one on each side will increase the total length of the antenna by about 25" (64cm) and allow a slight decrease in the required inductance - resulting in slightly lower loss and increase the aperture of the antenna slightly.  These additional screw-on sections are typically available from the sellers of the antenna for between US $10 and $15 each but are often called something like "Dedicated lengthened vibrator for JPC-7 (JPC-12)" or similar due to quirks of the translation.

60 meter operation

Figure 7:
The elements may be lengthened by clipping a lead to each
end of the telescoping sections, reducing the amount of
needed inductance - and also allowing resonance on lower
bands - in this case, 60 meters.
Click on the image for a larger version.

While adding two additional sections (on on each side, between the coil and the whip) - and rearranging the antenna with the coils located next to the feedpoint (rather than the usual configuration in which the coils are located away from the feedpoint) - will bring the resonant frequency down to about 5.7 MHz with full inductance and extension of the telescoping sections.  In this configuration, the added length beyond the coil adds significant capacitance, lowering the resonant frequency as compared to the normal coil location

The antenna can be made to cover 60 meters by clipping on short (18" or 46cm) jumper leads to the very end of the antenna elements and let them hang down.  Despite this being a less desirable configuration in terms of RF current distribution, in testing on the air, the signals were about 1 or 2 "S" units below a full-sized dipole, but still quite good for a fairly compact antenna that was  close to the ground in terms of wavelength.

If you wish to use the "stock" antenna on 60 meters rather than buying two extra screw-together sections, you'll need about 48" (1.25 meters)  of wire on each end:  For this I simply used two pair of 24" (approx. 100cm) clip leads connected end-to-end, each pair hanging from the tips of the telescoping section.

Longer is better

Of course these "extension" leads can be used for all bands for which the coils are needed to lower the inductance and reduce losses:  As it will always be the parts of the antenna that carry the most RF current that radiates the vast majority of the signal - and since those portions will always be the sections right near the coils for this type of antenna - adding these dropping wires at the ends won't appreciably affect the antenna pattern or its polarization.

As there is plenty of room to do so in the zipper case, I have since added two extra sections and two sets of "clip leads" permanently into the kit.

Get extra telescoping sections!

Finally, I would order at least two extra telescoping sections as these are the most fragile parts of the antenna kit.  These can also be ordered from the same folks that sell the antennas for US $12-$16 each and are typically referred as something like "304 stainless steel 2.5M whip antenna for PAC-12 JPC7 portable shortwave antenna". 

The reason for ordering two of them is that if the antenna falls over, both whips are likely to be damaged (ask me how I know!):  The cost of getting two extra whips is likely to be less than the cost of fuel for even a modest road trip to wherever you are going, so their price should be kept in perspective.  As the zippered case for the antenna has plenty of extra elastic loops inside, there is ready storage for these two extra whips with no modification.

A word of caution:  However you store them, do not allow the telescoping whips to lay loosely in the case:  If they bash into something else they can be easily dented which may make it impossible for them to be extended/retracted.  For this reason they should be secured in the elastic strap, or individually in a tubes or padded cases.

Note:  There are also available much heavier and longer telescoping whips with the same M10x1.5 thread that would easily allow 60 meter coverage:  I have not tried these to see how well they would work, mechanically, or if it would even be a good idea to do so (e.g. extra stress on the tubes, coils, mounting point - or how stable such a thing might be on a tripod).

Figure 8:
The mounting of the balun, just below the feedpoint mount.
The index holes allow flexibility in the orientation, the
connection being made by 2.5mm banana plugs.
Here, the antenna is shown with the elements configured
one hole higher than "flat", forming a lazy "Vee"
shape as seen in Figures 9 and 10.
Click on the image for a larger version.

Additional comments:

"To vee, or not to vee"

The feedpoint mount has a number of indexed holes that allow the elements to be mounted in a variety of configurations, from flat, in a number of "Vee" configurations, or even an "L" or vertical configuration.  

Personally, I use the flattest "Vee" configuration as seen in Figures 8, 9 and 10.  This configuration keeps the drooping ends of the telescoping whips higher than the feedpoint and helps clear any local obstacles (trees!)  - and just looks cool!

As can be seen in Figure 8, the connection between the balun and the feedpoint is made by plugging 2.5mm miniature banana plugs into the brass receptacles on the feed.  Shown in the photo are connections to the two sides, typically used for a dipole arrangement, but the third, unused connection on the top could be used to hold an element horizontal while one of the side connections hold it vertical - more on the use of this antenna as a vertical in the next section.

It should be no surprise that these 2.5mm miniature banana plugs are quite small and fragile and if one isn't careful - say, by allowing the weight of the balun to be supported by the wires rather than using the hook-and-loop strap - they can be broken.  For this reason I ordered a pack of ten 2.5mm banana plugs from Amazon and made a pair of short (4", 10cm) leads - one end with a small alligator clip and the other with a 2.5mm banana plug - to allow me to make a temporary connection should one get broken off in the field - something that could torpedo an activation if you didn't have spare parts! 

Operating as a vertical antenna

Because of the flexibility of the mounting point, it is possible to use this same kit as a vertical antenna with the second element as a resonant (rod) ground "plane" if - due to space or personal preference - emitting a signal with a vertically-polarized component is desired.  While this will certainly "work", if you do plan to operate with vertical polarization its recommended that you add several (2 or more) wire "radials" or counterpoises.

Because of the included balun (more on this in a moment) the coaxial feedline itself will not act as an effective part of the counterpoise network so rather than connecting additional radials to the shield, the ends of the wire should be clamped under the washer/bolt that holds the horizontally-configured element in place.  Of course, one need not use the balun and connect the coaxial cable directly, but if you choose this option you will be on your own to supply the means to make such a connection.

For best results with the fewest number of radials, choosing lengths that are odd-number quarter wavelengths long (1/4, 3/4, 5/4) and keeping them elevated a foot (25cm) or more off the ground is suggested as this will help minimize "ground" losses.  Having said this, almost no matter what you do, you will probably be able to radiate a useful amount of signal:  Operating CW or digital modes offers an improvement in "talk" capability owing to their efficiency - but if you are planning to operate SSB, it's worth taking a bit of extra time and effort to maximize performance.

Would I operate this antenna in "vertical" mode?  While I don't have plans to do so, I have purchased an extra ground stake of the sort used on the JPC-12 vertical, and the short banana plug/clip lead jumpers that I made could be used to make a temporary connection directly to a coaxial connector.

Nature of the balun

The supplied balun has a 1:1 impedance ratio and has DC connection between the input and output - but since there is a DC connection between all of the conductors, it is more than a simple current balun (e.g. transmission line wound on ferrite).  As the balun seems to work well, I have no reason to break it open to figure out what's inside, but I did a bit of "buzzing" of the connections with a meter to measure inductance and here are the results:

  • Between coax shield and center conductor:  16.9uH
  • Between red and black (on antenna side):   16.9uH
  • Between center coax and black:  38.5uH
  • Between center and red:  3.4uH
  • Between Shield and black:  3.4uH
  • Between Shield and red:  3.4uH
  • The DC resistance between any combination of the leads is well under 1 ohm.

What does this tell us?  The inductance readings of about 16.9uH indicate that this may be a voltage balun providing about 500 ohms of inductive reactance at 5 MHz - more than enough for reasonable efficiency.  The interesting reading is the inductance between the center coaxial connection and the black wire which is only twice the inductance of the input or output windings:  If there was a direct connection between one of the coax and one of the output wires this would imply twice the number of turns and four times the inductance - but since it is only twice, this indicates that the total number of turns in the "center coax to black" route is about sqrt(2) (or 1.414x) as many turns as the primary/secondary - or there is another inductor in there.

Figure 9:
The JPC-7 backgrounded by red rock during a POTA
operating in K-0010.
Click on the image for a larger version.

While I'm sure that the balun is very simple, its exact configuration/wiring escapes me at this time.

Coil losses

As mentioned earlier, the coil is wound with 18 AWG (1mm diameter) type 316 stainless steel wire.  Fortunately, this wire appears is austenitic - which is to say that it is not of the variety that is magnetic and thus has a permeability of unity:  Were it magnetic, this would negatively impact performance significantly.

Knowing the diameter of the coil form and the fact that there are 34 turns, we know that the total length of the wire used is approximately 180 inches (457cm) and measurement shows that the stainless steel wire coil has a total DC resistance of about 4 ohms.  Using Owen Duffy's online skin effect calculator (link) and assuming 1mm diameter, 316 Stainless we can calculate the approximate RF resistance including skin effect - the tendency for RF to flow on the outside skin of a conductor rather than through its cross-section - versus frequency:

  • 3.5 MHz = 5.2 ohms
  • 7 MHz = 7.2 ohms
  • 14 MHz = 9.6 ohms
  • 28 MHz = 13.6 ohms

If I make a very broad assumption that the feedpoint resistance at each coil is about 25 ohms (the two in series being around 50 ohms) we can see that in this hypothetical situation about a third of the total resistance could be due to the coil, and since P = I2R - and if we presume that the current is consistent throughout the coil (it probably is not) we can roughly estimate that the total power loss will be proportional to the resistance implying that about 1/3rd of the total power is lost in the coil.  In practical terms, a 33% power loss is around 4.8dB - still less than one "S" unit, so this loss may go unnoticed under typical conditions.

In operation, we would be unlikely to need all - or even most of the turns of the coil for operating on the higher bands, so the overall coil losses are likely to go down as the need for loading inductance at these frequencies is also significantly reduced:  Since we actually use only about 2/3 of the turns of the coil on 40 meters, the loss is more likely to be something on the order of 5 ohms rather than 7.2, reducing the loss even more.

Note:  K6STI's "coil" program - Link - calculates the loss for this coil as being closer to 8 than 5 ohms - a bit higher than the simple loss calculation of Owen Duffy's wire calculation and likely more representative of in-situ measurements.

When operating on 40 meters with 100 watts of CW or SSB, the coils definitely do get quite warm - but not dangerously so and thus I would presume that the very rough estimates above are likely in the ballpark:  If you operate heavy duty-cycle modes like RTTY or FT-8 and insist on running 100 watts key-down I would occasionally check the coils to be sure that they aren't getting too hot.

By comparison, the calculated DC resistance of  the same length of 18 AWG bare copper wire is under 0.5 ohms, but the RF resistance due to skin effect at 28 MHz is around 2 ohms and about an ohm at 7 MHz - roughly a 7:1 difference meaning that if the above analysis is in any way close to being correct, our losses at 7 MHz when using the full coil (again, we don't!) and presuming that the feedpoint of the individual coil stayed at 25 ohms (it probably won't) our losses would drop from about 30% to less than 5%.

As a consequence, if wound with copper/silver plated I would expect that the not only would the antenna become narrower than the 40 meter 2:1 bandwidth of about 120 kHz - which would make it slightly trickier to tune - I would also expect the feedpoint resistance to drop, possibly increasing the VSWR at the feedpoint.  From a practical standpoint, even a modest antenna tuner capable of handling only 3:1 mismatch should be able to cope with this, but it is likely that some of the gains from using lower-loss wire might be offset by the increase in losses caused by feedline mismatch and the losses within a tuner - both of which could easily exceed 3dB in a portable set-up with moderately-long, small-diameter coax.

Would it be worth rewinding the coil with (readily-available) 18AWG (1mm dia) silver-plated or bare copper wire?  Maybe

Note:  I have since rewound a coil with 18 AWG silver-plated copper jewelry wire and am in the process of doing direct comparisons with it and the original coil wound with stainless-steel wire - expect a blog entry on this in the near-ish future.

Final comments

Figure 10:
Operating 20 meter CW from POTA entity K-6085, with the
Conger mountains and the JPC-7 dipole in the background.
Click on the image for a larger version.

Is this an antenna that is worth getting?  I would have to say "yes".

Remembering that you will also need to supply a suitable tripod mount (e.g. an inexpensive "light stand" ) this antenna is quite portable and, if you have a bit of practice, quick to set up and adjust.  Unlike a vertical antenna, it doesn't need a set of ground radials and it is likely that the antenna itself will be up and above everyone's heads when it is deployed.

Best used on the higher bands (20 and higher) its efficiency will be quite good - certainly equal to or better than a typical mobile antenna.   As this is a large-ish antenna on a tripod, be sure to weigh down the legs and/or attach simple guying to it to prevent it from blowing over in the wind or being knocked over by tripping over the coax:  I can attest personally that the latter can easily happen!

* * *

I also have the JPC-12 vertical (discussed in the link below) and I find this antenna (the JPC-7 loaded dipole, that is) to be far more convenient to use than the vertical (e.g. no radial system), particularly if you plan to change bands several times during the operation - something that is quite likely to happen on the higher bands as propagation varies over the course of a few hours.  For the vertical, best performance requires adjusting the radials as well as the antenna itself, although it would probably work "just fine" if the radials are left at maximum length.  Another advantage of the JPC-7 loaded dipole being a (largely) horizontally-polarized antenna is that in an urban environment it is likely to intercept less noise on receive than a vertical - and it can be inconspicuous in its deployment as compared to a taller vertical.

For the lower bands (40 and 30 meters) the JPC-7 works quite well - particularly if one operates CW or digital modes.  As mentioned, it can also work competently on 60 meters as well with the addition of extra length of the elements by the purchasing of extra rods and/or simply attaching "drooping" wires to the ends of the telescoping rods.

Over the course of several POTA and related activations I have made about 500 contacts with this antenna on the band 60 through 15 meters - on CW and voice:  I'm sure that the antenna works well on 12, 10 and 6 meters as well, but I just haven't tried it on those bands.

Overwhelmingly, the sense has been "If I can hear them, they can hear me." with this antenna as I have worked quite a few QRP and DX stations that I could barely copy above the band's natural QRN level.  Admittedly, some of these times I was on the receiving end of the frenzy - being the activator during POTA operation - but there were many times when I had to stop operating not because I ran out of people to work, but because I ran out of time.

* * * * *

Related articles:

  • Silver-plated coils:  Since this article was posted I have added an article describing the effects of the stainless-steel coils in the JPC-7 and JPC-12 and how to rewind with silver-plated copper, found HERE.  This article also discusses the effects of stainless-steel versus copper loading coils in general.
  • I posted an article that gives a description of the JPC-12 loaded vertical - and ways to improve it - and it may be found HERE.

This page stolen from ka7oei.blogspot.com

[End]