Extending a Stellar Labs 30-2476
 

Interesting project, Bob.

Yes true.

Many models that I have seen, especially at UHF, fail to account for everything that can affect results.
Partially because NEC2 makes it difficult and partially because of failure to appreciate small details.

What modeling can do (if the model is accurate) is to provide insight and guidance.
Insight into how parameters affect results. Vary this, to see effect on that.

Unless one is seeking the last 0.1 dB of gain, small variations in director spacing are not critical.

Start with existing factory spacing, or the wide spacing could be OK.

Have used round aluminum tubing to sleeve inside of these booms

DX Engineering sells 6063-T832 that should be strong enough for this application.

https://www.dxengineering.com

Another option would be to omit one director and shorten the boom appropriately to use factory couplers.

Optimum spacing depends on antenna aperture (related to gain) and desired channels.
Higher gain and Lower channels would indicate wider pacing.

Maybe maximum spacing would be about 7 feet.

Wide spacing tends to increase gain and raise sidelobes.
Conversely, closer spacing yields lower gain and a better overall pattern (less sidelobes).

Vertical spacing behaves much different than horizontal spacing.
There is more coupling between antennas stacked vertically,
this affects antenna impedance, usually in a undesirable way.
Wider spacing vertically tends to minimize the undesirable coupling.

For performance of vertical stack to be mostly predictable, in terms of gain and pattern,
the antenna should be high above ground (preferably about 10 spacings high or more).
This height helps to make each antenna see the same signal.

Horizontal stacking is less afflicted by mutual coupling.
And, each antenna essentially sees the signal relative to ground.

If mast mounted relatively close to reflector, not much affect on gain.

It can affect the pattern by filling nulls.

Given that you have provided some vertical spacing with the U-pipe;

The further away from the driven element toward front of antenna,
the more likely horizontal pipe is to interfere.

For horizontal metal, more vertical spacing and further back is better.

The idea is to minimize exposure to the antenna aperture.

You could cut the reflectors and make directors of them.
Probably need a longer boom to effectively use more directors.

The Y10-7-13 has two dipoles (driven elements) to achieve sufficient bandwidth to cover channel 7 thru 13.

High band VHF (Ch 7-13) is considered moderately wideband for a Yagi antenna.

The Y10-7-13 was designed before modern computer modeling was feasible.
Computer modeling helps with designing Yagis for wideband performance.

The additional driven element in the Y10-7-13 actually cost a small amount of gain,
that was a tradeoff.

Driven elements generally are not as effective as parasitic element in increasing antenna gain.


Question:

What is the bending strength of your 1 & 3/8 exhaust pipe?

Comment: There can be a strong rotation force due to wind,
trying to rotate the U-pipe and antenna assembly downward.
Need sturdy non-rotatable mast to U-pipe mount.

.

Build details...
 

I'm jealous of all that sky. My home is surrounded by oaks, maples, red pines, and white pines. The red pines are tall trees with a tuft of needles at the top that sway in any wind.

How is your reception?

Mediocre reception so far.
 

The blue trace (antenna pair) shows a dip in middle of RF 8 spectrum.
That is indicative of two signals interfering with each other (out-of-phase).

An out-of-phase condition can result from:

* Multipath - the signal from the station arriving from two different paths,
one path delayed or advance relative to the other.

* Phase unbalance in Antenna system

One antenna providing a signal that when combined with phase opposing
the signal from the other antenna.

There are about five places in your setup (shown in initial post) where
phase misalignment could occur:

1. The balun at the antennas
2. Unequal coax lengths from antenna to preamps
3. RCA Preamps, Internal filters aligned different yielding phase differences
4. Combiner - Internal circuitry unbalanced
5. Unequal coax lengths from preamps to combiner

Seems, most likely suspect is the preamps (#3 above).

There are several Inductors (L) and capacitors (C) at input and output
of the RCA preamps. The L & C have to be hand tweaked for best signal
amplification, but there is no attempt to make one preamp identical to another.

If the preamps could be bypassed with F-81 connectors (barrels).
The signal would be lower, but if the spectrum was flatter, that might add insight.
May need to turn off power to avoid damage to power source.

If spectrum was flatter, then compare dual antennas without preamp
to single antenna without preamp.

------------------

If it appeared that cancellation was still occurring, check #1 above:

Might try switching baluns from from spare Driven elements (Folded dipoles) ,
to determine if the baluns were source of phase difference.

Doesn't seem that would have much affect on phasing. Very tiny
(almost not measurable) affect on gain.

Have seen similar performance with another different set of long Yagis.

.

Todays Testing
 

Reference 2nd and 3rd images in previous post:

http://forum.tvfool.com/showpost.php...91&postcount=8

Since the 'East Antenna Only' and the 'West Antenna Only' showed evidence
of phase cancellation (dip in the spectrum), that seems to indicate
that the signal self interference is outside the antenna pairs.

---------------

Reference 2nd image (green trace):

http://forum.tvfool.com/showpost.php...88&postcount=5

This antenna does not show the phase cancellation.

What could be the reason? Is this antenna at a different height?

Each of the dual antennas exhibit symptoms of phase cancellation and
the 3rd does not. If the 'blue-trace' antenna is at a different height,
then possibilities:

* It is at a height where ground reflection reinforces signal, whereas the
dual antennas are at a point where ground reflections interfere.

* Or, Some other source of external reflection may be present.

----------------------

The final scan with the Winegard CC-7870 coupler shows increased strength.
Most signals do vary if the station is not LOS and considerable distance
away. Could this be because the signal from the station varies with time?

Otherwise, it could mean that the Winegard CC-7870 coupler performs better
(less loss) than your previous combiner.

.

More baseline testing
 

Thanks for the image.

The location of the single antenna may be a better location with respect
to multipath (likely ground reflections).

Best current guess; the problem is external and related to height.

Height is often a consideration:

The following graph from posts starting here:

http://forum.tvfool.com/showpost.php...8&postcount=79

illustrates this height effect, to a small degree:


http://forum.tvfool.com/attachment.p...8&d=1411221833

Note:

The graph shows the average signal strength of 27 individual channels combined
versus height.

The heights in the image around 20ft and from 25 to 30 feet
showed evidence of multipath cancellation of signal.

The effects will be different for every location and channel, but is often present.

Mostly the interference comes from a ground reflection (which is delayed
due to longer path) being received simultaneous with direct path signal.
There will be heights where these two signals are in phase and other
heights where the two signals are not in phase.

This graph for three individual channels shows effects more clearly:


http://forum.tvfool.com/attachment.p...2&d=1411227205

---------------------------

If the horizontal stack were mounted at the location where the single
antenna is located, that could provide more insight.

.

Got the dual VHF moved up !
 

I almost didn't move the pair, rain was on the radar.
I definitely wanted to get it done though because I think we're going to have rain the next couple days.

I was only about halfway done when the rain started to fall. I couldn't haul both of the antennas up by myself so I took one of them off the crossbar. Next time I'll take both of them off the crossbar.


I wish I could have measured the spacing between the booms a little bit better but the rain began to fall quite a bit harder. They're kind of pigeon-toed right now judging by looking at the pictures.

So, I ended up plugging in the two VHF antenna array into the VHF half of my RCA PREAMPR1... I used the CC 7870 as the combiner. I'm going to have to make two new cables as they are stretched to their limit. I don't want to break the plastic baluns.

I took the ladder down and right after it stopped raining for a few seconds so I put the single 30-2476 back up on the mast in the lower position. The Spectrum analyzer pattern looks similar to when I had the pair down low like that so maybe it's roof reflections or ground reflections.

I don't think the signal of the two antennas is any stronger than the single antenna even when mounted up high. Still, there's room for improvement I feel.




http://forum.tvfool.com/attachment.p...1&d=1586654511

Antenna pics
 

Some pictures.

http://forum.tvfool.com/attachment.p...1&d=1586659854


http://forum.tvfool.com/attachment.p...1&d=1586659741

http://forum.tvfool.com/attachment.p...1&d=1586666112

2020-04-12 no changes better results. hmmmm...
 

Combatting losses
 

Interesting idea, using RG8 with RG11 connector.

Guessing from your image, probably RTV silicone would be a good candidate.
(Avoid silicone that contains acid, some use acetic acid, smells like vinegar)

Two things to consider:

1. Waterproofing
2. Impedance & Loss

Waterproofing, two approaches:

* Externally sealing the closed case. Any of the three items would probably work.

* Internally sealing. Also any of the three items, but there can be impedance
and loss considerations.

2. Impedance & Loss

If open internally, sealed externally, air dielectric, low loss, but may result in
high impedance due to wide spacing and small wire size.

Silicone has relatively low loss and dielectric higher than air. when using
relatively small wires, can help lower impedance.

For your combiner, the length of the wires internally are relatively short at VHF
frequencies, so impedance bumps (near the connections) may not cause a
noticable effect on performance.

------------------------------------

From another thread:

http://forum.tvfool.com/showpost.php...13&postcount=3

No new ideas.

For a Yagi, if other dimensions are suitable,
the length of a populated boom has most effect on gain.

For wideband Yagis, such as for upper VHF, director spacing (within limits)
has small effect.

Director length is sized to accommodate the highest desired channel.
If director is too long, gain on upper channels will be lost.

----------------------

The 30-2476 was designed for channels above the US VHF band, so the
directors could be slightly lengthened to obtain more gain in lower VHF
channels.

One can model such things with NEC2, but to verify the performance, can
require a very stable signal source (such as antenna test range).

The change in gain would probably be difficult to verify with reception
of time varying signals (broadcast stations).

If one is within LOS of broadcast stations and know the expected signal variations.
One could make many measurements of the modified antenna and compare the
average to the average of many unmodified measurements.

If the modification involved lengthening directors, and wished to preserve
coverage of upper VHF, it would be good to at least test gain on
channel 13.

.

Tripleo,

You said no new ideas but you did give me some ideas. Today is supposed to be a nice day today and I want to get some progress done.

I'm trying to gather enough knowledge and yours is very helpful. You mentioned wider directors helps out with the low end of the hi VHF band. I have wider directors from the two donor reflectors. I could use those towards the back.

First thing I want to do his do some baseline testing with what I have. I have two project antennas I could modify one I'll make it a little bit longer with more directors compare that to my original.

I've got to come up with a plan first though. I'll post my findings here.

I also plan to test my new lossless combiner! I'm kind of excited about that. Thanks to Calaveras over on AVS for the idea.

http://forum.tvfool.com/attachment.p...1&d=1587303108

If I recall correctly that is a design that Calveras developed.

Calaveras low loss combiner
 

Oh yes, I edited my post to give Calaveras credit. His instructions were if it was a solid polyethylene insulator to make it 10 inches exactly including the connectors. However I missed that important detail. I used foam insulated RG 8. Apparently that measurement should be 12 in. This is due to the velocity Factor.

These are things that escape me. In fact, 12" may not even be the exact length because every manufacturer publishes their own velocity specs. I don't know the manufacturer of this particular RG 8.

So, back to the drawing board on the lossless combiner. I'll be back.

I've made it to the Forum circuit, asking for advice. Luckily for me a couple of guys offered help over here Tripleo and over on AVS Calaveras.

Both agreed that the length of the boom is the most important. Calaveras suggested I would need to double the size of my antenna. Tripleo also gave me some great rule of thumb advice on the directors.

I tried to contact some people who specialize in antenna modeling but I was unsuccessful. Without any antenna modeling I decided to forge ahead and just try add four more directors on for a total of 16 I kept the last four directors the same size which was 21". I used one of the director booms and the holes were already pre-drilled to 10 and 1/2 inches spacing.

How does it work? I don't know yet. I had a storm creeping on me during the final mounting. I had to barely tighten up the mast clamp, wait for the rain squall the stop, jump back on the roof and try to re angle the antenna in the dark. The 40 mile-an-hour gusts made it difficult. I'll have to get back on the ladder tomorrow morning check my connections.

http://forum.tvfool.com/attachment.p...1&d=1587443372