Phased Delta Flag Arrays.pdf

Phased Delta Flag Arrays

Dallas Lankford, 2/21/09, rev. 12/31/09 (with later minor revisions)

This article describes the development of my quad delta flag array. The invention of my quad delta flag array is described

in my article “Phased Flag Arrays.”

The delta flag antenna is a variant of the flag antenna. The earliest delta flag antenna I am aware of was designed by K6SE

and constructed by ON4UN for use by FO0AAA some time before June 2000. While it was designed for the 160 meter ham

band, it probably worked well in the MW band. I became interested in phased delta flag arrays because I wanted to

experiment with quad phased flag arrays for splatter reduction in the MW band; see my article “Phased Flag Arrays.” My

flag array experiments

were inspired by the

phased rotatable 160

meter band dual flag

arrays of NX4D and

N4IS , but inexpensive

masts are not good for

MW flag construction

because of sag

problems. A delta flag,

on the other hand, can

be kept taught with

only one mast and two

ground stakes at the

ends of the base of the

triangle. As a matter of

fact, dual flag arrays

are not cheap mast

friendly either, so a

dual delta flag array is

a better choice than a

dual flag array. At right is a photo of one of the delta flag antenna elements which

were used in a quad delta flag array. The mast was a telescoping 20' BREAM

STIX BS20 fishing pole, with the top two elements removed. The base was a

0.5” ID galvanized conduit. Clear flexible 0.5” and 1” tubing was split and

slipped over the end of the conduit, and the butt of the fishing pole was slipped

over the 1” flexible tubing. The fit was tight, exactly what was wanted. A ground

clamp was installed snug against the bottom of the tubing so that the clear tubing

would not move when the butt of the fishing pole was installed. Details of the

conduit, ground clamp, and split clear tubing (all bought at Lowes) on which the

pole was mounted are shown in a photo below. To prevent the telescoping pole

from collapsing, cable ties were used just above the end of each section.

Pole Clamps

Cheap telescoping fiberglass fishing poles like the BREAM STIX BS20 tend to

collapse when relatively little weight is attached to the top. I decided to try nylon

cable ties under tension (at the joints). With the top two sections removed, there

were three joints to be secured. I used a 6” cable tie (Radio Shack) for the

smallest joint, and 8” cable ties (Lowes) for the larger joints (one of them shown

in the photo above right). This may not be the best long term solution, but it

works well for the short term. It was mentioned on a hams web site that stainless

hose clamps work well for this purpose. However, I decided not to use that

approach because the fiberglass near the joint could be cracked if the hose clamp

is tightened excessively. To increase the cable tie tension if necessary or desired,

slip the cable tie up the pole section several inches, maybe more, pull the cable tie

tight, and slip the tightened cable tie back down.

LC Delay

When I began to consider testing a quad flag array with its potentially better nulls, the

prospect of multiple coax delay lines was not attractive. In theory, two capacitors and an

inductor can be used to do the same thing as a long length of coax, provided the right power

combiner is used. The first time I tried the LC delay circuit with the combiner used

previously for the coax delay circuit, the LC delay circuit was a failure... the nulls were

variously unstable or not as deep as they should have been. So a new combiner based on a

schematic in the 1992 MiniCircuits RF/IF Designer's Handbook was designed. After the

new combiner was tested, the LC delay circuit worked very well with dual flag arrays, and

later with dual and quad delta flag arrays.

Note that the LC delay phaser has no controls. The quad (or dual) delta flag array is

optimized for maximum splatter reduction by orienting the array. It does not matter if the

array maximum is not pointed exactly in the desired direction because the beam width is

quite broad. The goal is to orient the array so that as many undesired signals as possible are

nulled as deeply as possible.

The time delay T in nanoseconds along a ray with arrival angle θ connecting two antennas with centers spaced a distance s

apart in feet is T = 1.02 s COS(θ) (nanoseconds). For a 30 degree arrival angle and 70' spacing T = 62 nS. Previously this

was converted into a length of coax to provide the necessary delay for phasing. The coax length has been replaced by the

LC delay circuit at right, which resembles a low pass LC filter, and used in the dual and quad delta flag arrays discussed in

this article. Its input and output impedances Z are the same. For a 50 ohm system, such as the dual and quad arrays, take Z

= 50 which gives 2500 = L/C, or L = 2500 C. Taking T = 62 x 10^–9, which was calculated above, both sides of the time

formula at right are squared, namely 3844 x 10^–18 = LC, after which substitution of 2500 C for L by the equation above

gives 3844 x 10^–18 = 2500 C^2, or C = 1240 pF. Thus C/2 = 620 pF, and L = 2500 x 1240 x 10^–12 = 3.1 μH. The

capacitors should be mica, and the inductor may be two parallel Miller 6.2 μH inductors, Mouser 542-4610-RG. Or use FT-

50-61 toroids and an accurate inductance meter to make the required 3.1 μH inductors. L and C/2 values for other

frequencies can be obtained by multiplying the values for 70' spacing by the ratio of the spacings. For example, for 100'

spacing, L = (100/70) x 3.1 = 4.4 μH, and C/2 = (100/70) x 620 = 886 pF mica capacitors.

LC Phased Dual Delta Flag Arrays

The pattern of a 100' spaced dual delta flag array at 700 kHz is shown in the figure at right.

The schematic and diagram of a dual delta flag array with 100' spacing is shown on the

next page. For insensitive receivers 10 dB or more of preamplifier gain may be needed.

For example, when using Perseus at Grayland with the quad delta flag array an additional

20 dB preamp was definitely needed. Figures later in the article contain diagrams and

information for quad delta flag arrays. If you don't have the space for a quad delta flag

array, or simply don't want to deal with the complexity of a quad array, the dual array may

be adequate in some cases even though its null aperture is not as wide as the quad array.

As with dual flag arrays, the delta flag arrays are not intended as general purpose null

steering arrays, but rather as wide null aperture fixed arrays for use at coastal DX sites to

reduce splatter throughout the entire MW band. The original dual delta flag array was

implemented with coax delay (phasing) like the predecessor flag arrays; see my “Phased

Flag Arrays”article. Later it was replaced with an LC delay circuit and an improved

combiner as described above. The two 100' long twin leads (speaker wire) in the

schematic and diagram on the next page are part of the signal delays. So the two twin lead

lengths must be equal if you use lengths other than 100'.

Design Objectives

It should be mentioned that the ham versions of these antennas have different design objectives than the MW versions. For

example, NX4D and N4IS have designed their dual flag arrays for maximum RDF in the 160 meter ham band which

maximizes the signal to man made or atmospheric noise ratio. This is what is needed for 160 meter CW DXing because

narrow bandwidth (~20 Hz) CW sensitivity is generally limited only by man made or atmospheric noise. However,

maximum RDF is generally not a high priority for MW DXers. What usually limits MW DXers ability to hear foreign splits

at many MW DX sites is splatter from adjacent domestics or domestics themselves. Thus maximum splatter reduction and

attenuation of domestics due to wide aperture nulls is the MW analogy to maximum RDF for the 160 meter band. Both the

dual and quad array patterns in

this article and in other articles

in The Dallas Files have been

designed for maximum

attenuation over wide (dual

array) and extremely wide (quad

array) null aperture angles.

These arrays are effective

mainly when they are used at a

coastal MW DX site for which

most undesirable signals are

“behind” (in the null of) the

antenna, such as at Quoddy

Head, ME and Grayland, WA.

Using one of these wide

aperture null arrays is of

virtually no benefit for me here

in North Louisiana when trying

to hear TA's or TP's because

there are so many undesirable

signals coming from the same

direction as the desired signals.

The First Quad Delta Flag

Array (QDFA) Test

A 70' spaced QDFA became operational about 4 pm CST 2/20/09 using a coax delay phaser. 70' spacing between centers

was used because a 100' spaced quad array would not fit on my lot. After many hours of testing this short version of the

QDFA appeared to be operating correctly. At some times and for some signals, a dual flag array used for comparison

produced deeper nulls. But more often than not the quad delta flag array produced nulls as deep as the dual array or deeper.

The front to side ratio of the quad array (at +90° and –90° from the maximum axis of the array) was noticeably better than

the front to side ratio of the dual array, by about 10 dB. The delta flag array performed the same with both coax delay

phasers and LC delay phasers. Note that for the 70'spacing C = 620 pF and L = 3.3 μH. On nights when signals to the

North (the maximum null direction) were stronger than normal and signals to the South were weaker, the dual flag often

had better nulls at lower MW frequencies. But when signals to the south were stronger, nulls of the dual flag and quad delta

were often about equal. It was determined that this is normal, assuming that

EZNEC accurately models patterns of flag and delta flag arrays. The

schematic below and the pattern at right describe the quad delta flag array

which was taken to Grayland. Note that the spacing of the Grayland QDFA

is 100' and its LC values are different from the LC values for the 70' spaced

QDFA. Note also that 300' of twin lead from the phaser to the receiver is

used because of placement of the Grayland QDFA. Component tolerances

need not be 1% or better, but rather matched to within 1%.

Non-standard QDFA Phasing

Normally a quad delta flag array would be implemented by spacing the delta

flags 100' between centers, phasing the 1st and 2nd pairs identically (say, for

a 30 degree elevation null) in the standard way (delay equal to the spacing

along the 30 degree elevation null between two vertical lines spaced 100'

apart + anti-phase [180 degrees]), and then phasing the two pairs as if they

were two single antennas twice as far apart (also for a 30 degree elevation

null). However, EZNEC simulation shows a disappointing 120 degree 30 dB attenuation aperture for such an array...

hardly worth the effort compared to a single pair of phased delta flags. But as was the case for quad flag arrays, if the

phasing between the two delta flag pairs is the same as the phasing between each adjacent pair, then the 30 dB null aperture

is 150 degrees. There is about a 3 dB additional loss for this “non-standard” phasing compared to the “standard” phasing,

but that seems like a small price to pay for an additional 30 degrees of 30 dB or more null aperture.

Grayland DDFA And QDFA Tests April 19-22, 2009

The 100' spaced QDFA

was tested at the

Grayland Motel MW

DX site near the

Pacific Ocean in the

State of Washington.

An aerial view of the

Grayland area and the

quad array location is

given in the figure at

right. The DDFA was

in the same location,

but half the length of the QDFA. The Grayland aerial map shows the quad delta array pointed due West. However, for

optimum splatter reduction the array should probably be rotated 10 degrees to the north of west for better attenuation of

southern California stations.

Winner!!! (after a field change)

At first there was a problem: the QDFA had degraded nulls for the lower frequencies of the MW band. The problem was

eventually determined to be due to the untested long 300' twin lead which was required at Grayland because of the QDFA

placement. But where a fair comparison could be made, namely in the upper half of the MW band, and especially the upper

third of the MW band, the QDFA was clearly superior to a (dual ALA-100) Wellbrook array. Some DX was heard on the

QDFA which was not heard at all on the Wellbrook array. Other DX was merely heard better on the QDFA. A more

detailed account of the problem and the “fix,” and some of what was heard on the QDFA are contained in the article

“Grayland Quad Delta Flag Array Report” in The Dallas Files . A short account of the problem and the “fix” is given below.

The “fix” below is different from the “fix” which was done at Grayland because some of the impedances for the “fix” were

not matched correctly in the “Grayland fix.” In the “fix” below the impedances are matched correctly. In any case, after the

long 300' twin lead was isolated from the QDFA with a push-pull Norton transformer feedback amplifier, the QDFA

performance was outstanding.

The Fix

The schematic on the next page shows a fix similar to the one which was

done at Grayland to make the QDFA fully operational for the lower

frequencies of the MW band, namely a push-pull Norton transformer

feedback amplifier to isolate the 300' twin lead from the phaser. The photo at

right shows one version of the fix. The Norton amp is attached to the “front”

side of the phaser which contains the 12 volt DC feed via Pomona gold

plated banana jacks and the output to the 300' of twin lead also via Pomona

gold banana jacks. The “rear side” contains 8 banana jacks for 200' twin lead

each to the 4 delta flag elements. The transformers and LC phase delay

circuits are mounted on insulated standoffs. Faucet washers were used for

strain relief on the back sides of the Pomona banana jacks after several rear

insulators shattered during initial installation. All banana connectors are gold

plated to eliminate oxidization. Everything fits neatly inside a Hammond

1590E cast aluminum box. The Norton amp PC board was professionally

manufactured by ExpressPCB. This is the phaser which was made for the

beta test site at Kongsfjord. It was retrofitted with gas discharge surge

arrestors for static and other transient voltage protection after it failed at

Kongsfjord at the end of October 2009, and has been modified to a LIN with

14 dB gain and 0.9 dB noise figure vs. 10.3 dB gain and 1.7 dB noise figure

for the standard Norton.

Dual Delta Flag Array Test

As an intermediate step in fixing the QDFA, for the night of 4/21 the array was operated as a dual delta flag array (DDFA)

with two of the four delta flag antenna elements removed. All receptions in the Grayland log dated 4/21 were made with the

DDFA. In the log (which follows) you will find a number of good receptions with the DDFA, including the DXAM

Philippines 1017, VOA Philippines 1170, Malaysia 1475, Micronesia 1503, a few Hawaiians, and lots of Alaskans. If you

don't have space for a QDFA, you should consider a DDFA. A DDFA (or DFA) will work almost as well as a QDFA.

Nearby Antennas

Nearby antennas, especially beverages, can spoil

the pattern of a QDFA or DDFA. EZNEC

simulations at right, for an unterminated beverage,

with open circuit input, illustrate some of the

QDFA patterns for nearby beverages. For a

beverage connected to an independent 450 ohm

source, the QDFA pattern is not degraded as much.

However, it is not known if these EZNEC

simulations give an accurate description of the

skewed QDFA patterns. As a rule of thumb I

would recommend that no part of a beverage

antenna be closer than 300 feet to a QDFA or

DDFA, and even that may not be enough. I do not

know of any way to settle these issues. Other

antennas may not degrade the QDFA pattern as

much as beverages, but who knows?. Personally, I

would not use any other antenna near a QDFA.

EZNEC simulation also shows that nearby power

lines can significantly degrade a QDFA pattern,

especially for a power line at one end of and

perpendicular to a QDFA.

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: