That isn't an easy proposition at UHF frequencies due to the short
I may combine the 2 signals at the antennae location or inside of the
house but in this case I must consider the electrical length or the
This will feed a High Definition TV set but before I start the work I
would like your opinion on the subject.
the antennas need to be spaced apart an odd # of 1/2 wavelengths from each
other as well, then combined near antenna.
It's also not an easy proposition if your various local TV stations are
broadcasting from different directions: If the two antennas are in-line with
each other, there's some finite delay between the two received signals
(measured right at the antenna's feedpoint) if they're "looking" at a
transmitter that's also in-line with the antennas, whereas there's no delay
when they're looking at a transmitter broadside to the line.
Hence you need to be able to adjust the phasing based on the frequency
(channel number) as well as the direction the transmitter is in. While this
is certainly doable, in most cases I think it's beyond the the point where
it's easier to instead just get one 3dB higher-gain antenna. (Also keep in
mind that if you get your antennas too close to one another -- within a few
wavelengths -- they may start coupling enough to significantly distort each
If there's only one particular station you want to tune in, though, then I'd
say, yeah, go for it!
Connect the two antennas together with a piece of transmission line.
Connect your feeder (main) transmission line to the center of the line
that connects the two antennas. They will be in phase.
Now the feedpoint of the array will be some impedance which will
probably not match the main line which will cause large standing waves
and therefore will result in large signal loss. In addition, it will be
rather frequency-selective as to where it works best and where it almost
doesn't work at all.
Your best bet is to do what Joel said in his last sentence.
Correction: Do as Joel said, get a 3dB better antenna.
You did not state in which country you are living.
If you live in a country with HDTV services distributed with COFDM
(DVB-T/DVB-T2) the best results can be obtained by moving the antennas
as far away from each other as possible (spatial diversity). Of
course, proper diversity receivers should be used (e.g.as in mobile TV
reception), but even a simple RF combining will help the situation, it
is just like a single frequency network (SFN) with multiple
transmitters on the same frequency sending identical data.
However, if you are unfortunate enough to live in a country with 8VSB
(ATSC etc.) apparently the nth generation equalizer are finally living
up to the expectations with a single antenna, but adding an other
multipath path, does not help the equalizers, thus, it is
questionable, if the second antenna will actually help anything,
** A two-way splitter can also be used as a combiner.
Must be fitted near as possible to the antennas and fed from two equal
If you are using RG59 co-ax, over 10 metres the loss at UHF is about 3 to 4
Consider changing to RG6 and you would gain 2 dB.
The main line is impedance matched at the TV reciever and so there are no
Using a passive combiner for the two antennas means you get a good match for
You a radio ham ?
** FFS - wake UP and read the OP's story !!!
There are two identical antennas, on the same mast, FITTED with a rotator
" have 2 UHF antennae of the same model installed on the same mast
with rotors and amplifiers .. "
Yes, although note that splitters like this often have insertion losses of a
dB or two as well... at which point it's becoming pretty questionable as to
whether or not the whole exercise is really worthwhile, IMO.
Good point about coax cable losses...
You are correct. My mistake.
There won't be an impedance match at the TV receiver (assuming that
the TV receiver terminates the line with its characteristic impedance)
and thus there will be no reflected wave from that point, and no
development of standing waves on the feedline between the receiver and
the central joining point.
However, it seem to me that there will be an impedance discontinuity
at the joining point, since you may have (e.g.) two 75-ohm feedlines
coming in from the antennas, wired in parallel, joining a single
75-ohm feedline down to the TV receiver. This discontinuity will be
able to generate a reflection back up towards the antenna, creating
standing waves... and if the reflection back towards an antenna is
itself partially reflected from the antenna's impedance (if the
antenna isn't a perfect match to its feedline) you can end up with the
signal ricocheting back and forth between the antenna and the joining
This can result in a loss of signal strength (due to cancellation).
Possibly more troublesome, the delayed reflections can degrade the
intelligibility of the signal... just as multipath reflections from
surfaces near the antenna can do.
In analog (e.g. NTSC) TV applications, these short-period reflections
within the antenna system may not be delayed enough to cause visible
"ghosting", but they can result in some amount of horizontal blurring
of the signal (e.g. what should be sharp transitions in the signal
brightness can be blurred out).
In his classic book "TV and other Receiving Antennas", Arnold Bailey
discusses the effect of these sorts of in-antenna and in-feedline
reflections at some length, and suggests some limits on how much delay
and reflection can be tolerated without unacceptable degradation of an
The effects are different on digital TV signals, of course... you
don't see horizontal blurring or ghosting... but you can see an
increase in the error rate due to inter-symbol interference. Whether
it's severe enough to result in a loss of reception (or undo the
benefits of the signal-strength increase that the OP was trying to
achieve), I don't know... probably depends a lot on the details of the
There is, I believe, an effective way around this problem. Don't just
hard-connect the feedline segments together. Instead, use a good
2-way signal splitter (the sort you'd use to connect a single antenna
feed to two different televisions), in reverse. Hook the "out" ports
to the antennas (via identical lengths of the same batch of feedline),
and the single "to antenna" port to the feedline going down to your TV
set. These splitters contain a "hybrid" which has the effect of
isolating the two "output" ports from one another and maintaining an
acceptable impedance, minimizing reflections.
I agree with others, though, that it's probably less trouble, and more
reliable, to replace a single TV antenna with a larger one having more
gain, rather than try to use two separate antennas and combine the
Dave Platt <dpl...@radagast.org> AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!
I think your first sentence "There won't be an impedance match at the TV
receiver (assuming that the TV receiver terminates the line with its
characteristic impedance)" is correct if you replace match with
mismatch. I'd bet that's what you meant.
Other wise, I agree completely.
Hams, especially the moonbouncers, stack antennas all the time. I'm
not really sure how they get around the loss of the Wilkenson
I'd just get a bigger antenna. The results would be more predicatable
since that larger antenna in theory has some test specifications as
is, while you would be hacking with your stack. If there is a
particular station that is an issue, you could get a yagi cut for that
station. Log periodic antennas have less gain than a yagi for the same
This is a good place to discuss gain versus aperture. Take a half wave
antenna. It has a specific gain. As it is scaled for higher frequency,
the size of the antenna is reduced, but the gain stays the same. Now
field strength is measured is uV/m. The smaller antenna has less
aperture, so the amount of signal captured is less. That is why UHF
antennas tend to be multibay, i.e. get more aperture to keep the
signal strength the equivalent of a VHF station.
Most preamps are crap. The good news is they are cheap crap. For the
preamp to be very useful, it needs a noise figure better than your
I have a number of Chip's preamps. Not cheap and you have to wait for
delivery, but good stuff. He builds customs if there is something
close to what he sells but not exactly what you need.
If you have a lot of cable loss, you might consider one of those
Silicon Dust networked receivers. You could put the box right by the
antenna, though indoors unless you put it in a NEMA box.
MythTV and XMBC can interface to the box.
> the best results can be obtained by moving the antennas
>as far away from each other as possible (spatial diversity).
Such reception is only "true diversity" if it feeds a "diversity
receiver". Otherwise, there is no discrimination done by the receiver.
True diversity would choose the best signal from whatever array of
antennas it has. Not necessarily the strongest signal. A good diversity
receiver chooses the "best" signal, and strength is not the only factor.
>, but even a simple RF combining will help the situation,
Are you sure?
Where are you relative to the stations? Do you have Line Of Sight or
are you hidden behind buildings or hills? What antenna(s) are you
using? You say you have a rotator but that can be a big PITA if you
want to watch the east station but record the north station. Look up
'jointenna' which are channel selective combiners. Not cheap but you
end up with a system that all channels are available simultaneously
like a cable company.
Also check into DVRs built from PCs. I have 3 of those sometimes all
recording at once.
I reckon changing the cable would be his best bet unless he *really*
wants to play around with a twin antenna interferometer just for fun.
The lobe pattern of combined sensitivity will peak along the
perpendicular to the spar length D separating the two aerials and in the
direction that they point. The fringe pattern has a pitch proportional
to lambda/D. This always makes pointing more difficult.
And if the cable lengths are even slightly different then the main lobe
will shift with frequency which could be somewhat annoying.
Placing the aerials too close together also allows additional local
interactions that will be hard to predict but should be modest for Yagis
pointed at the transmitter. A vertical separation might be helpful if
you have dynamic multipath problems from low flying aircraft, but it
would require some experimentation.
I think the first question for the OP is why does he want to do this?
I doubt it is worth the effort for optimistically a measly 3dB gain
(when 2dB can be had reliably with a better cable run).
> I doubt it is worth the effort for optimistically a measly 3dB gain (when
> 2dB can be had reliably with a better cable run).
** The OP is already using mast head amplifiers.
" I have 2 UHF antennae of the same model installed on the same mast
with rotors and amplifiers "
>On Thu, 02 Jun 2011 02:26:37 +0300, upsid...@downunder.com wrote:
>> the best results can be obtained by moving the antennas
>>as far away from each other as possible (spatial diversity).
> Such reception is only "true diversity" if it feeds a "diversity
>receiver". Otherwise, there is no discrimination done by the receiver.
Apparently you do not understand how COFDM based Single Frequency
Networks (SFN) work. Multiple transmitters quite close to each others
are used, each transmitting exactly th same signal at exactly the same
frequency with accurate timing. The coverage areas overlap and it
possible that a single receiver takes some subcarriers from one
transmitter and some from other transmitter(s) if the signal from the
nearest station suffers from selective fading on some subcarriers.
From the receiver point of view, having two well separated received
antennas and simply combining the signal into a single coaxial cable
will look just like receiving from two separate transmitters in an SFN
This does not work in a single carrier systems like the ATSC used in
the USA, so such combining does not help. Even for that system, a
proper diversity receiver would help, if such thing exists in the USA.
With proper diversity receivers, the spatial antenna separation should
be as large as possible.
There has been discussions in this thread about impedance matching
issues when combining two antennas. While it might be possible for a
very narrow band system to directly connect the dipole terminals to
each other and then to a feedline, but the length of the feedline
would have to be a specific fraction of the wavelength to get correct
This approach can not be used with TV due to the huge bandwidth of the
UHF TV band. The practical approach is to use a balun at each antenna
and use ordinary 75 ohm TV coaxial cables with exactly the same length
from the antennas to a passive splitter/combiner and from the common
port then use any length of 75 ohm coax to feed the TV.
While in theory, stacking two identical antennas would give an
additional 3 dB gain, but in practice, due to the variation in
radiation pattern and the losses in the combiner, the practically
achievable gain increase is 2-2.5 dB due to stacking.
While the signal from the satellite is quite predictable and
fluctuates very little, the terrestrial signal fluctuates in many
ways. Most of the time, the signal level is close to the average
level, but then there are short dips (-5 .. -20 dB) in the signal
level, perhaps 1-10 % of the time. These short dips in analog TV was
not a big issue, since it only momentarily added some noise to the
If the signal drops below the digital receiver threshold, the raw bit
error rate will increase to 100 % (error burst). To combat this, most
digital radio systems use bit interleaving, thus, after
deinterleaving, burst errors are converted to random errors, which can
be corrected by an error correction code.
Depending on how long time span the bits are interleaved compared to
the length of the error burst, the signal might or might not be
recovered without any losses. If the error burst is too long (the
signal drop takes too long), the picture pixeliates and the sound is
While stacking closely two antennas in the same mast will no doubt
increase the _average_ signal level by 2 dB, while the signal is
already adequate most of the time, but during a -20 dB dip, stacking
would only result in a -18 dB dip in signal level. This might slightly
shorten the length of the error burst (and slightly helping the ECC),
I would not expect miracles from stacking two antennas close to each
One should try to find a better location for the antenna and increase
the height, in order to get a reliable, comfortable digital reception.
** You are a fucking lunatic.
Must be another retarded radio ham.
>Apparently you do not understand how COFDM based Single Frequency
>Networks (SFN) work.
Apparently you do not, since that it not the topic here.
>Most preamps are crap. The good news is they are cheap crap. For the
>preamp to be very useful, it needs a noise figure better than your
In terrestrial reception, the ground emits noise at about 300 K noise
temperature (noise figure 3 dB), thus the receiver noise figure does
not have to be that good. Receiver noise figures of 3 dB are not
uncommon, thus contributing the same amount of noise as the warm
These days, the main purpose of antenna mounted preamps is to
compensate for the feedline (and splitter and connector) losses. Thus
the preamp gain should be larger than the feed line losses. By
increasing the gain with a further 5-10 dB, the receiver front end
noise can be eliminated from the equation.
The gain should not be increased above this, since the generation of
spurious signals goes very quickly up, either in the antenna preamp or
in the receiver.
>I have a number of Chip's preamps. Not cheap and you have to wait for
>delivery, but good stuff. He builds customs if there is something
>close to what he sells but not exactly what you need.
I did not find any device directly suitable for UHF TV masthead
preamps. Perhaps the wide band 0.5 dB noise figure device would be
nice, but it would absolutely require proper high pass (450 MHz) and
low pass (850 MHz) filtering at the input, thus, the insertion losses
of these filters would have to be added to the system noise figure,
but still be well below 3 dB.
OTOH, for terrestrial reception, I do not see much point in trimming
off the last decibel. The signal variation due to various kinds of
fades can be more than 10 dB.
>I think your first sentence "There won't be an impedance match at the TV
>receiver (assuming that the TV receiver terminates the line with its
>characteristic impedance)" is correct if you replace match with
>mismatch. I'd bet that's what you meant.
You'd win! Let me know where to mail the nickel and the five vintage
box tops :-)
>Other wise, I agree completely.
Is it "antennae" or "antennas"?
A pet peeve of mine I guess.
May just be a US thing.?
My dictionary lists both, although at least in my experience people use
"antennas" for the things that send or receive EM radiation and "antennae" for
what's found on the top of insects.
Are you in the US?
Do you think extra height on the antenna can help?
Nothing beats line of sight coverage!
You should go away and attempt to plug all of your ofrifii with
> I have 2 UHF antennae of the same model installed on the same mast
> with rotors and amplifiers and I would like to combined the output
> signal in order to get of up to 3 dB of the signal gain.
> I understand that the signal must be in phase at the combining point.
> That isn't an easy proposition at UHF frequencies due to the short
> wave length.
> I may combine the 2 signals at the antennae location or inside of the
> house but in this case I must consider the electrical length or the
> 30' cable.
> This will feed a High Definition TV set but before I start the work I
> would like your opinion on the subject.
Just get a 2-to-1 splitter/combiner, and use two of the same length
pieces of coax. Plug these antenna leads into the combiner inputs and
the combiner output to the downfeed.
Probably not a good move. Firstly, best reception requires
freedom from multipath reception (i.e. your best steering
might be to null a reflection rather than to peak a signal strength).
So, a 'better' antenna is not one with higher signal power,
but with a well defined null direction.
Secondly, a good four-element Yagi has one dipole and
three directors or reflectors; two two-element antennae
is not similar to a good four-element one. It's a rather small
gain for the cost of multiple antennae and downleads.
There is a fairly good bargraph signal
strength indicator in most modern TVs; look at it
carefully, 3 dB is a TINY bit of difference.
** Not much of a problem with digital TV signals.
> Secondly, a good four-element Yagi
** Ever seen a UHF TV antenna for roof top use ?
Usually have dozens of elements and are very directional.
Off axis, reflected signals are ignored.