5

Is it possible to create a TV antenna that is composed of smaller antennas that are pointed to different broadcast centers? So in essence, many antennas for just a single digital TV.

Is it possible without any problem? or It is possible but with some problems?

Do antennas for digital TV still need to be tilted to the right directions?

Abel Melquiades Callejo
  • 153
  • 1
  • 1
  • 6
  • What kind of antennas and components (passives, BJTs/FETs, op-amps, etc.) are you using or want to use? Also, why use multiple antennas instead of an "omni-directional" antenna? – Shabab Oct 16 '13 at 21:12
  • What are you trying to achieve functionally? – Andy aka Oct 16 '13 at 21:46
  • Another important factor when combining two TV antennas is the coax cable running from each antenna to the antenna coupler/combiner should be of the same length. This will reduce the chances of a weak signal from one antenna interfering with the strong signal on the other antenna. The weak signal will still be present but it will phase properly with the strong signal. –  Dec 14 '13 at 10:18

2 Answers2

7

The electromagnetic wave and antenna don't know if the TV is analog or digital. To them, it's the same! This means that most of analog transmission experience applies to digital TV as well.

So in general, you should point the antennas correctly for best signal quality and use proper polarization as well.

You can use multiple antennas for best signal quality, but there are some bad sides to that as well. Let's say you have an antenna correctly set up for one signal source. It is designed for its frequency, has sufficient gain and is correctly pointed at the signal source. That antenna will still pick up signals from other sources as well, but they will be weaker. If you have another antenna matched to the weak signal source of the first antenna and you combine them both directly, first antenna's weak signals can interfere with second antenna's strong signals. Usual solution to that is to use a diplexer. It's a device that has a band pass filter (some implementations may just use low pass and high pass filters) and you set it up so that it passes through the frequencies which a particular antenna is receiving the best and blocks other signals. This way, at the output of diplexer you'll only have the best signals from all antennas you have and weak signal from one antenna won't interfere with a strong signal from another antenna. Of course, diplexers have their own insertion losses, so they'll decrease the signal quality. The idea is to use as few antennas as possible to get the best signal and to use diplexers sparingly in order to prevent their insertion losses from having a negative net effect on signal quality.

Some digital TV standards allow use of "auxiliary" broadcast locations so that you have one main transmitter and few smaller transmitters connected to the main transmitter. In such setups, the signals from the smaller transmitters shouldn't interfere with the signal from the more powerful transmitter. Unfortunately, I don't have any practical experience with this, so I won't go into any details.

The other important thing is the digital cliff and analog slope. In digital TV systems, as long as your signal quality isn't horrible, you'll get relatively good picture. Problems will occur once it gets borderline and then quickly after that, you'll completely lose signal. You can take advantage of that to have good picture quality even with not so good signal, but the downside is that you'll be more sensitive to bad weather than if you had good signal quality. On the other hand, in some situations setting up a proper TV antenna system can require expensive antennas, cables and expertize so exploiting the digital cliff effect can be economically justified.

AndrejaKo
  • 23,261
  • 25
  • 110
  • 186
  • that pretty sums up everything. great! – Abel Melquiades Callejo Oct 16 '13 at 22:05
  • I wonder if it would have been practical to have a small portion of the time bandwidth for each channel devoted to a pulse which was preceded and followed by a certain amount of dead time, such that TV receivers could have shown an "oscilloscope view" of the input signal at that time? On old analog sets, it was useful being able to watch the screen while adjusting an antenna to see how different adjustments would affect different kinds of ghosting, noise, etc. I've never seen a digital set with anything nearly as nice. – supercat Oct 14 '15 at 21:01
  • @supercat Switch to cable or competition from professional TV installers or just a perceived lack of profit from that feature? Or could it be that there's no nice way to get statistics from the demodulator inside of the TV? I know that some LG TVs do show BER, which is a good tuning indicator. On some other TV sets, the update rate for BER is too low. It would be nice if we could see the raw spectrum on the screen and be able to tune antenna using that, but again that would depend on what is actually available to the TV's OS. – AndrejaKo Oct 15 '15 at 00:09
  • For suggested TDMA system, it just seems way too complicated to actually affect transmitters like that and then bolt on something that would just be maybe useful for tuning antenna. I mean, internally TVs would already have adaptive filters for channel equalization and so on, so there should be information already inside of the unit on which we could base tuning. – AndrejaKo Oct 15 '15 at 00:11
  • @AndrejaKo: I don't know how the modulation works, but in many kinds of digital wireless system data is divided into packets, and receivers are essentially agnostic as to what goes between packets. Inserting dead time before a packet would thus reduce the effective available bandwidth by a time proportional to the amount of dead time (and e.g. 16us every 16ms wouldn't be much of a loss). In the days of analog television, I remember that there would sometimes be multiple ghosts that would come and go as the antenna was rotated, and... – supercat Oct 15 '15 at 15:08
  • ...the trick was to find a spot where no ghosts were too big. From what I understand of digital television, one of the biggest sources of reception problems is multi-path interference; a "quality" indicator may regard a signal with four ghosts at -10dB as inferior to one with a single ghost at -8dB, but I would think the former would be more resistant to weather-related interference. – supercat Oct 15 '15 at 15:12
  • @supercat But with modern systems multipath isn't as big issue. In fact, in DVB-T/T2 systems using COFDM, it is actually helpful. There are even synchronized single-frequency networks that broadcast on the same channel and signals add up. Packets are used, but there is no free air-time between packets. Instead, they are continuously sent out and the receiver if always getting new data, so there isn't any free time to get timing impulses. Also, there's big pressure to provide any free bandwidth to error correction coding, which would be helpful even when the antenna is correctly tuned. – AndrejaKo Oct 15 '15 at 15:53
  • @supercat As far as I understand it, in 8VSB systems, equalization is used to get rid of multipath and it seems that its proponents seem to believe that it's sufficient solution, so again we end up with 'virtually' no multipath scenario. – AndrejaKo Oct 15 '15 at 15:55
  • @AndrejaKo: How well do such approaches deal with changes in signal paths? Are there any tuner designs that can make use of multiple spacially-separated antennas to improve reception? For both OTA TV and satellite, I would think that if one didn't mind having more than one antenna, the probability of random weather phenomena simultaneously affecting antennas separated by a few meters would be much lower than the probability of affecting just one. – supercat Oct 15 '15 at 17:49
  • @supercat Well the idea in COFDM is that we have lots of very low speed carriers. They are so low that ISI from major multipath components shouldn't be an issue. As for tuner designs, I do not know of any commercially implemented tuner designs that allow for multiple antennas, but I do know that at least in DVB-T2, standard does provide option of having receivers with multiple antennas. The keyword for research is DVB-T2 SIMO. – AndrejaKo Oct 15 '15 at 22:51
  • There is research in that field, for example this paper: http://www.inue.uni-stuttgart.de/forschung/veroeffentlichungen/pub_2013/ahmed_ITG_Feb.pdf Also it seems to be a popular idea for use in mobile devices, since mobile extensions seem to be something relatively interesting here in Europe. – AndrejaKo Oct 15 '15 at 22:53
  • There's also the Alamouti scheme used for transmission sites, where two antenna systems are used and signals are combined in a single receiver. Here's a whitepaper about that: http://www.lstelcom.com/fileadmin/content/marketing/flyer/general/LS_WhitePaper_DVB-T2_en.pdf Unfortunately, I do not know of any research efforts being done in the USA on these topic. – AndrejaKo Oct 15 '15 at 22:57
-2

First of all. unless you have the math, you should avoid combining two antennas to one coaxial down-lead. Even so, if the signals are weak and need boosting, the ALWAYS use a mast-head preamplifier. NEVER ever use a preamplifier next to the receiver. All that will happen is that you will amplify the noise of the coaxial cable to the detriment of the wanted signal. Phasing of coaxial lines to two antennas is very complex and can get you into serious technical trouble.

I would make the suggestion that you buy two mast-head preamplifiers.

Buy two new runs of good AIR CORE DIELECTRIC (lowest loss and less noise) coaxial cable. Never skimp on VHF or UHF coaxial cable, it is false economy to buy poor cable that is cheap. You end up paying twice. Cut the coaxial cable to multiples of an ODD HALF WAVELENGTH. The math is simple- 299792458 / Carrier Frequency x 0.66 /2 = XXXXXXXXX gives you the half wavelength in meters. Example 299792458/560MHzx0.66= 353.3268255cm (full wavelength) /2 = 176.666341275 cm (1/2 wave) .

Measure your run from the mast-head preamplifier to the receiver. Let's say it is 10 meters. 10 x 353.3268255= 353.3268255 meters x 3 =10.597104765 meters plus your half wavelength of 176.66634127 cm = 10.5971 meters or 34.76728 feet (417.1/4 inches). I rounded this up a little as makes no difference. You now have a true 75:Ohm match to both the preamplifier and the receiver. The losses will be minimized and the noise factor (with the preamp at the mast head near as damn it, 0). The preamplifier drowns out the cable noise with a strong captured signal. You have two antennas and to cable runs. Next buy a cheap and cheerful coaxial relay you place next to the receiver and switched via your remote control. Viola!

One word of caution. You only need enough signal to give a fully quieting signal to the receiver, that is just enough plus a little extra to overcome the natural cable losses, that is, about 10db of gain MAX, okay?

Otherwise you will overload the receiver and you won't like what you see. If you would like to give me the exact measurements of the cable run and the frequency the two transmitters that you will find on Google for your two stations and I will send you the trimming details.

Please leave a comment on this answer.

Community
  • 1
Tony Bradbury
  • 1
  • 3
    This wall of text is virtually illegible. Some line breaks would make it slightly better. – David Oct 04 '14 at 08:12