Quote:
|
Would it be a uvsj or a diplexer?
|
A UVSJ is a diplexer. No, that is used to combine a UHF antenna with a VHF antenna.
You can combine the two 91XGs with a splitter used in reverse as a combiner. The two pieces of coax from the antennas to the splitter must be the same length. This is what works for two antennas, that are identical and both aimed in the same direction, that have built-in baluns with coax output.
Some people think that using a splitter as a combiner would result in a loss of 3.5 dB, which would make two antennas worse than one. When used as a splitter, the loss is 3.5 dB. When used as a combiner the loss is about 0.5 dB, giving you a gain of 2.5 dB for two antennas IF they are identical, properly phased together, aimed in the same direction, and the wavefront presented to the antenna array is uniform (which you might not have if there are trees in the signal path).
http://www.avsforum.com/forum/25-hdt...l#post12978153
Quote:
It needs to be remembered that if a splitter that measures 3.5 dB on the split ports is used as a combiner with identical and properly phased antennas, that the loss is only 0.5 dB. IOW, you'll get 2.5 dB gain by phasing 2 antennas, which is the number most typically quoted for phasing 2 antennas.
I have a pair of phased VHF antennas and a pair of phased UHF antennas using the ferrite splitters as combiners. The splitters measured 0.4 to 0.6 dB loss across the bands. I used a Fluke 6062A signal generator and a Boonton 9200 power meter for the measurements.
|
also:
http://www.avsforum.com/forum/25-hdt...ml#post5768148
http://www.avsforum.com/forum/25-hdt...ml#post8101234
http://www.sandiegohdtv.org/discus/m...tml?1181781042 (Posted on Monday, March 26, 2007 - 05:03 pm)
For antennas that have 300 ohm output, you use a 4:1 matching transformer (balun) for each antenna and then run the two coax lines to the splitter. It is necessary to correctly phase together the two antennas so that they work in unison and their outputs are additive, so you have to reverse the 300 ohm wires on ONE of the baluns to see which way works the best. When the antennas are NOT correctly phased together you get two peaks as you rotate the antenna, one each side of the direction you want, with a null between at the center. Another way to say it: The main peak is split in two. Since your two 91XGs probably will have built-in baluns, you don't need to worry about this phasing test.
The output of the splitter/combiner goes to the input of the 7777.
My own experience is with stacking two original CM4221 antennas side-by-side. They were 300 ohm output, so I used two baluns and a splitter in reverse. I got a gain of 2.5 dB (over one antenna) as measured with my Sadelco signal level meter. Theory says max possible gain is 3.0 dB, but there are losses in the combiner. It is possible to build a lower loss combiner, but I don't think it's worth the trouble, unless you need every 1/10 of a dB you can get. Lindsay used to sell one for about $100. This is what the combiner would look like. These are for 50 ohms, but could be made for 75 ohms by changing the size of the inner conductor:
http://www.gare.co.uk/6m_antenna/divider.htm
http://www.qsl.net/dk7zb/Stacking/splitter.htm
http://home.teleport.com/~oldaker/power_dividers.htm
http://www.packratvhf.com/power_di.pdf
I took another look at the Calaveras antenna diagram:
http://images.aa6g.org/AVSForums/Antennas.jpg
I'm guessing, but it looks like he is using an ingenious way to reduce combiner loss. Just before the preamp you see the notation 1/4 Wave Matching. If he has connected the two coax lines from the antennas together in parallel, you get 37.5 ohms. If you use a 1/4 wave section of coax between that point and the input of the preamp, it can convert the 37.5 ohms to 75 ohms. Calaveras didn't make a housing for his combiner, he just joined the coax lines together in a "Y" configuration with the center conductors soldered together and the braids soldered together.
The proper impedance for that matching section is given by this formula: the square root of the product of the two impedances to be matched. In this case it is the square root of 37.5 times 75, which is 53 ohms, close enough for 50 ohm coax. The disadvantage of this matching technique is that it is narrowband.
http://www.antenna-theory.com/tutori...nsmission5.php
http://www.qsl.net/w4sat/qtrwavtr.htm