Joining 4 OTA antennas?

[ppauper]

If I wanted to join 4 identical antennas, is it better to use a 4-way splitter or three 2 way splitters ?
and if 4-way is the answer, can someone recommend a good brand/model

[keef]

You'd need three 2-way splitters. As a 4-way may very well be three 2-way splitters internally, I doubt there's much difference.

If I were determined to join 4 antennas, I'd probably want to join/balance them in pairs, then join/balance the two pairs together; get each pair working well, then get the two pairs working together. Three 2-way splitters would give you that flexibility.

Note: I'd have to be determined. I don't know that I'd even attempt it without some pretty expensive testing and measuring equipment.

[roger1818]

Originally Posted by ppauper

If I wanted to join 4 identical antennas, is it better to use a 4-way splitter or two 2 way splitters ?

Will any or all the antennas be pointing in the same direction? This will affect the decision.

[stampeder]

I was thinking the same thing, Roger - is this for a diamond or quad array, ppauper?

Those 4-antenna configurations are almost always done with channel-cut Yagis for CATV/MATV head ends.

[ppauper]

Originally Posted by roger1818

Will any or all the antennas be pointing in the same direction?

all in the same direction.
It's a pair of stacked CM4228s, and I was going to try disconnecting the feedlines and using 2 separate baluns on each half of each 4228 (as in the 4228HD hack, to see if there was any improvement) so that I sort of have 4 identical antennas forming a square

[stampeder]

ppauper, Ken Nist says not to bother with a 32-bay (4 ganged 8-bays) and he says to read his 16-bay page first:

http://www.hdtvprimer.com/ANTENNAS/16bay.html

http://www.hdtvprimer.com/ANTENNAS/32bay.html

Note that at the bottom of the 32-bay page he mentions the CM4251 parabolic antenna, but he wrote that page before the various Gray-Hoverman variants came along, so you might want to investigate building something like a GH10.

[ppauper]

stampeder: It's 16 bay, I've got 2 x 4228's stacked, but I was going to split each 4228 into a pair of 4221's as it were making 4 x 4221

keef: you're right, 3 x 2-way versus 1 x 4-way. I was asleep I guess

[stampeder]

Ken's warnings would apply then. You'd probably hit the law of diminishing returns i.e. reception does not improve with effort expended. Stacking original CM4228s was commonplace, but what you want to do is not.

If you give it a try can you post some before and after signal strength results here?

[holl_ands]

Quad arrays need a low-loss Stripline Coupler (0.2 dB loss, irrespective of matching)
rather than a typical Hybrid Transformer (e.g. RF Splitter) which has 1-7+ dB of loss,
depending on antenna/combiner mismatch wrt amplitude and phase (no-way)...

For instance, if one of the antennas in the Quad Array has a dip in its frequency
response (could be a broken or bent whisker), then it will ALSO degrade the signal
on those frequencies from the OTHER antenna going into the first 2-way combiner.
And THAT imperfectly summed signal will degrade the 2-way combiner trying to
sum the output of the other pair of antennas...a double whammy....

Here is a summary of (mostly partial band) Stripline Couplers:
http://www.avsforum.com/avs-vb/showt...php?p=16072450

[ppauper]

it'll be a while before I try it (I'm on the road but thinking about OTA while I'm out of town)
I was thinking about this because if you look at the charts linked in the CM4228HD threads, the hacked CM4228HD with 2 baluns seems to outperform the original 4228 with 1 balun, so I was wondering whether doing the hack to the original 4228 would help.
I may instead buy what appears to be a (from NZ so probably $$$ for shipping) 1:1 balun I posted in the balun thread and again remove the feedlines and hook all 4 "4221s" together to a single 1:1 balun

[Lineloss]

Well I got it working. (see http://www.digitalhome.ca/forum/showthread.php?t=112071 for reference)

I did a fast scan with the top 4228hd connected to the 4 to 1 splitter; two of the inputs were left unconnected since I was only looking for a proper polar response at this point. Connecting both 4228hd's ,in phase this time, only seemed to help increase signal strength on my weak stations from Rochester as all the Buffalo and TO stations are coming incredibly strong. I was pulling in 45 digital channels pointing due south. When I have time I'll get around to a proper comparison. This setup is definitely smoking the single 4221 I used to have mounted in the same spot!

[Lineloss]

I was pondering if my stacked 4228's (lines removed and baluns added) are impedance matched properly as a whole (75 ohms).

I inserted a 1 to 2 splitter between the antenna and my pre amp input. I only used the splitter input and one output. I know there is some loss here, but the point was to double the so called 75 ohms antenna to 150 ohms and then to lower it to 37 ohms to see the results. Both connections should reduce my signal right? I know an antenna's impedance is not constant with frequency so I picked two middle "real" channels... When the antenna was assumed to be 150 ohms I got higher signal strength???

Any thoughts on this one?

I can't complete any full testing because "this ugly antenna thing is stupid" haha....yep you guessed it, the wife was helping me. I use the word helping loosely..hehe!

[ppauper]

so lineloss, does that mean we need to put a 2:1 balun on each antenna or an 8:1 balun ?

[Lineloss]

That would make it four 2:1 baluns.... Something tells me it may be between a 4:1 and a 2:1 somewhere, but this would need a much better investigation than I have done thus far! Hopefully this weekend

Must have more performance

[roger1818]

Lineloss, I am not convinced that leaving one input on a 2-1 splitter open will simulate either a 37 ohm or 150 ohm input. What the exact behavior is will depend on the design of the splitter you are using.