TV Fool  

Go Back   TV Fool > Over The Air Services > Special Topics > Antennas

Notices

Reply
 
Thread Tools
Old 10-Jun-2013, 5:53 PM   #21
tripelo
Senior Member
 
Join Date: Jan 2010
Location: Dallas, TX
Posts: 122
Loss of Gain of S/N, Combining Two Antenna

Quote:
Originally Posted by Pete Higgins View Post
A compelling advertised feature of the HDB8X is its ability to align the 4-Bay panels to receive broadcasts originating from different directions.

Combining the two antennas resulted in the total loss of 9 LA stations and dramatically reduced the others. ...nearly useless for leveraging its advertised multi-direction feature.
Thanks Pete, for testing this feature. I’m sure that you know this:

There are at least three fundamentally different ways of combining all channels from two antennas. Each involves two or more significant sources of S/N loss.

In all cases, if the antennas point in different directions there is:

- Loss of the stacking gain (up to ~ 3 dB) that the antennas might have (assuming transmission lines matched).

Additional Loss for Each Case:

Case 1. Separate antennas pointing different directions connected to a hybrid combiner (usual splitter reversed) have:

- Loss through the combiner, typically more than 3.5 dB at UHF

Note: When the signals at the combined (input) ports are non-identical, hybrid combiners have 3 dB additional loss above normal circuit losses. Not-identical signals occur when antennas point toward differing stations.

Case 2. If the signals are combined with transmission line techniques that avoid the 3 dB hybrid combiner additional loss, then there is no isolation between input ports.

Without the isolation; The second antenna loads the first, and re-radiates away ~3dB of signal, vice-versa for the first antenna loading the second one.

Case 3. If both antennas have their separate matching preamplifiers prior to combining, then there is isolation between input ports, so:

* No loss thru re-radiation,
* No S/N loss through combiner (single signal input to output)

This is because both signal and noise have been raised above noise floor by preamplifier, so combiner loss in effect just lowers gain of both signal and noise. Not affecting S/N ratio.

But, now there is:

- Twice the noise, compared to one antenna, one preamp, and no combiner.

This is a result of each preamp adding equal noise power above noise floor. Two independent noise sources adds at combiner output.

Depending on preamp noise figure, this Case 3 could be better, or much worse than the first two cases.

Choose your poison.

Each method has some advantages and disadvantages.

Note: Matching two transmission lines is most important when antennas face same direction. When antennas face different directions, phase control of the two signals is usually lost, therefore matching transmission lines typically does not provide gain.

Note: Further in this thread is attempt to add more detail.

-----------------------------------------------------
Diplexers such as a UVSJ, or a Join-tenna, get around some of the S/N losses mentioned in Cases 1 thru 3 by providing lower-loss frequency dependent isolation.
-----------------------------------------------------

Quote:
I suppose one advantage it still has over a conventional 8-Bay ... use it as 2 separate 4-Bays and run separate cables to a coaxial switch?
Yes.

Last edited by tripelo; 16-Jun-2013 at 7:56 PM. Reason: typo & clarify
tripelo is offline   Reply With Quote
Old 11-Jun-2013, 7:38 AM   #22
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
HDB8X Harness

tripelo,

I knew some of what you mentioned in your last post but definitely not all.

When I assembled the antenna, I noticed that the coax was unusually thin and very flexible with rubber boots and molded connectors installed @ each end.

Yesterday, I decided to document the coax harness and test it and the supplied combiner (by substitution). Fortunately, the coax was marked as 3C-2V, which made looking up its specs. relative easy. It is just another form of 75 ohm coax.



I was surprised to find a ¼” difference in length when measured from screw-on edge to screw-on edge. First, I substituted two length matched 36” pieces of RG-6 between each of the HDB-8X’s BALUN's to the combiner. Worst case loss for 34” of 3C-2V@ 700 MHz = ~.268 dB and should be insignificant compared to RG-6 @ .217 dB. Needless to say, this test didn’t yield any noticeable difference. Next, I left the RG-6 connected to the BALUN’s and substituted an ordinary 1 GHz 2-way splitter for the combiner. Again, no noticeable difference in SNR.

Here is a picture showing the installation and comparing antenna sizes.



I want to try some ideas suggested by your post but I’ve got to finish refurbishing a CONDO.

I bought a 5 year old 3 Br./3bath CONDO for my wife’s son just after I retired and he trashed the whole place. Just a few more hours work & we’ll have it ready to rent –to someone else. Maybe tomorrow or Wednesday I can get back to playing.
Attached Images
File Type: jpg 5. Closeup HDB8X-CM-4228.JPG (108.2 KB, 7015 views)
File Type: jpg Harness Specs 2.jpg (84.3 KB, 5813 views)

Last edited by Pete Higgins; 17-Jun-2013 at 7:56 PM. Reason: Add pictures back in.
Pete Higgins is offline   Reply With Quote
Old 11-Jun-2013, 6:03 PM   #23
329049
Junior Member
 
Join Date: Jun 2013
Posts: 5
I also recently purchased this antenna (HDB8X). I have used the 4228 for a few years and my situation is more difficult than most, 70 miles from the transmitters and in hilly terrain.

Construction of the antenna was straight forward, but I also found it to be weak in physical strength, the reflector end caps came off easily. I was also surprised that the elements were strickly pressure points between wires with screws and plates (no rivets or welds). Here in the Northwest, exposed to allot of water, these points will oxidize and become resistive over time (have already seen this design and used it, with the oxidation results) I doubt this antenna will with stand a 50mph wind (get them in the fall and spring).

Putting the 4228 against the HDB8X with a 10Db (2.7DB NF) amp (have too), the 4228 out performed the HDB8X by about 6Db (measured, I have a RFM 151)
I substituted the cheap combiner with a transformer type and picked up 3Db, but with the 300 ohm to 75 ohm baluns on each bay (HDB8X), it just couldn't match the performance of the 4228.

If I was closer to the transmitters by even 15 miles, and they were in different directions, this antenna would rock, but for fringe, (extreme fringe) it just wont hold up. As a side note the 4228 stand alone is weak also, for me, I must use the preamp, and to bring in a couple of weak stations I have doubled the 4228's (use two of them) into the preamp, allot of plumbing but it works. I did run a test with 2 HDB8X's (yes I bought 2, I had high hopes) into the same situation, and only received half as many channels. Those supplied combiners really bite. (HDB8X)

My next try will be to modify the HDB8X's with 300 ohm combiners (feedline method) and see what happens.
329049 is offline   Reply With Quote
Old 11-Jun-2013, 7:17 PM   #24
329049
Junior Member
 
Join Date: Jun 2013
Posts: 5
I got curious on the combiner. Wow, super cheap, only a 1/2 turn for each leg, good ones will have a min of 2 turns, up to 4 typically. Measured the loss, 4.9Dbm, ouch (before I took it apart)
Attached Images
File Type: jpg cheap balun.jpg (242.4 KB, 884 views)
329049 is offline   Reply With Quote
Old 11-Jun-2013, 9:49 PM   #25
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
Re: 329049

329049,

With your own RFM 151, it sounds like you are a lot better equipped to more rigorously test than I am. Depending on your definition of performance, I suppose my SNR equivalence could be masking relatively large differences in antenna gain. I know it’s less important with digital, but I would dearly love to be able to characterize individual channel signal strengths right at the antenna. I’ve toyed with buying a Spectrum Analyzer to do that and to look through the spectrum for interferer’s, but, just can’t justify the $1,000 – 2,000 cost for my “playing”.

My HDB8X, from Solid Signal, assembled seems to be pretty stout. I’ve carried it by the end caps and they don’t show any signs of coming off (well, I’ll stop that now!). We get higher winds than 50 mph here all the time, in fact, we’ve already had a couple of days with 45 mph gusts and it seems to handle that just fine. Do you recall if all the reflector tube lengths were the same or are the tubes longer where they extend into the wider centers of the end caps?

I have a 40 year old 300 ohm CM-4228 (not the newer –HD) and my wire “whiskers” are riveted through plastic stand-offs to the wires interconnecting the bays. Since it is so old, I wish I could take it apart to clean the “pressure points” although performance wise it was still competitive with my new 91XG.

All of my LA stations are 1 & 2 edge @ ~ 52 miles and San Diego stations are 1 & 2 edge @ ~ 88 miles. Oriented towards LA both antennas showed very similar SNR. Oriented towards San Diego the CM-4228 posted 3-4 dB stronger SNR numbers. I thought at the time this was due to not having the HDB8X alignment optimized.

http://www.tvfool.com/?option=com_wr...1dda169109ca5c

I hope you will continue to report your results and detail any improvements you develop.
Pete Higgins is offline   Reply With Quote
Old 13-Jun-2013, 8:04 PM   #26
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
HDB8X multi-directional? w/dual amps.

329049,

Quote:
I substituted the cheap combiner with a transformer type and picked up 3Db
I write off-line and then post. I didn’t see your combiner post until today. Can you tell us the brand & part number of the lower loss combiner you used for your tests? In my location, 3 dB would make all the difference in the world!

tripelo,

I was able to try your panel isolation suggestion yesterday. I don’t have a pair of matched amplifiers so, since the San Diego stations are much stronger, I started with a PCT MA2-M +15 dB drop amp. on the LA panel and the unamplified San Diego panel signal into the combiner. That combination killed all three San Diego UHF stations. As soon as I disconnected the LA panel from the combiner input all three San Diego channels came back strong.

Then I tried the PCT MA2-M +15 dB amp. on the LA panel and a PCT MA-B1010-1PN + 10 dB single output drop amp on the San Diego panel. That marginally brought back the San Diego stations but killed off or lowered all of the LA channel’s SNR’s. Disconnecting either panel from the combiner caused the other panel’s channels SNR’s to improve, usually by 3-6 dB, but one channel even came up by 10 dB.

These PCT drop amps. work wonderfully in a mixed -13.4/-15.6/-100 or lower environment, consequently, they were my first choice. Your isolation point made perfect sense to me when I read it so I was surprised it didn’t work. After I sat down and thought it through, I realized that these are bi-directional cable TV drop amps. and are designed to pass signals in both directions –duh!

Fortunately, I have two runs of coax going to the push-up. I setup 2 power inserters in the garage and combined their output signals. At the mast I hooked my Winegard HDP-269 to the San Diego panel and my Winegard AP-2870 to the LA panel. I connected the Winegard YA-1713 through an Antennas Direct -20 dB FM trap to the AP-2870 VHF input. Things got a little better but I was still missing channels and sometimes would have to retune several times to get lock on a channel.

My last trip to the roof I installed my Channel Plus NF-471 55dB channel 24-29 notch filter on the UHF input to the AP-2870. For the first time ever this completely eliminated RF channel 26 (24.1 KVCR @ -15.6 dBm). My SNR meter pegs out @ 30 dB and two of the three San Diego UHF stations pegged the meter. The third showed 29.5 dB. That was a dramatic improvement. LA reception also appeared to improve but I’m still missing channels. I was too worn out to keep going so I went in and ate ice-cream and watched TV with what’s-her-name. I don’t know if this will give me a final solution but it definitely improved my situation dramatically. So, until you’re better paid, thanks.
Pete Higgins is offline   Reply With Quote
Old 13-Jun-2013, 9:49 PM   #27
329049
Junior Member
 
Join Date: Jun 2013
Posts: 5
The combiner was on the HDB8X, with a 1/2 turn explains why there is more than 3.7Db loss.
I understand why they did it this way, converting to 75 ohms on each bay, so they could maintain the flexibility (azimuth), but the loss's add up, I should have my 300 ohm conversion done in a couple of days, will be interesting to see if I get any difference at that point.
329049 is offline   Reply With Quote
Old 13-Jun-2013, 11:18 PM   #28
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
329049

I’m looking forward to your results. I expect the HDB8X factory combiner is attempting to match 37.5 ohms (two paralleled 75 ohm inputs) to a 75 ohm output.

Do you recall if all the reflector tube lengths were the same or are the tubes longer where they extend into the wider centers of the end caps?

Can you tell us the brand & part number of the lower loss combiner you used for your tests? Not the one that came with the antenna.
Pete Higgins is offline   Reply With Quote
Old 14-Jun-2013, 7:13 PM   #29
329049
Junior Member
 
Join Date: Jun 2013
Posts: 5
I made a quick 300 ohm lead in, I did use the 75 ohm adapter that was on the antenna as well. Just using 300 ohm into a Diplexer (VHF/UHF) did not work very well.
Overall it gained about 4Db into the RFM151, I compared the stock HDB8X (including it's combiner) to my modified 300 ohm HDB8X (pictures after I post this, because of the delay). I was a little surprised I didn't have a bigger difference, but 4Db when you are in fringe is huge. I also did not use a preamp.
I am figuring there is a better way than the quick way I did the 300 ohm conversion. I have not compared it to the 4228 yet either, but I suspect they will be close to the same. The flexibility of the variable azimuth would give the HDB8X and advantage for those that are closer to the transmitters.
With the stock HDB8X pulling the left side back about 4 degrees, did increase the signal level by 1.5Db, this did not work with my 300 ohm version.
The test was also don at ground level, same mast and coax, just adjusting the antenna for peak readings.
One other note, I tried different lengths of 300 ohm lead in (from the joining to the adapter) and it appears picking a wave length of a channel for the lead in length, will help that channel
329049 is offline   Reply With Quote
Old 14-Jun-2013, 7:35 PM   #30
329049
Junior Member
 
Join Date: Jun 2013
Posts: 5
here are some pic's of what I did.
I use the Wineguard adapters, CC7870 (75 ohm) and SD3700 (300~75)
Attached Images
File Type: jpg 300 ohm test.jpg (118.2 KB, 1406 views)
File Type: jpg 300 ohm test2.jpg (115.3 KB, 947 views)
File Type: jpg 300 ohm test3.jpg (131.8 KB, 1079 views)
File Type: jpg 300 ohm test4.jpg (108.4 KB, 1493 views)
329049 is offline   Reply With Quote
Old 14-Jun-2013, 9:03 PM   #31
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
329049,

Based on earlier posts and posts in other forums, I thought the DIGIWAVE ANT-7288 and Solid Signal antennas were the same. After looking at your pictures it appears they are not. What I bought was a “Solid Signal Xtreme Signal HDB8X High Definition Blade 8 Bay Xtreme Antenna (HDB8X)” for $59.99 with $0.05 shipping.”

While very similar at a glance, it appears our two antennas have completely different elements and mounting frames. Your elements appear to be made from wire, similar to my CM-4228 while mine are made from much wider 1.66 mm (~0.065”) thick stamped, or die-cut, 8” lengths of sheet aluminum. Please look at post #22 above for a better comparison. This may explain why our CM-4228 comparisons yielded such different results?

The picture you posted of the combiner you opened got me thinking that it might pass power through both ports. I just got down from the roof, after pulling mine off the cross brace. I show open from the case (ground) to all of the center conductors and continuity between all three center conductors. Not knowing this, I ran separate (different length) powered coax’s to each amplifier and combined their outputs in the garage. Later today, I’ll rewire taking advantage of the power passing this style combiner affords. Thanks for sharing that picture.

My combiner is labeled “FREQUENCY MIXER, 470-862 MHz, INSERTION LOSS less than or equal to 4.8 dB”. If accurate, that seems like an excessive amount of loss to incur for combining two 4-Bay outputs, especially in a flat panel configuration, but might explain why you saw so much improvement when you substituted a different type of transformer combiner?

At least if I can amplify each panels signal pick-up before it reaches the combiner it should help compensate for that 4.8 dB of loss.
Pete Higgins is offline   Reply With Quote
Old 16-Jun-2013, 4:39 PM   #32
Stereocraig
Senior Member
 
Join Date: Sep 2012
Posts: 457
Very interesting, Tripelo.

Would the isolation provided by one preamp per antenna, also allow for dissimilar antennae and/ or different feed lengths to the combiner?
Stereocraig is offline   Reply With Quote
Old 16-Jun-2013, 5:57 PM   #33
tripelo
Senior Member
 
Join Date: Jan 2010
Location: Dallas, TX
Posts: 122
Tradeoff Gain as Example

Quote:
Originally Posted by Stereocraig View Post
Very interesting, Tripelo.

Would the isolation provided by one preamp per antenna, also allow for dissimilar antennae and/ or different feed lengths to the combiner?
Yes, but

There are tradeoffs. Some mentioned earlier, but not every detail.

Depends on objectives.

An example:

If the two antennas (either alike or different) are receiving signals of similar power level from different stations, then probably the preamps should be well matched in gain and noise figure. In general, transmission line length doesn't matter (but in specific cases, see below, it can matter).

After combination, if amplifiers are not gain matched, the noise from amplifier with the high gain will dominate the noise floor for both signals. Then the signal from the amplifier with low gain may suffer a noise floor that is higher that it may have had before combination (thus lowering S/N ratio for that particular signal).

As with most things, this (preamp gain mismatch) can either be an disadvantage or an advantage.

If one had a really strong signal, then that particular preamp gain could be chosen to be lower than the preamp for the weaker signal. This could help the weaker signal after combination because the noise floor would not be quite as high as otherwise. This is at cost of some reduction in S/N for the strong signal.

----------------------
Signals received from more than one antenna

If both the antennas (different or not) are receiving appreciable signal power from a desired station, when combined there can be phase interference, somewhat like short delay multipath. Since this uncontrolled phase relationship between signals is somewhat random, odds are that it may not be severe.

For consistent results, each desired station's signal power should be pretty well eliminated from all but one antenna.

Conceivably, one could adjust transmission line length to help phase align the signal from one station that was being received by both antennas.

Last edited by tripelo; 16-Jun-2013 at 6:00 PM.
tripelo is offline   Reply With Quote
Old 16-Jun-2013, 6:48 PM   #34
Stereocraig
Senior Member
 
Join Date: Sep 2012
Posts: 457
Thank you Tripelo, for the in depth response.
I am running A/B right now and will probably just add more runs for C/D if I expand.
Stereocraig is offline   Reply With Quote
Old 16-Jun-2013, 7:46 PM   #35
tripelo
Senior Member
 
Join Date: Jan 2010
Location: Dallas, TX
Posts: 122
Combiner Effect - Signal & Noise (above kTB)

Referring to hybrid combiner in earlier post:

A hybrid splitter/combiner schematic may look something like this schematic from Mini-Circuits.

- Ports A & B are splitter outputs or combiner inputs.

- Port S is the combined port.



It seems that most common splitters available are not configured as above. To find a splitter with good isolation between ports, it's probably best to look for a splitter with specified isolation. An isolation value between input ports of 30 dB is good number.

Attempting to add clarification to Case 3 in the previous post. Both the TV signals and the noise, being independent, suffer ~3 dB attenuation from one input port to output, or combined port, of the hybrid combiner. This ~3dB loss results from internal circuitry to provide isolation. There are also additional normal circuit losses that ranges from tenths of a dB to a few dB.

Noise power from two independent sources above the noise floor (kTB) is additive in a hybrid combiner. Equal noise power above noise floor at each input adds at output. So relative to signal power, the noise experiences no loss (3 dB loss through single port, plus 3 dB combined gain). The signal has no combined gain since the other input port does not contain that particular signal.

About Noise floor & kTB:

Because of the preamplifier’s gain, the noise power at combiner inputs is above the noise floor.

KTB, thermal background noise is a major contributor to the noise floor at UHF and above. Man-made and galactic noise can contribute to the overall noise floor but mostly these affect VHF more than UHF.

The overall result is that S/N is reduced 3 dB (hence the twice the noise mentioned in the previous post). Different signal frequencies from the separate antennas pointed in different directions do not coherently add in the combiner. Combiner circuit losses affect both the signal and the noise so this is where the gain of the preamplifier in front of the combiner can help with improved signal to noise ratio.

Assuming all three cases utilize identical preamplifiers, with preamplifiers after combiner in Case 1 & 2, then:

In general compared to single antenna, single preamp, and no combiner, all 3 cases lose S/N at combiner output (by about 3 dB plus).

Case 3 differs:

- Mitigates 2nd antenna loading losses (Case 2) and
- Via gain of preamplifiers, overcomes the additional normal circuit losses in the combiner (Case 1). Although they exist, no transmission line losses are assumed in Case 2.
- The noise figure of two preamps must low enough, so as to be a small contributor to final S/N, else Case 3 could make situation worse.

As the example in the previous post, All cases have interesting specific applications in which they may have some advantage.

-------------------
Will make a small edit to Case 3 in earlier post to help with clarification.
Attached Images
File Type: jpg Hybrid Splitter Combiner.jpg (10.1 KB, 7235 views)

Last edited by tripelo; 16-Jun-2013 at 8:17 PM. Reason: Typo
tripelo is offline   Reply With Quote
Old 16-Jun-2013, 11:29 PM   #36
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
tripelo,

I sent you a PM -I think?

Using two amps ahead of the combiner works better than anything else I’ve tried. Unfortunately, my signals are so weak (all 1-edge & 2-edge) that 4-Bays only works when conditions are moderate to good. I'll keep playing.

Is there a way for me to post graphics that will show & stay? Mine seem to disappear.

Last edited by Pete Higgins; 16-Jun-2013 at 11:46 PM.
Pete Higgins is offline   Reply With Quote
Old 17-Jun-2013, 1:07 AM   #37
tripelo
Senior Member
 
Join Date: Jan 2010
Location: Dallas, TX
Posts: 122
Thanks Pete for your interesting PM.

You may post images that will automatically open for the reader by using the BBcode [IMG].

Place the following in the post where you want the image to appear.

[IMG]URL link to the image[/IMG]

'URL link to the image' is the exact link you see in your browser when you click to display your uploaded image.

After uploading your image, in the Control Panel, select 'User Options', then select 'Attachments', then click to display the desired image. When your image displays, the URL for the image then shows in the URL address bar in your browser. Copy the link and paste it in above.

To get the URL link to the image in Firefox without displaying image:

Right-click the image attachment title in the Control Panel and select the "copy link location".

Paste this link between the URL designations as above.
tripelo is offline   Reply With Quote
Old 21-Jun-2013, 8:48 PM   #38
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
HDB8X multi-directional config.

tripelo & all,

A gentleman on the digital home forum (holl_ands) identified the combiner picture that 329049 posted as a “Magic Tee” and posted several links for me to bone up on Magic Tee theory & design. 329049’s would basically be the T2 portion of the Mini-Circuits diagram that you posted and appears to be missing the port-to-port (2 X R) 150 ohm mis-match dissipation resistor (RINT). According to what I read, the 75 ohms on each input port should transform to 37.5 ohms @ the combiner’s output, necessitating the output impedance matching auto transformer T1 (also missing from 329049’s combiner). The only way I could tell if my model has a resistor across the windings would be to de-solder the metal cover, since a 150 ohm resistor in parallel with a couple of inches of wire is going to appear as a short to an ohm meter. Based on my continuity measurements, I can tell that mine doesn’t have a grounded auto transformer on the output side but maybe if I’m lucky its designed something like this:



I wanted to show how I have the HDB8X configured for multi-directional reception. At my location, I really need the extra gain of an 8-Bay vs. a 4-Bay but when the signals aren’t faded this configuration seems to work very well –without a rotor. When the signals are faded, even my CM-4228 & amplified 91XG tend to drop in & out.

http://www.tvfool.com/?option=com_wr...1dda169109ca5c

As you can see from my TV Fool report, KVCR, RF channel 26 @ -15.6 dBm (and an FM station on 99.9 @ -13.4) would severely overload my pre-amps., but all the rest of my San Diego (~168 deg.) and LA (~292 deg.) stations are 1 & 2 edge @ -85 dBm and below. That’s why I added the 55 dB NF-471 notch filter that rejects off-air UHF TV channels 24 to 29 and the Antennas Direct 20 dB FM Band reject filter to the AP-2870 LA panel amp.


[IMG]http://forum.tvfool.com/attachment.p...1&d=1371843183[/IMG]
A couple of things to note are that:

1. I am powering both amps. through the two panel signal combiner that came with the HDB8X.
2. The NF-471 completely removes RF channel 26 @ -15.6 dBm with the LA panel pointed right at that station and the amplified & unfiltered San Diego panel that’s rotated ~123 deg. is also completely rejecting KVCR. I wasn’t expecting this. This panel does bring in RF channel 30 (KPBS @ -96.3 dBm) from San Diego with an SNR in the mid 20’s.
Attached Images
File Type: jpg MagicT.jpg (9.5 KB, 5879 views)
File Type: jpg HDP8X Multi-Setup -3.jpg (61.5 KB, 1473 views)

Last edited by Pete Higgins; 24-Jun-2013 at 4:58 AM. Reason: Add Title
Pete Higgins is offline   Reply With Quote
Old 25-Jun-2013, 12:08 AM   #39
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
HDB8X Frequency Mixer

329049,

I wanted to let you know that while they still could be electrically equivalent, the HDB8X “Frequency Mixer” and the combiner that came with your antenna are physically housed in different cases. This at the least suggests they are supplied by different manufactures.





How is your testing going?
Attached Images
File Type: jpg Combiner from 329049.jpg (242.2 KB, 6944 views)
File Type: jpg HDB8X Frequency Mixer.jpg (36.4 KB, 7847 views)
Pete Higgins is offline   Reply With Quote
Old 30-Jun-2013, 2:21 AM   #40
Pete Higgins
Member
 
Pete Higgins's Avatar
 
Join Date: Apr 2013
Location: California, 58 miles @112 degrees from Mt. Wilson
Posts: 82
HDB8X multi-direction progress update

tripelo & all,

I wanted to update my progress to show how I had the HDB8X configured for multi-directional reception. For some reason, I haven’t been able to connect to the TV Fool or FM Fool sites for the last several days. (“the site TV Fool is not responding”)

At my location, I really needed the extra gain of an 8-Bay vs. a 4-Bay, but when the signals aren’t faded this configuration seemed to work very well –without a rotor. When they are, even my CM-4228 & 91XG tend to drop in & out.



A couple of things worth noting are:

1. I am powering both amps. through the two panel signal combiner that came with the HDB8X.

2. The NF-471 completely removes RF channel 26 @ -15.6 dBm with the LA panel pointed right at it and the amplified & unfiltered San Diego panel that’s rotated ~123 deg. is also completely rejecting KVCR. I wasn’t expecting this. This panel does bring in RF channel 30 (KPBS @ -96.3 dBm) from San Diego with an SNR in the mid 20’s, so it obviously has good forward gain near channel 26 and very good rejection of it to the rear.

http://www.tvfool.com/?option=com_wr...1dda169109ca5c

As you can see from my TV Fool report, RF channel 26, 294 deg.@ -15.6 dBm (plus I also have an FM station on 99.9 MHz also @ 294 deg. & -13.4 dBm) would severely overload my pre-amps., but all the rest of my San Diego (~168 deg.) and LA (~292 deg.) stations are 1 & 2 edge @ -85 dBm and below. That’s why I added the Channel Plus NF-471 notch filter that rejects off-air UHF TV channels 24 to 29 and the Antennas Direct FM Band reject filter to the AP-2870 LA panel amp.

Subsequent to the single HDB8X multi-direction experiments, I decided to try the CM-4228 in conjunction with the HDB8X. My reasoning being that the theoretical 3 dB of additional gain (~ doubling the signal power) would help compensate during the signal fades that always seem to interrupt the San Diego evening news.

Because it was the only device I’d found that would pass power to both amplifiers from a single coax, I queried Solid Signal about purchasing an additional HDB8X Frequency Mixer. They currently don’t list them for sale, but generously offered to send me one (@ no cost) from an antenna damaged in shipping.

I wanted to reconfigure the test installation to raise the HDB8X above the rotor to the same height as the CM-4228 so I could align its pattern for best SNR but unfortunately I can’t bring down my pushup mast by myself. When originally implemented I had two sons, a teenager and a college student, living at home to help. Now, one lives in Oregon and the other retired last year and lives 60+ miles away. Undaunted, I revamped the test configuration as shown below:



As with previous testing, I “bump aligned” the CM-4228 for best SNR on LA RF channel 36 (4.1 KNBC @ -107.4) and using its alignment as a guide, aligned the HDB8X as close as possible to the same direction. (This makes the assumption that as mounted, both antennas have the same radiation pattern –which may or may not be valid.)

After everything was in place & hooked up, with both antennas pointed towards LA, I measured an SNR of 24.7-25.3 for RF channel 36 (4.1 NBC). I also measured a SNR of 30 (as high as my meter registers) for RF channel 26 (24.1 KVCR) the -15.6 dBm PBS station 3.5 miles from my house and in line with all the LA stations.

Next, I rotated the CM-4228 towards San Diego, stopping @ ~168 deg. As the CM-4228 turned I noticed 2-3 dB fluctuations in SNR. I “bump aligned” the CM-4228 for best SNR on San Diego RF channel 40 (39.1 KNSD @ -95.2 dBm). I went back and measured an SNR of 23.1-23.4 for RF channel 36 (4.1 NBC). Interestingly, having both 8-Bays pointing the same direction added an additional 1.6-1.9 dB to the SNR. As with previous testing, with the CM-4228 8-Bay turned 123 deg. away from my local PBS station, it didn’t show any back lobe response. Unfortunately LA RF channel 31 (5.1 CW @ -103.5) falls within the Channel Plus NF-471 notch filter’s skirt and was lost. (Rotating the CM-4228/HDP-269 back towards LA brings channel 31 back without any signs of overload)

UHF signal-to-noise ratios from both LA & San Diego are usually 20 dB or greater and appear to stay about the same as a single 8-Bay when aligned with a rotor. More importantly, with a sample size of three afternoons, my San Diego news station dips from an SNR of ~25 to a low of ~18-19 in late afternoon to early evening. It had been dropping into the low 15.x’s where signal was lost. Most nights it holds the 18-19 and stays error free, climbing back up after sundown.

With my limited evaluation capability, last summer my new 91XG performed only slightly better than my “original” CM-4228. This spring my new HDB8X, similarly, performed slightly better than my “original” CM-4228. Therefore, I have to conclude that for my location there isn’t a significant performance difference between the three designs. It also suggests that the advertised gain is very optimistically over stated. Unlike long Yagi’s or LPA’s, Bow Tie designs are easily mounted to a tower leg or off the side of a house or balcony (J-Pole). Hinged designs, also afford convenient reception of multiple markets where a 4-Bay is sufficient.

Because of their price and observed performance, if I needed to design a system from scratch to feed UHF to multiple TV’s without the inconvenience of turning a rotor to view different markets, depending on TV Fool’s reported values, I’d be real tempted to try an unamplified HDB8X with the panels aligned for both markets. If signals were such that you couldn’t tolerate the “re-radiation loss” I’d order two HDP-269’s to help isolate the panels and overcome combiner, coax & splitter loss. Finally, if signals are really weak, (deep fringe) combining two HDB8X’s with two matched pre-amps. (HDP-269’s in an overload prone environment) might offer a cost effective solution. I did find this Splitter/Combiner which should support powering two pre-amps. through a single coax. lead-in:

http://www.solidsignal.com/pview.asp...ce=google_base

SNR and antenna gain are not the same thing. I would really like to see gain comparisons and real world test results between the HDB8X that I have and some of the other hinged designs –some costing almost 4 times as much.
Attached Images
File Type: jpg HDP8X Multi-Setup -3.jpg (61.5 KB, 6260 views)
File Type: jpg 2 Ant Test-04.JPG (66.6 KB, 6005 views)

Last edited by Pete Higgins; 30-Jun-2013 at 2:24 AM.
Pete Higgins is offline   Reply With Quote
Reply

Bookmarks

Go Back   TV Fool > Over The Air Services > Special Topics > Antennas


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT. The time now is 2:39 AM.


Powered by vBulletin® Version 3.8.8
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Copyright © TV Fool, LLC