[re_nelson]

I performed a test involving the Winegard HD7696P (which I obtained a week ago) compared to an existing duo of the Antennas Direct 91XG and C5.

Here's the TV Fool plot for my location to give you the essential context for this shootout performed in my attic on 27 NOV 2010 between 15:00-16:00 on a day with no tropo to skew the measurements.

http://www.tvfool.com/?option=com_wrapper&Itemid=29&q=id=81a3ba3ab771ef

I removed the existing 91XG/C5 combo (feeding a CM-7777 in separate V/U mode) and, after optimizing the location, put the HD7696 into place (feeding the CM-7777 in combined mode). Here are the Sencore SLM-1456 readings with the HD7696P:

+----+--------+-----+-----+------+
! CH.!  TYPE  !LEVEL! MER ! BER  !
!    !        !POWER! CCN !      !     
!    !        !dBmV ! dB  !      !     
+----+--------+-----+-----+------+
!  8 !8VSB    !  6.9! 34.0! <10-9!
!  9 !8VSB    !-  .4! 33.2! <10-9!
! 11 !8VSB    ! 11.0!>36.0! <10-9!
! 14 !8VSB    ! 18.1! 33.6!4x10-7!
! 18 !8VSB    ! 12.9!>36.0! <10-9!
! 19 !8VSB    ! 16.0!>36.0! <10-9!
! 23 !8VSB    ! 10.6! 32.7! <10-9!
! 25 !8VSB    !- 2.4! 35.1! <10-9!
! 27 !8VSB    !- 7.3! 34.0! <10-9!
! 29 !8VSB    !- 1.8! 31.9! <10-9!
! 30 !8VSB    ! 15.6!>36.0! <10-9!
! 31 !8VSB    !- 9.0! 35.9!1x10-6!
! 32 !8VSB    !  7.4!>36.0! <10-9!
! 34 !8VSB    !- 9.0!>36.0! <10-9!
! 35 !8vsB    !  9.2!>36.0! <10-9!
! 36 !8VSB    ! 11.4!>36.0! <10-9!
! 38 !8VSB    !-12.0! 35.0!1x10-9!
! 39 !8VSB    ! 14.7! 34.8! <10-9!
! 40 !8VSB    ! 13.0! 34.6! <10-9!
! 41 !8VSB    ! 13.1! 34.9! <10-9!
! 42 !8VSB    ! 10.6!>36.0! <10-9!
! 43 !8VSB    !  5.6!>36.0! <10-9!
! 44 !8VSB    !  3.2!>36.0! <10-9!
! 45 !8VSB    !  7.1!>36.0! <10-9!
! 46 !8VSB    !  2.4!>36.0! <10-9!
! 48 !8VSB    !  7.1!>36.0! <10-9!
! 50 !8VSB    !-12.4!>36.0! <10-9!
! 51 !8VSB    !-12.1! 33.3!1x10-7!
+----+--------+-----+-----+------+

I then took the HD7696 down and replaced it with the 91XG/C5 combo (into the CM-7777 switched back to separate U/V). Here are the Sencore readings with the AntennasDirect 91XG/C5 pair:

+----+--------+-----+-----+------+
! CH.!  TYPE  !LEVEL! MER ! BER  !
!    !        !POWER! CCN !      !     
!    !        !dBmV ! dB  !      !     
+----+--------+-----+-----+------+
!  8 !8VSB    ! 13.2!>36.0! <10-9!
!  9 !8VSB    !  7.6! 35.0! <10-9!
! 11 !8VSB    ! 14.7!>36.0! <10-9!
! 14 !8VSB    ! 17.0! 33.6!4x10-8!
! 18 !8VSB    ! 13.7!>36.0! <10-9!
! 19 !8VSB    ! 18.6!>36.0! <10-9!
! 23 !8VSB    ! 10.6! 32.2! <10-9!
! 25 !8VSB    !- 1.4! 32.3! <10-9!
! 27 !8VSB    !- 3.2! 31.8! <10-9!
! 29 !8VSB    !- 1.2! 30.7! <10-9!
! 30 !8VSB    ! 20.5!>36.0! <10-9!
! 31 !8VSB    !- 5.8!>36.0!2x10-7!
! 32 !8VSB    ! 13.3!>36.0! <10-9!
! 34 !8VSB    !- 1.4!>36.0! <10-9!
! 35 !8VSB    ! 16.4!>36.0! <10-9!
! 36 !8VSB    ! 14.3!>36.0! <10-9!
! 38 !8VSB    !- 3.6! 35.3!2x10-9!
! 39 !8VSB    ! 24.0! 34.9! <10-9!
! 40 !8VSB    ! 20.6! 34.4! <10-9!
! 41 !8VSB    ! 21.1! 35.7! <10-9!
! 42 !8VSB    ! 16.7!>36.0! <10-9!
! 43 !8VSB    ! 12.1!>36.0! <10-9!
! 44 !8VSB    !  8.1!>36.0! <10-9!
! 45 !8VSB    ! 13.8!>36.0! <10-9!
! 46 !8VSB    !  8.1!>36.0! <10-9!
! 48 !8VSB    ! 17.9!>36.0! <10-9!
! 50 !8VSB    !- 5.3!>36.0! <10-9!
! 51 !8VSB    !- 8.0! 34.5! <10-9!
+----+--------+-----+-----+------+

As much as possible, I wanted to give the HD7696P a fair test (by tweaking the location with the Sencore before performing the readings shown here).

The expectation was that the HD7696P would fare a bit worse than the 91XG. Aside from RF-14 where it bested the 91XG, all other power readings were down, especially as the frequencies went higher (see RF-50 and RF-51 in particular).

Most surprising to me was that the smallish C5 beat the HD7696P on the VHF side. The published gain for the C5 is around ~7 dBi whereas the larger Winegard shows a high of 10.9 dB. The RF-9 difference was rather astounding since that's where the HD7696P shows peak gain. I'm at a loss (no pun intended) to explain why the 15-element VHF side of the HD7696P didn't at least match the little C5 loop antenna.

To be fair, both the HD7696P and the 91XG/C5 combo provided more than adequate signals for "real world" viewing (as shown by the BER column). My interest was to compare the raw power level (dBVm) between the two configurations.

 

[re_nelson]

...but would you be able to do the test with a different meter?

I did do an A/B comparison yesterday using a Perl script I wrote which iterates through the channels capture measurements from a Silcondust HDHR. I have no other test gear (aside from the signal quality meter on a consumer-grade television and the aforementioned Sencore).

These supplement the readings made in my initial post with the hope they'll be of interest. Both were made within the same time window of 15:00-16:00 on 27 NOV 2010:

Here's the report for the Winegard HD7696P:

CH  SS SNQ SEQ
-- --- --- ---
 8  95  98 100
 9  92  91 100
11 100 100 100
14 100  82 100
18 100 100 100
19 100  82 100
23 100  85 100
25  91 100 100
27  87  92 100
29  89  96 100
30 100 100 100
31  77  77 100
32 100 100 100
34  83  87 100
35 100 100 100
36 100 100 100
38  77   0   0
39 100  98 100
40 100  82 100
41 100  97 100
42 100 100 100
43 100 100 100
44  99 100 100
45 100 100 100
46 100  99 100
48 100  99 100
50  74  75 100
51  66  61 100

...and with the AD 91XG/C5 pair:

CH  SS SNQ SEQ
-- --- --- ---
 8 100 100 100
 9  96  92 100
11 100 100 100
14 100  82 100
18 100 100 100
19 100  86 100
23 100  86 100
25  94 100 100
27  88  98 100
29  90  99 100
30 100 100 100
31  80  82 100
32 100 100 100
34  90  90 100
35 100 100 100
36 100 100 100
38  84  77 100
39 100 100 100
40 100  91 100
41 100 100 100
42 100 100 100
43 100 100 100
44 100 100 100
45 100 100 100
46 100 100 100
48 100 100 100
50  87  91 100
51  85  80 100

As I stated in the initial post, the high-VHF performance of the HD7696P didn't meet my expectation. My intent with the purchase of the Winegard was to perhaps keep using the 91XG for its good UHF performance and use the VHF side of the larger HD7696P to get a little boost in gain on RF-9.

 

[300ohm]

I'm at a loss (no pun intended) to explain why the 15-element VHF side of the HD7696P didn't at least match the little C5 loop antenna.

I dont see a real channel 11 on your TVFool.

Channels 8 and 9 are really strong at above 45 NM. In cases like that, the low SWR of the C5 is what is making the difference I think. The HD7696P most likely has a significantly higher SWR.

 

[re_nelson]

I dont see a real channel 11 on your TVFool.

The owner, CBS, is still simulcasting KTVT on both RF-11 and RF-19 via an STA. Like many VHFs suffering from viewer loss on 12 JAN 2009, CBS opted to light up pre-transition channel 19 while maintaining 23 kW RF-11 (while bumping co-owned KTXA back to pre-transition channel 18. It's actually a complex three-way situation ultimately involving KTVT, KTXA and low-power KPFW.

Here's an old TV Fool report showing what was intended to be the final KTVT facility:

Channels 8 and 9 are really strong at above 45 NM. In cases like that, the low SWR of the C5 is what is making the difference I think. The HD7696P most likely has a significantly higher SWR.

For what it's worth, RF-9 (KFWD) is another oddity. The 55 kW plant reflected in the current TV Fool report hasn't been built (and most likely won't be built according to a very reliable source). Thus, the old TV Fool report posted above with the 13 kW is what's in use with the somewhat lower NM figure (40.3 vs. 45.8).

My interest in tweaking high-VHF was to overcome occasional co-channel QRM on KFWD which, unfortunately is on the same channel as KCEN, a mere 146 km south of KFWD's Cedar Hill, Texas tower. My hope was that the greater gain figures and narrower lobe of the HD7696P would make KFWD more resilient against the QRM in these grievously short-spaced stations that are very close in azimuth at my site.

 

[re_nelson]

The 1454 and 1456 also have a spectrum analyzer function that allows the operator to "see" the waveform of the signal coming off the antenna. The additional data in the BER/MER readings and the peak-to-valley values give the experienced operator information about the QUALITY of the signal rather than only the power of the signal.

The spectrum analyzer is probably my most used tool on the SLM-1456. Judging the height and shape of the waveform is how i optimize the location and azimuth angle. I've found that, as a rule of thumb, "the higher and flatter...the better" yields the best overall signal quality when connected to the television. And that "high and flat" approach of mine tracks to a good P/V value.

By the way, I have no connection whatsoever with either AntennasDirect nor Winegard. I want to make it clear that this little shootout I performed wasn't meant to hype the C5 or denigrate the HD7696P. I was just somewhat surprised that the HD7696P didn't outperform the little C5 on the VHF side. I would have tried the larger HD7697 or 7698 but there's just not enough room in attic for those larger boom sizes.

 

[300ohm]

My interest in tweaking high-VHF was to overcome occasional co-channel QRM on KFWD which, unfortunately is on the same channel as KCEN, a mere 146 km south of KFWD's Cedar Hill, Texas tower. My hope was that the greater gain figures and narrower lobe of the HD7696P would make KFWD more resilient against the QRM in these grievously short-spaced stations that are very close in azimuth at my site.

Well, the technically weaker in gain C5 should decrease the chance of getting distant stations. The two stations are only about 4 or 5 degrees apart from each other and an antenna with that narrow of beamwidth is going to have about 20 to 25 dbi in gain. (none such made for vhf-hi) Even with a very narrow beamwidth, thats not going to stop tropos.

Unfortunately, Texas and the south are in a very high tropo zone from what Ive observed over time at http://www.dxinfocentre.com/tropo.html , and theres not much that can be done about that except to marvel at the distant stations that you can get occasionally.

Just look at the forecast and youll know in advance which days youre going to have problems. I think the tropo forecasts are more accurate than the weather forecasts, heh.

 

[re_nelson]

Just look at the forecast and you'll know in advance which days youre going to have problems. I think the tropo forecasts are more accurate than the weather forecasts, heh.

I don't need to visit the tropo and VHF maps over at DXInfoCentre. All I have to do is wait for the Dallas Cowboys to lose a game so that I can enjoy watching the late night replay of WFAA's caustic sportscaster Dale Hansen on KFWD. That's inevitably sets things up for KCEN to start blasting in for me.

And given the Cowboys' woeful record this season, the tropo has been VERY active! Heh, heh, heh.

All kidding aside, the behavior of KFWD is quite interesting to watch when the tropo pours in co-channel KCEN's QRM. The signal meter on the television starts dropping in a very jumpy, jittery fashion. Then there's severe pixelation, audio dropouts and finally the dreaded "No signal" which remains for periods ranging from 30 seconds to as long as a few hours. KCEN has never been strong enough to overpower KFWD for a lock so I've never seen that Temple/Waco facility. From what I surmise, the two ATSC signals conflict in such a manner that neither can be decoded.

This, of course, begs the question as to why two co-channels a mere 146 km apart were ever allowed during the channel elections. Should KFWD ever build the 55 kW facility, the 6.2 dB boost over the current 13 KW plant might be just enough to give the Dallas-Ft. Worth station an edge over the channel 9 Temple/Waco co-channel during tropo action.

There's another co-channel situation involving KLEG (RF-44) in the Dallas area with KWKT also in Temple/Waco. The former is a low power (15 kW) operation closer to my location. During tropo, KLEG's reading on the signal meter does erode but I suppose the sheer signal density of our local channel 44 keeps the co-channel visitor at bay since KLEG never suffers dropouts nor does it vanish totally.

 

[mclapp]

I wonder if the larger winegard antenna is affected more by the surroundings of the attic.

Due to its size it would have a higher probability of being closer to other objects that could skew the reception pattern.

 

[re_nelson]

I wonder if the larger winegard antenna is affected more by the surroundings of the attic.

Due to its size it would have a higher probability of being closer to other objects that could skew the reception pattern.

That's a valid concern. In the coming days ahead, I'll give the Winegard HD7696P another shot by locating it elsewhere up in the attic that may be less prone to reflection from ventilation ducts.

On the other hand, when I used the SLM-1456 to optimize the location for RF 8, 9 and 11, the waveform was nice and flat. It just never got as high on the spectrum display as the C5. Multipath is generally easy to see on the waveform, since the desired flat top is distorted by notches or angled instead of a nice horizontal top.

ProjectSHO89: Please correct me if I wrong about my interpretation of the spectrum's waveform.

 

[ProjectSHO89]

ProjectSHO89: Please correct me if I wrong about my interpretation of the spectrum's waveform.

You have interpreted it correctly. The flatter the top of the 8VSB waveform, the better off you are. Bigger peaks and valleys translates into more multi-path and greater probability of decoding errors.

Keep in mind that when you switch from the MEAS to the SPECT screens, the unit will auto- range and will usually recalculate the REF LEV (top of the display) based on the currently measured level of the signal. Personally, I find it annoying to have a ref level at some point that is fractional and find that I often re-position the ref level to a multiple of 5 or 10 if I'm going to be switching channels via the SPEC screen.

 

[Jase88]

It wouldn't surprise me about the C5 performing better.

Despite my initial reservations about this antenna, it has proved to be a strong performer in my setup.