Canadian TV, Computing and Home Theatre Forums banner

3481 - 3500 of 4417 Posts

·
Registered
Joined
·
3,396 Posts
Just found yet another "Max Total Input" spec for HDP-269, this time it's 175,000 uV, but no details re how MANY strong signals nor Crossmod level, so it's incompletely specified. I have to go all the way back to a 2005 W-G Catalog to find this important information in a Preamp spec summary footnote (prior to HDP-269). And should we interpret "TOTAL" to mean the RMS sum of these 7 (UHF) or 5 (VHF) signals at -46 dB Crossmod or the signal level for EACH input signal (as I have assumed, per industry standard IMD measurement spec):
http://www.winegard.com/dealer/catalog-sheets/WC-939 OTA Product Cat.pdf [April 2010 Catalog]

Also note that W-G Catalog confirms that there is indeed a Fixed -12 dB FM Trap in HDP-269, but no info re whether this is valid for the entire FM Band, or is the Max attenuation SOMEWHERE in the FM Band....

FYI: Other W-G Catalog downloads can be found here:
http://www.winegard.com/dealer/materials.php
 

·
Registered
Joined
·
200 Posts
holl_ands,

I have the Winegard product cat. that I downloaded when I bought my HDP-269. In the table it lists 175,000 uV for VHF and 175,000 uV for UHF and to the left of the table it says:
AP Series & HDP-269
• Made of high impact ABS material, mast
mounted for easy installation
• Low noise
• Specialized circuitry protects against
lightening pulses
• AP Series - variable and switchable FM Trap
HDP-269, AP-4700, AP-4800 - no FM Trap

When I had it hooked to my Winegard 10 element FM antenna, I couldn't tell any difference in station count/quality from it and the 10 dB PCT cable drop amp that I was also playing with.

I am still concerned that the "high impact ABS material" doesn't offer any shielding? Both the HDP-269 on a UVSJ and the AP-2870 severly overloaded when hooked to my 91XG/Y10 7-13 array even with a 55 dB channel 26 trap connected directly to the 91XG UHF antenna.

Thanks for your reply on SFDR & intermods last week. My Time Warner Internet & phone have been down since last Friday.
 

·
Registered
Joined
·
6,257 Posts
One thing about the HDP-269 is, unlike the AP Series, it has one amp for both VHF and UHF so when counting the number of strong signals, you need to combine both. When separate VHF and UHF amps are used, you count the number of strong VHF signals separately from the UHF ones.
 

·
Registered
Joined
·
366 Posts
Make of this what you wish

The HDP-269 that I bought and tested several months ago did indeed have an FM trap, roughly 12 dB. I swept it at 5 MHz intervals, so I don't have anything accurate as to where the attenuation begins and ends. Unfortunately, I no longer have the amp on hand for further testing.

I printed out and then scanned a noise figure and gain sweep of that sample (sorry, the software is limited, as are my Excel skills).

https://dl.dropboxusercontent.com/u/23201192/HDP269_Gain_Noise_Figure_graphs.pdf

Several things are quite noticeable:

`1) FM trap is present around 100-120 MHz.

2) Measured noise figure isn't anywhere near the claimed value of 3 dB average. Measurements on low-VHF averaged just below 5 dB, on high-VHF around 4.5, and UHF averaged 6 dB. Excluding ingress, it probably would average 5.0-5.5 dB on UHF.

3) There are consequences in using an ABS plastic case instead of metal shielding for an amplifier. Since I don't have a screened testing facility, ambient signal ingress is plain as day on the noise figure graph. I usually see major ingress on plastic-housed amps in the FM band, at each local St Louis Broadcast channel, in the LTE band, local test equipmen that might be running, and noise from PC and laptop computer clocks. I also have a Verizon 4G phone on my desk, but even with it turned off, there are still major amounts of local wireless signals that show up as spikes on the noise figure graph.

Observations 2 & 3 were similar for the AP8700 & AP8800 that I've also tested although noise figures were a little better.. I do have samples of each of them still in house.

Make of this what you wish.

Cheers!
 

·
Registered
Joined
·
200 Posts
ADTech

Thanks for posting your HDP-269 sweeps. The gain display clearly indicates FM trap existence between ~95 MHz & ~130 MHz. Ironically, last summer I couldn’t tell any difference between using my HDP-269 and a well shielded PCT MA B-1010 +10 dB CATV “drop amp.” when either was hooked to my Winegard 10 element FM antenna. That may help to explain why the HDP-269 works with my 8 Bay Bow Tie when my AP-2870 overloads.

Before I tried my AP-2870 on my tower it gave me fits just trying to align the tunable FM trap. Hooked to my 91XG & Y10 7-13 it overloaded so badly so as to be unusable.

I ordered 3 20 dB FM traps from Antennas Direct a couple of weeks ago (I believe from a link you might have posted) to use with the 2870 but think I’ll probably have to build a shielded housing before traps on the input will do any good. I have a -13.4 dBm FM station and a -15.6 TV station on a mountain top 3.5 miles right out my back door. I doubt my 55 dB channel 26 notch filter and the FM traps will have much effect if the circuit board is bombarded with RF.
 

·
Registered
Joined
·
556 Posts
One thing about the HDP-269 is, unlike the AP Series, it has one amp for both VHF and UHF so when counting the number of strong signals, you need to combine both.
Roger, what information makes you believe that the AP2870 has two separate amps in it? They are very similar in price to the single-input amps (like the AP8700), so I assumed that the only difference was the two inputs were just coupled by an internal UVSJ before the amplification stage. The Winegard spec sheet says nothing about there being two amps (although it does claim to eliminate the insertion loss of a UVSJ).
 

·
Registered
Joined
·
6,257 Posts
Roger, what information makes you believe that the AP2870 has two separate amps in it?
While not proof, it seems obvious from the specs since they quote a different Maximum Total Input# for VHF than they do for UHF. That wouldn't make any sense if it was one amp.

They are very similar in price to the single-input amps (like the AP8700), so I assumed that the only difference was the two inputs were just coupled by an internal UVSJ before the amplification stage.
I figure the AP8700 is the same as an AP2870, but the AP8700 splits the VHF and UHF from a single input (similar to the old CM-7777 in combined mode). Most likely both models have a LPF before the VHF amp and an HPF before the UHF amp so combining them is just a matter of wiring the two inputs together (that is all the switch in the old CM-7777 did).

Looking at the specs for the AP series, it is pretty obvious that Winegard has 4 amplifier designs (2 VHF and 2 UHF, a high and low gain of each) that they mix and match. They likely all use the same PCB but have different stuffing options.
 

·
Premium Member
Joined
·
1,139 Posts
Preamp Overload Test 3

In this preamp overload test I have tried to make some simplifications and learn a little more about overload. I used 3 orig. CM7777 preamps: the first one brings my weak signals up to fairly strong, the second makes them very strong, and the third is used to demonstrate overload. For this simulation the signals out of the second preamp are to be considered the signals from an antenna in a strong signal area that are fed to a preamp (#3).

Code:
                                                           / SLM
CM4221> 7777#1> 50ftRG6> 0747> 7777#2> ATT1> 7777#3> 2-way> 
                                                           \ ATT2> TVorAPEX502

Code:
     Output7777#1      Output7777#2             Output7777#3                    
                         Gain              ATT1   Gain   ATT1    ATT1-25dB
Real SLM  Sig SNR   SLM  over1 Equ SNR    at 0dB  over2 at 22dB  Sig  SNR  
 CH  dBmV Str  dB   dBmV  dB   dBm  dB     dBmV    dB    dBmV    Str   dB

16   +19  79  25    +43   24    -6  24     +50      7    +42     79    24
29   +15  79  25    +40   25    -9  25     +49      9    +41     79    24
40   +13  79  26    +37   24   -12  26     +48     11    +36     79    25
31   +12  78  25    +36   24   -13  25     +46     10    +36     78    25
33   +12  79  25    +36   24   -13  25     +44      8    +34     79    24
46    +4  79  25    +27   23   -22  25     +38     11    +27     79    24
50    -3  74  25    +20   23   -29  24     +34     14    +19     72    23
17    -9  57  18    +15   24   -34  18     +32     17    +15     59    16
45   -11  69  22    +13   24   -36  21     +30     17    +13     70    18

13    +6  79  29    +26   20   -23  29     VHF channels not
 7     0  79  29    +22   22   -27  29      affected by UHF
 9   -10  72  23     +9   19   -40  23      signal overload
The 7777#1 output was fed directly to the TV for SNR figures. Outputs from 7777#2&3 were reduced by 21 dB attenuator #2 before the TV to protect it from overload.

You will notice an anomaly: the #1 dBmV readings of CH 17 & 45 are almost equal, but the signal strength readings from the TV are 57 & 69. My guess is that since CH17 is adjacent to the very strong CH16 signal, the AGC is influenced by CH16 to reduce the tuner sensitivity for CH17. This is called "desense."

The gain of 7777#2 over #1 and SNR indicate that #2 is operating in a linear mode.

The gain of 7777#3 over #2 indicates gain compression with ATT1 at 0 dB. When ATT1 is set at 22 dB, the gain of 7777#3 is offset and the dBmV figures are very close to those of 7777#2.

The SNR numbers are used as a guide, because the spurious signals from IMD raise the noise level and reduce the SNR, first for the weakest signals and then for the stronger ones. The Signal Strength and SNR numbers for 7777#3 were made with ATT1 set at 25 dB as a starting point, where it is operating in a linear mode.

Starting with CH45, I adjusted ATT1. I made a second run with the APEX DT502 connected for a comparison with the SONY TV in case I wanted to make some tests at another site and wouldn't be able to bring along my 22-inch TV in the car. And then CH16:

Code:
     Channel 45 Results              Channel 16 Results
ATT1    SONY      APEX DT502            ATT1    SNR
 dB     SNR       SQ     SS              dB      dB

27       19       100    71              20     24-25                           
26       18       100    71              17      24             
25       18       100    71              15      23                  
24       17       100    71              13      21 
23      16-17      94    72              12      20               
22       16        45    72              10      19
21      15-14      41    72               8      18
20      12-13      15    73               6      17
                                          5      16
                                          4      15
                                          2      15
                                          1      14 with errors
                                          0      picture freeze 
                                          All other CHs show NO SIGNAL  
                                          This is the NONE point
SQ is Signal Quality and SS is Signal Strength. Note that as the attenuation is reduced the signal gets stronger, but the signal quality goes down, because the strongest signals generate more spurious signals in the preamp that reduce the SNR of the weakest signals. If ATT1 is set lower than 20 dB, the CH45 signal cannot be decoded.

I was able to capture some images with the scanning mode of my Sadelco DisplayMax 800 that show the rising noise floor that reduces the SNR of weak signals by scanning CH45 for the signal and then adjacent CH44 (not in use) for the noise. The scanning mode does one channel at a time showing amplitude VS frequency just like a spectrum analyzer. Note again that as ATT1 is reduced, the signal gets stronger, but the noise floor comes up to reduce the SNR. At 20 dB, the difference between -3.2 and 9.1 is 12.3 dB, close to the SNR of 12-13 dB. At 29 dB, the difference between -17.9 and 1.1 is 19 dB SNR.



The other channels between the weakest and strongest required at least 12 to 17 dB for ATT1 to reach 16 dB SNR.

The CM3410 Compared to the CM7777 for Overload Resistance:

7777#3 was replaced by a CM3410:

Code:
       Output    Gain        Results for CH45      Results for CH16   
Real    3410     over2         ATT1    SONY          ATT1     SONY
 CH     dBmV      db            dB      SNR           dB       SNR

16      +58       15            10       19            4        25
29      +54       14             9       19            3        22
40      +51       14             7       18            2        19   
31      +49       13             5       17            1        17
33      +48       12             4       15            0        16
46      +42       15             3       13
17      +31       16
45      +27       14
Conclusions:

The CM3410 is more resistant to overload than the oriqinal CM7777.

There is no gain compression even with +58 dBmV (+9 dBm) output.

The fall of SNR with decreasing attenuation is more rapid with the 3410 probably because of the greater influence of increasing 3rd order products and the lesser influence of clipping distortion which is the opposite of the 7777.

Each of the 3 preamps adds some noise (its NF) to the signal chain, but most of that increase doesn't reduce the SNR readings much because it is below the noise floor that has been elevated by spurious signals from preamp overload.

You might think that you are giving up some gain using the 3410, but the 7777 doesn't deliver its rated gain when near overload because of gain compression.

The maximum input at my location for the 7777#3 for CH45 SNR of 16 dB is about +21 dBmV (+43 dBvm for Ch16 minus 22 dB of attenuation)(-28 dBm), or in terms used by Winegard about 11,220 uV (microvolts).

The maximum input at my location for the 3410 for CH45 SNR of 16 dB is about +38.5 dBmV (+43 dBmV for CH16 minus 4.5 dB of attenuation)(-10.3 dBm), or in terms used by Winegard about 84,000 uV (microvolts).
 

·
Registered
Joined
·
200 Posts
Rabbitt73

You’ve done a tremendous amount of work and it is very insightful.

I’ll need to go through it a few more times to digest it fully, but it is much easier to understand the way you formatted it today than what I was able to cut & paste from the DigitalHome email.

You have taken an ingenious approach to clearly demonstrate the effect of signal overload on a common and very popular commercial antenna pre-amplifier. I’m sure that your results can be extrapolated to other brands/models with similar specs.

I was especially pleased to see the comparison/evaluation of the Channel Master 3410. As you probably already know, CM-3410 is Channel Master’s part number for the PCT MA2-M +15 dB, 2.7 dB NF drop amplifiers that I use. Here is a link to the specs. for that series of amplifiers:


If you can, I’d like a little more clarification on your last paragraph. I have a Winegard AP-2870 rated @ “Maximum total input 110,000 microvolts VHF, 93,000 microvolts UHF” and an HDP-269 rated @ “Total Input UHF/VHF 350,000 microvolts” both of which overloaded noticeably connected to my antenna array. The PCT MA2-M has never shown any signs of overload, so when you say “The maximum input at my location for the 3410 for CH45 SNR of 16 dB is about +38.5 dBmV (+43 dBmV for CH16 minus 4.5 dB of attenuation)(-10.3 dBm), or in terms used by Winegard about 84,000 uV (microvolts)” I guess, based on my empirical results with the Winegard’s, I would expect that 84,000 microvolts to be much higher, or, is it all about just maintaining your SNR of 16 dB?

Other than the difference in initial gain, (which I partially overcome by cascading amps.) can you think of any reason not to use the drop amplifiers? or any advantage to using pre-amps? It looks like at least the MA2-M drop amps have a lower noise figure than the pre-amps by a few tenths of a dB.
 

·
Registered
Joined
·
4,966 Posts
How practical is cascading 3 Amps trying to reproduce overload?
Wouldn't it be better to just use 1?
 

·
Premium Member
Joined
·
1,139 Posts
Pete:
You’ve done a tremendous amount of work and it is very insightful.
Thank you. I was curious to the point that I had to try something to answer the questions that I had about overload. And, I thought that my results might help others who live in a strong local signal area but want to receive some weak signals which seems to be common in Canada.
I’m sure that your results can be extrapolated to other brands/models with similar specs.
Probably true, but I would have to buy some other preamps just to test them unless someone wanted to send me their preamp for testing. I would, however, think that the preamps with plastic housings would not be suitable in a strong-signal area. In order to test a preamp with a plastic housing, I would not be able to use my simulated test and would have to setup in a strong signal area that would challenge the preamp with signal ingress. Maybe the inside of the housing could be sprayed with a conductive paint to provide shielding.
I have a Winegard AP-2870 rated @ “Maximum total input 110,000 microvolts VHF, 93,000 microvolts UHF” and an HDP-269 rated @ “Total Input UHF/VHF 350,000 microvolts” both of which overloaded noticeably connected to my antenna array.
I don't believe what they say until I test it for myself; the marketing department often makes statements that an engineer wouldn't agree with. They have revised their preamp chart to say 150,000 μv for the HDP-269:
http://www.winegard.com/kbase/upload/chart29.pdf

but the spec sheet for the HDP-269 still says Total Input UHF/VHF 350,000 μv:
http://www.winegard.com/kbase/upload/WC-809%20_HDP-269.pdf
I guess, based on my empirical results with the Winegard’s, I would expect that 84,000 microvolts to be much higher, or, is it all about just maintaining your SNR of 16 dB?
As you said previously, every reception situation is unique. I have to make tests to find those answers because it is the tuner itself that decides how good a signal is. I use a SNR of 16 dB as a yardstick, just as the tuner does, because it is at the edge of the "digital cliff." In a location that requires a SFDR greater than can be obtained with a preamp and an attenuator, more exotic solutions are needed like a bandstop filter for a very strong local signal, or a separate antenna with a single-channel bandpass filter and preamp for a needed weak signal.

When you are making an estimate of the needed SFDR, it isn't just the difference in dB between the strongest and weakest signals. It is the difference in dB from the top of the strongest signal to the bottom of the weakest signal, so you must add 16 dB for the required SNR of the weakest signal. And to that you should add a "fade margin" because weak signals vary in strength more than strong ones.

In my simulation it would be: a difference of 30 + 16 for SNR + 5 for fade margin = 51 dB
Other than the difference in initial gain, (which I partially overcome by cascading amps.) can you think of any reason not to use the drop amplifiers?
A drop amp is designed to have a flat response over a broad frequency range; some preamps have circuits that have low gain in parts of the RF spectrum that are not needed for OTA reception to resist signals from other services that might cause interference. IIRC the AD CPA19 is one of those preamps.
 

·
Premium Member
Joined
·
1,139 Posts
majortom:
How practical is cascading 3 Amps trying to reproduce overload?
It's the best that I can do with what I have at my location, and I was so curious about overload that I couldn't wait for the ideal test location. I got it down from 4 to 3 amps, and I was careful to make sure that the first two amps were in their linear range. There is a location nearby that would get it down to 2 (a half-day project). I would need to drive a long distance to get it down to one amp (at least a full day driving around the Norfolk area with a map book and tvfool reports).

The fact that I needed to cascade 3 amps doesn't make my conclusions wrong. It doesn't really matter if it is 1, 2, or 3 amps, the concept remains the same: If you can give up a little gain to reduce overload, then it is possible to improve the SNR of the weak signals. If I haven't made that clear, then my explanation wasn't good enough.

As for my 3-amp simulation, better an imperfect setup than none; I might learn something, which I did. Did you?

Anyone who doubts my conclusions is welcome to make their own tests and post the results. I might learn something from them. That is the scientific method.

If you find that using an attenuator with your preamp doesn't allow you to maximize your SFDR as much as you need because there is no extra gain to sacrifice, then more exotic solutions are required, like separate antennas and filters.

When balm told me that he needed to use an attenuator with a preamp to receive WVNY CH13 at his summer location in Saint Anicet, I didn't understand why, so I asked him. Why would he increase gain with a preamp and reduce gain with an attenuator? He said, because it works. After looking at my CH44 & CH45 scans made with my signal level meter I see why.

Wouldn't it be better to just use 1?
Yes, it would, but my signals are too weak at my location to do that. As I said previously:
Maybe this summer I will have a chance to drive to a strong signal area and repeat this test with real signals.
I still want to do that.
 

·
Registered
Joined
·
645 Posts
Rabbit:

Instead of placing an attenuator ahead of a high gain amplifier, would it not be more prudent to use an amp with lower gain and higher input capability such as the CM7778?

It seems counter productive to use a high gain amp and them attenuate all signals (weak and strong) before amplification.
 

·
Registered
Joined
·
4,966 Posts
relax...it's all good.:p I was only curious why ya used three amplifiers in cascade is all. And only because no one is gonna have that in the real world.

More realistic would be one amp in an overload situation then pad a little bit, noting the attenuator doesn't have to be an attenuator. Just use a length of coax, which in the real world would simply imply bringing the preamp further downstream from the antenna.
 

·
Premium Member
Joined
·
1,139 Posts
Rabbit:

Instead of placing an attenuator ahead of a high gain amplifier, would it not be more prudent to use an amp with lower gain and higher input capability such as the CM7778?

It seems counter productive to use a high gain amp and them attenuate all signals (weak and strong) before amplification.
Yes, it would be more prudent to use an amp with lower gain and higher input capability. That was the point of my post when I said that the CM3410 had better overload resistance than the 7777. Also, the gain of the 3410 is almost as much as the 7777 because the 7777 suffers from gain compression during overload and doesn't deliver its rated gain. The 3410 is a better choice than the 7778, especially the new 7778.
 

·
Registered
Joined
·
200 Posts
Winegard AP-2870 has one amplifier.

I’ve seen several Winegard AP-2870 posts recently debating how many amplifier IC’s it contained. Since I had a brand new one out in the garage I decided to take a look. Here is the best image I could get with my wife’s PowerShot G2.



It appears that Winegard uses this same circuit board for multiple amplifier models. Down the right (output) side are a double series of solder pads labeled 8275, 8700, 3800, 4800, 2870, 8780, 8800 and three more across the top labeled 4700, 2880 & 3700. Mine has a chip part that measures 0 ohms (short) across the pads labeled 2870.

The circuit board contains 24 microstrip line filters (gold squares on photo) and a number of surface mount chip capacitors & inductors. The filters rely on the trace inductance and the distributed capacitance (between traces & between the traces and the ckt. board backplane).

The unloaded voltage from the power brick measured 34.0 VDC. When connected to the Winegard supplied power inserter, the unloaded voltage out of the power inserter measured 17.42 VDC. With the amplifier connected, the input voltage dropped to 17.38 VDC. Power from the RF Out/PWR In ‘F’ connector goes through a 2-terminal device marked with 3 Dots (Red centered @ one end & brown & blue @ the other end) that I assume is a resettable circuit breaker and then passes through a chip diode (protects amp from reverse voltage) and is applied to the input pin of the 78M08 regulator IC. Accounting for diode drop, 16.56 VDC is applied to the regulator. Regulator output measured 7.76 VDC and travels to three chip parts (a chip cap & two parallel chip inductors). From their it goes through another 2-terminal device marked with 3 Dots (Red centered @ one end & brown & blue @ the other end) that I again assume is a resettable circuit breaker and is applied to the amplifier IC. The amplifier carries the designation QS printed on one end and 8Y on the other. An internet search for “QS8Y” didn’t turn up any useful information.

The VHF & UHF inputs are electrically isolated and follow different paths across the ckt. board. UHF flows through the smaller microstrip line filters across the top of the board and LowVHF, FM & High VHF through the medium & large microstrip line filters in the middle and along the bottom. I assume the FM filters are configured in a band stop/tunable frequency stop configuration controlled by the switch near the inputs.
 

·
Registered
Joined
·
3,396 Posts
Here are in-focus interior photos of similar W-G AP-8700 Preamp, with multi-purpose circuit board.
I think that some of those Spiral Inductors form L-C and even Pi Filters using discrete Capacitors:

BTW, here is an important photo tip: Even if the camera has a MACRO function, the MINIMUM Focus distance may be more than you think, so move the camera away from the object and crop the resultant photo image after uploading to your computer. And you can never have enough light...so try turning on the flash and shoot at a slight ANGLE so the camera doesn't see the reflection of the flash.



 
3481 - 3500 of 4417 Posts
Top