Antenna Advice (Rochester, MN) - AVS Forum
Forum Jump: 
 2Likes
  • 1 Post By Calaveras
  • 1 Post By Calaveras
 
Thread Tools
post #1 of 29 Old 06-22-2014, 01:47 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Antenna Advice (Rochester, MN) [Resolved]

EDIT:

NORTHWEST ROCHESTER, MN FINAL SETUP

Winegard 7694P Antenna pointed straight south in Attic
Winegard LNA-200 Pre-amplifier
Antennas Direct CDA8 Distribution Amplifier (Optional, for long runs)

Results: Good signal on KIMT (3-1 CBS), KAAL (6-1 ABC), KTTC (10-1 NBC, 10-2 CW), KSMQ (15-1 Austin PBS), KXLT (47-1 FOX), and K25NK (58-1 3ABN), Moderate signal on KYIN (24-1 Iowa PBS)

ORIGINAL POST:

http://www.tvfool.com/?option=com_wr...e1c6868e146ac6

I'm not concerned with any stations weaker than KIMT, so the target stations are all within a 21 degree window to my south. I don't really care if I lose access to K25NK either.

I'm working on trying to get better/more reliable signal in my area. I've tried a few cheap antennas and got nothing aside from K25NK. I got the Amazon Basics version of a Mohu Leaf Ultimate and it's passable.

However the signal is weak and flaky. I get artifacts on occasion at it's best, but at it's worst, I lose KTTC altogether and sometimes KXLT is bad, but not completely gone. Interestingly, KIMT and KAAL have no issues. I'm mounting the Leaf on a south facing wall about 20' over ground level.

When I plug it into my computer with a Hauppauge 950q and use their Signal Meter, I get about 22db SNR on KTTC and KXLT in good conditions and 24-26db on the rest. I had run things through a -7db splitter before but started losing signal with just one more split, so I got a CDA8 drop amp and that problem went away. I haven't seen any noticeable drop in SNR through that amp.

I was thinking about going to a 4228HD. That antenna is likely total overkill on UHF but I'm concerned it might not be enough in VHF-HI to receive KTTC. I'm not opposed to mounting in my attic, which is what I'm planning on doing with the 4228HD.

I'm also not sure if I'm going to need a pre-amp. I'll likely need 35' of cable to reach the drop amp but I may have to go 50' since that's the next step up from 25 and I don't have a crimping tool / know how to properly crimp a coax cable.

Last edited by Phane; 07-20-2014 at 07:58 AM.
Phane is offline  
Sponsored Links
Advertisement
 
post #2 of 29 Old 06-22-2014, 02:13 PM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
Those flat indoor antennas are intended for strong LOS stations, not the 1 and 2 edge stations you have. You should follow the TV Fool recommendation of at least an attic antenna. You need both VHF and UHF. Take a look at the Antennacrat HBU series and the Winegard HD-769xP series to see what you can fit and go with the largest one. Point the antenna at about 180 degrees and that should be good enough for the varying directions you have.
Calaveras is offline  
post #3 of 29 Old 06-22-2014, 02:50 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Calaveras View Post
Those flat indoor antennas are intended for strong LOS stations, not the 1 and 2 edge stations you have. You should follow the TV Fool recommendation of at least an attic antenna. You need both VHF and UHF. Take a look at the Antennacrat HBU series and the Winegard HD-769xP series to see what you can fit and go with the largest one. Point the antenna at about 180 degrees and that should be good enough for the varying directions you have.
Would the 4228HD be inadequate? I would be able to wedge it against a flat wall in my attic (south part of my attic ends in cathedral ceiling for living room).

I think I have maybe 6-7' of length from north to south in there from the access to the start of the cathedral ceiling. I'm guessing that that measurement is what is being referred to as 'boom length'.

Edit: I looked into the antennas you recommended and the HBU22 looks like as big as I can go. The Wingard HD-7694P would fit but it claims to only have 30 miles UHF range which would be inadequate and the only model up that they haven't discontinued would be too large. Would the HBU22 be better than a 4228HD? Would I need a pre-amp or would my drop amp be adequate?

Last edited by Phane; 06-22-2014 at 03:28 PM. Reason: Additional Info
Phane is offline  
post #4 of 29 Old 06-22-2014, 04:12 PM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
The 4228HD is a UHF antenna. Some people are able to get by using it on VHF if they have strong signals. You need something that is VHF and UHF. Antennas don't have mileage limits. That is marketing. Is there some reason why you can't mount an antenna outdoors?

You probably don't need a preamp with your short coax run. The distribution amp is probably good enough.
Calaveras is offline  
post #5 of 29 Old 06-22-2014, 04:30 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Calaveras View Post
The 4228HD is a UHF antenna. Some people are able to get by using it on VHF if they have strong signals. You need something that is VHF and UHF. Antennas don't have mileage limits. That is marketing. Is there some reason why you can't mount an antenna outdoors?
Personal preference for one, for two I'm really not comfortable getting on my roof, and for three I think my association forbids visible antennae / weather vanes anyway.

I'll give the HBU22 a try. I was getting all the channels with the flat antenna, just not quite well enough to my high standards. This will probably be miles beyond what I had, even in my attic, unless someone has reason to believe the 22 won't be adequate.

Edit: Should I be concerned that the CEA rating on that antenna is only green and I should be needing violet according to antennaweb.org?

Last edited by Phane; 06-22-2014 at 07:02 PM. Reason: Extra question
Phane is offline  
post #6 of 29 Old 06-22-2014, 07:31 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
After reviewing specs, the Winegard 7964P seems a better match. It has much higher gains, although a narrower beam. Still a CEA rating of only red, though.

The 7694P has somewhere in the neighborhood of 8-9 dBd gain around channel 10 and higher in the UHF ranges with a beamwidth at half of 34 degrees at it's worst.
The HBU22 has a generic VHF gain of 4.1 dBd with a generic UHF of 7.2 and a beamwidth at half of 62.5 degrees (VHF) and 46.8 degrees (UHF)

That means the HBU22 has roughly 1/3rd of the gain on VHF and 1/2 the gain on UHF but has a much wider beam.

My stations are all tightly packed in a 21 degree arc. Better to get the antenna with the tighter beam and higher gains, I think. Based on the polar pattern in the specs, 10 degrees off axis is only a loss of 0.9 of the gain.

References:
http://www.antennacraft.net/pdfs/HBU22_.pdf
http://www.winegard.com/kbase/upload/HD7694P.pdf

Last edited by Phane; 06-22-2014 at 08:15 PM.
Phane is offline  
post #7 of 29 Old 06-23-2014, 04:25 AM
AVS Special Member
 
ProjectSHO89's Avatar
 
Join Date: Feb 2009
Posts: 2,660
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 70
The Amazon Basics flat antenna is a Winegard Flatwave, not a Mohu Leaf antenna.

No one here uses the CEA color coding system. It's far too simplistic and is close to useless except, perhaps, by the sales persons at Radio Shack stores. Best advice is to ignore it.

Either the HBU-22 or the 7694 should work fine, there's no need to slice and dice the specifications down to the dB or fraction thereof. However, if you're going into an attic, the AntennaCraft antennas are much easier to re-fold than are the Winegards. Also, you must separately purchase a balun for the AntennaCraft.

Of equal (or even greater) importance to selecting an antenna is to select its location. You can get the world's best antenna, place it in a crappy location, and get crappy reception from it and it won't be the fault of the antenna. Attics have plenty of stuff that can interfere with easy reception mainly due to metal used in the home's construction. Outside factors include trees, the neighbor's house, or any other structure in or near the signal path.

You shouldn't need a preamp if you keep the coax run to the distribution amp as short as possible.
ProjectSHO89 is offline  
post #8 of 29 Old 06-23-2014, 06:20 AM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by ProjectSHO89 View Post
Either the HBU-22 or the 7694 should work fine, there's no need to slice and dice the specifications down to the dB or fraction thereof. However, if you're going into an attic, the AntennaCraft antennas are much easier to re-fold than are the Winegards. Also, you must separately purchase a balun for the AntennaCraft.

Of equal (or even greater) importance to selecting an antenna is to select its location. You can get the world's best antenna, place it in a crappy location, and get crappy reception from it and it won't be the fault of the antenna. Attics have plenty of stuff that can interfere with easy reception mainly due to metal used in the home's construction. Outside factors include trees, the neighbor's house, or any other structure in or near the signal path.
Thanks for the extra info. I knew about the AntennaCraft folding but didn't know about the balun. I was thinking it was odd that it was 300 ohm output and was wondering about that. Now it makes sense. It looks like I can keep the elements of the Winegard folded in and fit it through my attic access that way. It's a little too wide to fit through based on it's full specs, but the box specs are fine.

I live in a split level, so all the duct work is in between my floors. I think I saw one vent up there from a bathroom exaust. Other than that there's just the metal braces holding up the joists, I didn't see any other metal, just cellulose and fiberglass insulation. Other factors (neighbors, trees, etc) should be shared with the Flatwave antenna, which was almost adequate. The attic mount will be about 5 feet higher up, so it might have a little more clearance.

50' should be more than enough to get me from the antenna to the pre-amp. I'm figuring on the following cable, since it's really highly reviewed and has good ends, etc, although I'm not sure if a generic RG6 Quad would be better.

http://www.amazon.com/gp/product/B00...?ie=UTF8&psc=1
Phane is offline  
post #9 of 29 Old 06-23-2014, 08:10 AM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
Quote:
Originally Posted by Phane View Post
I'm figuring on the following cable, since it's really highly reviewed and has good ends, etc, although I'm not sure if a generic RG6 Quad would be better.
Don't bother with quad shield. It's not lower loss and the extra shielding would only be required in the most extreme RF environment.
Calaveras is offline  
post #10 of 29 Old 06-23-2014, 08:11 AM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Calaveras View Post
Don't bother with quad shield. It's not lower loss and the extra shielding would only be required in the most extreme RF environment.
Great! Thanks for all the advice. I appreciate it. I'll let you guys know what kind of results I get when the goods show up and I get it all installed.
Phane is offline  
post #11 of 29 Old 06-28-2014, 02:22 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Phane View Post
Great! Thanks for all the advice. I appreciate it. I'll let you guys know what kind of results I get when the goods show up and I get it all installed.
Short version:

I think I might need a pre-amp. Suggestions?

Long version:

I got the antenna and attempted to mount it in a few locations in the attic and the signal was only slightly better than I was getting with the flat antenna. The place that ended up working best for me was just laying it on top of the rafters over my vaulted ceiling (no mount). That got me an additional channel (KYIN) that I'd like to keep, but as is it is barely viewable with a signal strength of about 30 (according to a TiVo Premiere) and about 15-16db SNR (according to Hauppauge Signal Monitor). The other signal strengths of other channels are in the mid 50's to mid 70's with SNR's of 25-27db (I think it maxes out at 27).

Another thing I noticed is that my two digital converter boxes (iView 3500STBII's) are cutting out despite the strong signal. Taking a laptop up to the attic and plugging it directly to the antenna is producing similar readings. Good SNR on everything but KYIN which is around 15-16db. Everything else is mostly 25-27db.

As an experiment, I took the antenna out on my deck thinking that if I mounted it underneath it would probably be concealed enough to not piss off the association but even sitting on top of my deck the readings were worse than in my attic.

If I got a pre-amp, do they get hot? Would there be a fire issue laying it on top of cellulose insulation? Would it help boost the signal of KYIN considering the SNR is pretty low to begin with?
Phane is offline  
post #12 of 29 Old 06-28-2014, 05:36 PM
AVS Special Member
 
ProjectSHO89's Avatar
 
Join Date: Feb 2009
Posts: 2,660
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 70
You haven't told us anything about how the signal from the antenna is being distributed to all these devices. Important details.
ProjectSHO89 is offline  
post #13 of 29 Old 06-28-2014, 05:50 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by ProjectSHO89 View Post
You haven't told us anything about how the signal from the antenna is being distributed to all these devices. Important details.
50 foot cable drop down to CDA8 Distribution Amp to 7 outlets with varying cable lengths. 1 TiVo Premiere, 1 TiVo HD, 1 HDTV with built-in tuner, 2 iView 3500STBII's, 2 other unusued runs that are capped with 75ohm terminators and one unused port on the CDA8 that's also got a 75ohm terminator on it.
Phane is offline  
post #14 of 29 Old 06-29-2014, 06:58 AM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Been doing some more research and I think I might have a good idea as to what's going on here. I'm learning as I go, so feel free to correct me if I'm wrong (I probably am). The longest run I have is probably about 10' from the CDA8 to the wall jack and then another 50' around the perimiter of the room to the TV. This would be a total of around 7db loss due to cable attenuation from the antenna and the CDA8 only compensates for 4.5db resulting in a net -2.5db from the antenna.

In fact, the 50' drop from the antenna to the CDA8 is -3db from the antenna due to attenuation, so the CDA8 is already picking up a -3 db loss in signal before amplifying it +4.5 leaving only 1.5db gain left for cable runs from the amp, which isn't likely enough even before you consider that it's amplifying from a poor location in the run anyway.

A preamplifier seems like a good idea. Based on research, I have two options that look good. Apparently the CM7777 isn't as good as it used to be, and maybe the 10G212 would be better at this point? (it's certainly cheaper).

I could use that and get +30db gain at the antenna and toss the CDA8 out altogether since it'd only introduce more noise and go with splitters instead. -3.5 db from a Y splitter then another -7 from four ways to get the desired 8 outputs would net a 10.5 db loss and then another -7db from cable attenuation means I make my runs with plenty of signal to spare and since I can knock the 10G212 back 10db or so I could (theoretically) get less noise by amping the signal less at the lower setting and be amping 20db and losing 17.5, coming out pretty close to ideal margins. The downside to this is that it's a pretty noisy amplifier (3db) compared to my second option.

The other option would be to get an LNA-200 since it is really low noise (1db for UHF) and still gets me +15-17db gain at the antenna. The downside is that that isn't quite enough to make my longest cable runs with splitters, so I'd likely still need my CDA8, which might add more noise by also amplifying attenuation from the 50' drop as well as amplifying the 1db of noise from the preamp.

I'm inclined to go with the 10G212 since it's doing all the amplification I need at the antenna and I could use splitters from there. I think that even though it's adding 3db of noise through amplification it's doing it all by itself and a lot of that noise might be lost through attenuation anyway. I'm thinking that the CDA8 might take the 1db of noise from the LNA-200 and amplify it by 4.5db as well. I'm not sure how the math works out on that.
Phane is offline  
post #15 of 29 Old 06-29-2014, 08:19 AM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
Quote:
Originally Posted by Phane View Post
Been doing some more research and I think I might have a good idea as to what's going on here. I'm learning as I go, so feel free to correct me if I'm wrong (I probably am). The longest run I have is probably about 10' from the CDA8 to the wall jack and then another 50' around the perimiter of the room to the TV. This would be a total of around 7db loss due to cable attenuation from the antenna and the CDA8 only compensates for 4.5db resulting in a net -2.5db from the antenna.
RG-6 has about 5 dB loss per 100' at channel 51 so your 60' is about 3 db loss plus any additional connector loss which shouldn't be much, maybe 4 dB total. Loss is less on lower frequencies.


Quote:
Originally Posted by Phane View Post
In fact, the 50' drop from the antenna to the CDA8 is -3db from the antenna due to attenuation, so the CDA8 is already picking up a -3 db loss in signal before amplifying it +4.5 leaving only 1.5db gain left for cable runs from the amp, which isn't likely enough even before you consider that it's amplifying from a poor location in the run anyway.
This is not how it works. Once signal is lost it cannot be recovered by adding an amplifier somewhere down the line. The 4.5 dB gain of the CDA8 compensates for cable loss after the amplifier.

The mistake is that almost everyone new to this issue looks at gain when they need to look at noise figure. Maintaining the lowest system noise figure possible is the reason a mast mounted preamp is used. The highest signal-to-noise you can have is at the antenna terminals. Anything you do after that reduces the SNR (adds noise and/or reduces signal). The idea is to reduce the SNR as little as possible when the signal arrives at the TV.


Quote:
Originally Posted by Phane View Post
A preamplifier seems like a good idea. Based on research, I have two options that look good. Apparently the CM7777 isn't as good as it used to be, and maybe the 10G212 would be better at this point? (it's certainly cheaper).

I could use that and get +30db gain at the antenna and toss the CDA8 out altogether since it'd only introduce more noise and go with splitters instead. -3.5 db from a Y splitter then another -7 from four ways to get the desired 8 outputs would net a 10.5 db loss and then another -7db from cable attenuation means I make my runs with plenty of signal to spare and since I can knock the 10G212 back 10db or so I could (theoretically) get less noise by amping the signal less at the lower setting and be amping 20db and losing 17.5, coming out pretty close to ideal margins. The downside to this is that it's a pretty noisy amplifier (3db) compared to my second option.

The other option would be to get an LNA-200 since it is really low noise (1db for UHF) and still gets me +15-17db gain at the antenna. The downside is that that isn't quite enough to make my longest cable runs with splitters, so I'd likely still need my CDA8, which might add more noise by also amplifying attenuation from the 50' drop as well as amplifying the 1db of noise from the preamp.

I'm inclined to go with the 10G212 since it's doing all the amplification I need at the antenna and I could use splitters from there. I think that even though it's adding 3db of noise through amplification it's doing it all by itself and a lot of that noise might be lost through attenuation anyway. I'm thinking that the CDA8 might take the 1db of noise from the LNA-200 and amplify it by 4.5db as well. I'm not sure how the math works out on that.

Measurements have shown that the old and new CM7777s have about the same noise figure. The LNA200 is not nearly as good as claimed running about 2 dB or a bit higher on UHF and 4-6 dB on high VHF. I don't know about the 10G212 but it's unlikely to be better than claimed.

With your short cable run you don't need or want 30 dB of gain. There's a rule of thumb I use to determine how much gain the preamp needs to limit the degradation of the system noise figure to around 0.5 dB:

Gain = All losses before the first amplifier + Noise Figure of 1st amp + 8 dB

In your case I'll assume that the CDA8 noise figure is 3 dB since it's not specified.

Gain = 4 dB + 3 dB +8 dB = 15 dB

That's all you need. More gain will not get you anything and much higher gain risks overload.

I'd go with the LNA200 since the gain is more in line with what you need and the noise figure is good enough. You should notice some improvement on UHF and probably no improvement on high VHF since the system noise figure will be about the same with or without the LNA200.

Last edited by Calaveras; 06-29-2014 at 08:23 AM.
Calaveras is offline  
post #16 of 29 Old 06-29-2014, 08:58 AM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
I don't think I explained my logic well enough. I have 50' of cable to the CDA8 and then another 60' beyond the CDA8 to the furthest receiver. That's a total of 110', which is where I'm claiming a net loss of 7db from.

I understand that I cannot truly regain the 3db loss from the 50' cable drop to the CDA8, which is why I was saying that I was getting the 4.5db gain from it in a poor location on the run.

I was saying that the LNA-200's 15 db would be adequate WITH the CDA8 (like you were saying), but I was bringing up an alternative configuration of using the 20-30db 10G212 WITHOUT using the CDA8 and using splitters instead. The splitters would add another 10.5db of loss on top of the 7db of loss from the 110' RG6 (and some might be RG59, not sure) would total 17.5db of loss. Then I wouldn't have to worry about the 3db of noise from the CDA8 on top of the 2+db from the LNA-200.

I'm assuming with your rule of thumb that the +8db you throw in at the end is for loss after the amp in RG6 and connectors. With the 10G212, I'll have zero loss before, +3db noise, +8db which is 11db, but then I have to compensate for the 10.5db in splitters I'm using in place of the CDA8.
Phane is offline  
post #17 of 29 Old 06-29-2014, 10:54 AM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
Quote:
Originally Posted by Phane View Post
I don't think I explained my logic well enough. I have 50' of cable to the CDA8 and then another 60' beyond the CDA8 to the furthest receiver. That's a total of 110', which is where I'm claiming a net loss of 7db from.

I understand that I cannot truly regain the 3db loss from the 50' cable drop to the CDA8, which is why I was saying that I was getting the 4.5db gain from it in a poor location on the run.

It doesn't work the way you think it does. You're looking at it from a gain perspective when you need to be looking at it from a noise figure perspective. A mast mounted LNA200 and the CDA8 mounted where you have it now will work perfectly. The primary noise figure is that of the LNA200. The CDA8 or the TV tuner (in the absence of the CDA8) introduces something called 2nd stage noise contribution. There's a formula to calculate this which is what my rule of thumb approximates. The 2nd stage noise contribution is added to the preamp noise figure and shows how much the preamp noise figure is increased. I like to shoot for no more than 0.5 dB.

In the case of a preamp and a distribution amp down the line, 3rd stage noise contribution enters the picture. Fortunately in almost all cases it is a very tiny number and has negligible contribution to the preamp noise figure. The rule of thumb here is that as long as the loss after the distribution amp is not much more than the distribution amp gain, you can ignore it. In your case 60' of cable is 3 dB loss max and the CDA8 gain is 4.5 dB so you can safely ignore it.


Quote:
Originally Posted by Phane View Post
I was saying that the LNA-200's 15 db would be adequate WITH the CDA8 (like you were saying), but I was bringing up an alternative configuration of using the 20-30db 10G212 WITHOUT using the CDA8 and using splitters instead. The splitters would add another 10.5db of loss on top of the 7db of loss from the 110' RG6 (and some might be RG59, not sure) would total 17.5db of loss. Then I wouldn't have to worry about the 3db of noise from the CDA8 on top of the 2+db from the LNA-200.
Sorry, but again, no. Instead of 3 dB from the CDA8 you need to use the 6-7 dB noise figure of the TV tuner. Using 7 dB for cable + connector loss, 10.5 dB for a splitter, 6 db for the TV and my additional 8 dB, you need 31.5 dB gain. You're better off to split it up between two devices which will minimize the possibility of overload.


Quote:
Originally Posted by Phane View Post
I'm assuming with your rule of thumb that the +8db you throw in at the end is for loss after the amp in RG6 and connectors. With the 10G212, I'll have zero loss before, +3db noise, +8db which is 11db, but then I have to compensate for the 10.5db in splitters I'm using in place of the CDA8.
No. The 8 dB is the amount of gain above and beyond the cable losses and 2nd device noise figure that is required in the 2nd stage noise contribution formula to keep the preamp noise figure from degrading more than 0.5 dB in most typical setups.

I normally don't do this because people's eyes glaze over at the sight of formulas but:

Ft = F1 + (F2-1)/G1

Ft = System Noise Figure in dB
F1 = Preamp noise figure
F2 = 2nd device noise figure
G1 = Preamp gain

Each term in the equation first has to be converted from dB to a linear number using 10^(dB/10).

Perform the calculation then convert back to dB using 10*Log(linear result)

And if you really want to know, add this to the equation above to determine the 3rd stage noise contribution:

(F3-1)/(G1*G2)

The variables are exactly what you think they are. The large denominator produced by G1*G2 is the reason the 3rd stage noise contribution is normally very small and can be ignored.

Attached is the measured gain for the LNA200 I tested. UHF gain runs from about 16-18 dB.
Attached Thumbnails
Click image for larger version

Name:	LNA200 Freq Response.gif
Views:	14
Size:	21.4 KB
ID:	141178  
Phane likes this.
Calaveras is offline  
post #18 of 29 Old 06-29-2014, 11:59 AM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Calaveras View Post
I normally don't do this because people's eyes glaze over at the sight of formulas but:

Ft = F1 + (F2-1)/G1

Ft = System Noise Figure in dB
F1 = Preamp noise figure
F2 = 2nd device noise figure
G1 = Preamp gain

Each term in the equation first has to be converted from dB to a linear number using 10^(dB/10).

Perform the calculation then convert back to dB using 10*Log(linear result)

And if you really want to know, add this to the equation above to determine the 3rd stage noise contribution:

(F3-1)/(G1*G2)

The variables are exactly what you think they are. The large denominator produced by G1*G2 is the reason the 3rd stage noise contribution is normally very small and can be ignored.
Hey, I like math. Although I don't understand the science behind the formulas as I would think that noise would have a cumulative effect downstream and any subsequent amps would be amplifying noise created by the preceding amps and adding their own noise on top of that instead of having their own noise effectively nullified at an exponential rate.

In that regard, it seems like the noise profile of the very first amp is the only one that truly matters, the noise of any additional amps will be nearly cancelled out by the gain of the first in just about any scenario.

Based on the math you provided, the noise of the LNA-200 + CDA8 system would be 2.04db and the noise of the 10G212 alone with splitters would be 3.16db, if I don't fail at math... it's been 17 years since I took calculus.

Last edited by Phane; 06-29-2014 at 12:22 PM.
Phane is offline  
post #19 of 29 Old 06-29-2014, 12:51 PM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
Quote:
Originally Posted by Phane View Post
Hey, I like math. Although I don't understand the science behind the formulas as I would think that noise would have a cumulative effect downstream and any subsequent amps would be amplifying noise created by the preceding amps and adding their own noise on top of that instead of having their own noise effectively nullified at an exponential rate.

In that regard, it seems like the noise profile of the very first amp is the only one that truly matters, the noise of any additional amps will be nearly cancelled out by the gain of the first in just about any scenario.

Based on the math you provided, the noise of the LNA-200 + CDA8 system would be 2.04db and the noise of the 10G212 alone with splitters would be 3.16db, if I don't fail at math... it's been 17 years since I took calculus.
I probably wasn't clear on how to calculate F2.

In the 10G212 example: (log/linear conversions are omitted)

F1 = 3.5 dB (per Antennacraft)
G1 = 30 dB
F2 = Coax/Connector Loss + Splitter Loss + Tuner NF = 7 + 10.5 + 6 = 23.5 dB

Ft = 2.24 + (223.9 - 1) / 1000
Ft = 3.9 dB (meets my <0.5 dB degradation guideline)


In the LNA200 with CDA8 example:

F1 = 2.0 dB
G1 = 17 dB (ave)
F2 = Coax/Connector Loss + CDA8 NF = 3 dB + 3 dB = 6 dB
G2 = 4.5 dB
F3 = Coax/Connector Loss + Tuner NF = 4 dB + 6 dB = 10 dB

Ft = 1.58 + ((4.0 - 1) / 50.1) + ((10.0 - 1) / (50.1 x 2.8))

Ft = 2.3 dB (meets my <0.5 dB degradation guideline)
Phane likes this.
Calaveras is offline  
post #20 of 29 Old 06-29-2014, 01:53 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Calaveras View Post
I probably wasn't clear on how to calculate F2.

In the 10G212 example: (log/linear conversions are omitted)

F1 = 3.5 dB (per Antennacraft)
G1 = 30 dB
F2 = Coax/Connector Loss + Splitter Loss + Tuner NF = 7 + 10.5 + 6 = 23.5 dB

Ft = 2.24 + (223.9 - 1) / 1000
Ft = 3.9 dB (meets my <0.5 dB degradation guideline)


In the LNA200 with CDA8 example:

F1 = 2.0 dB
G1 = 17 dB (ave)
F2 = Coax/Connector Loss + CDA8 NF = 3 dB + 3 dB = 6 dB
G2 = 4.5 dB
F3 = Coax/Connector Loss + Tuner NF = 4 dB + 6 dB = 10 dB

Ft = 1.58 + ((4.0 - 1) / 50.1) + ((10.0 - 1) / (50.1 x 2.8))

Ft = 2.3 dB (meets my <0.5 dB degradation guideline)
Ok, so you include signal loss from connectors/cable/splitters in the equation for F2. I was only using noise and I thought the 10G212 was 3db, not 3.5db. I also brain farted and used the gain (I used 16db) for the LNA-200 when calculating noise for the 10G212. I can follow the math now, thanks.
Phane is offline  
post #21 of 29 Old 07-02-2014, 12:42 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Well, I think I might have to redo some cable runs. Other than that, I'm stumped. The amp came and it did improve things significantly at some TV's, but not others.

Before the amp, the cable jack by my computer was pegging at 27db SNR on KIMT, KAAL, KTTC, and KSMQ, but was at 21 on KXLT, 23 on K25NK and only 16.5 on KYIN. After the amp, it's now pegged at 27db SNR on all channels except for KYIN (which is fluctuating between 19-20). My TiVo HD downstairs reports KAAL and KSMQ pegged at 100 signal strength (they were mid 90's), KIMT, KXLT and KTTC in the mid 80's (were in the 70's) and KYIN in the low 50's (was low 40's). That's seems darn good to me.

The problems are that the TiVo Premiere upstairs is still reporting signal in the 50's-60's across all stations except KYIN (which is now in the mid-upper 30's. I also see artifacting on occasion on that unit. I've replaced all the connectors to that TV with compression connectors but it did nothing to improve the situation. Since adding the LNA-200, KAAL actually got worse than it was before on that TiVo (was mid 70's). The two iView 3500STBII's downstairs are both still cutting out on KIMT still, as well. They now get KYIN loud and clear, though. They didn't before the amp.

I don't think I'm over-amplifying, otherwise I'd see drops in quality on all TV's on the problem channels. When I have more time, I'll try swapping the TiVo Premiere and HD units and see if they report the same signal strength on the different runs. I'll also bring one of the 3500STBII's upstairs and hook it to a known good run and see if it still flakes out on KIMT. If it turns out to be the runs, I'll just replace them. Any other ideas on what might be causing issues?
Phane is offline  
post #22 of 29 Old 07-02-2014, 11:16 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Found more info...

http://www.tivocommunity.com/tivo-vb...d.php?t=446075

Apparently the TiVo Premiere's are really touchy when it comes to signal strength. I had no idea how touchy until tonight. I hooked my laptop to all the runs that I thought were giving me problems and they all tested out pegged with 27db SNR across the board except for KYIN. I tried taking out the CDA8 and running everything from the LNA-200 with splitters. The signal on both TiVos was mostly the same but when I measured SNR, KXLT dropped way down to the 18-20 range. I then wired it back the way it was except I threw combinations of splitters on the run to the TiVo Premiere. -3.5 db improved a bit... -7 improved more... -10.5 made it drop back down again. So the Premiere has a REALLY tight window of what kind of signal it likes. And I'm talking a difference of signal strength from the mid 50's to mid 80's.

Part of the other issue is that while the signal strength of the stronger channels improved after being split, the strength of the weaker ones dropped. KAAL, KSMQ and KIMT come in real strong. KXLT, and KTTC are real flaky. KYIN is hard to test because it stops broadcasting at night, which is when I have the most free time to tinker.

Right now, I put a -7db splitter just past the wall-jack by my TiVo Premiere and split one feed straight to the TV. Are there attenuators that only effect certain channels so I can filter down the strong ones and keep the weak ones?
Phane is offline  
post #23 of 29 Old 07-03-2014, 07:04 AM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
Quote:
Originally Posted by Phane View Post
Apparently the TiVo Premiere's are really touchy when it comes to signal strength.

From what I've read it's not signal strength but that the TiVo Premiere is not very good at handling multipath. This problem is fixed with the Roamio. I have a Roamio Basic and the tuners are equal to all the others I have.
Calaveras is offline  
post #24 of 29 Old 07-03-2014, 07:53 AM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Calaveras View Post
From what I've read it's not signal strength but that the TiVo Premiere is not very good at handling multipath. This problem is fixed with the Roamio. I have a Roamio Basic and the tuners are equal to all the others I have.
Is there a way to fix that short of building a faraday cage around the antenna?

The Premiere has a lifetime sub, so upgrading to a Roamio would cost hundreds of dollars even if I sell the premiere on ebay.

Last edited by Phane; 07-03-2014 at 08:04 AM.
Phane is offline  
post #25 of 29 Old 07-03-2014, 10:09 AM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
Quote:
Originally Posted by Phane View Post
Is there a way to fix that short of building a faraday cage around the antenna?

The Premiere has a lifetime sub, so upgrading to a Roamio would cost hundreds of dollars even if I sell the premiere on ebay.

Yeah, that's a TiVo gotcha isn't it? Even putting the antenna in a cage isn't a guarantee because multipath can come from the same direction as the main signal if buildings and vegetation are in line with the transmitter. The cage isn't very practical in most situations either.
Calaveras is offline  
post #26 of 29 Old 07-03-2014, 10:34 AM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Calaveras View Post
Yeah, that's a TiVo gotcha isn't it? Even putting the antenna in a cage isn't a guarantee because multipath can come from the same direction as the main signal if buildings and vegetation are in line with the transmitter. The cage isn't very practical in most situations either.
That's why I said "short of" with the cage idea. I don't think it'll help. I look south of my house out of my window and a few hundred feet of there is a line of 40' tall bushy trees from 160º to 240º. So winter reception will be fantastic, I'm sure, it's just the rest of the year that's the problem. The TiVo HD is getting 99-100 signal strength on KAAL and KIMT, getting 85 or so on KXLT and KTTC and 65 on KYIN. I figured out how to view SNR on the TiVo and the main channels are all in the 29-33 range, which is stellar. If getting a Roamio means I will never miss cable TV again then I have to look at it from the 'saving me $75 a month' perspective.

I'll order the Roamio with no service. I want to verify it's resolving my issue before I cough up a non-refundable $400 for lifetime on it.

Thanks for all of your help, Calaveras. I really appreciate it. Assuming it's just the multi-path issue left, I should have awesome reception for years to come.
Phane is offline  
post #27 of 29 Old 07-09-2014, 12:29 PM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Curious. The Roamio isn't reporting much better signal than the Premiere (and a bit lower than the HD) but the key seems to be that the Roamio and HD have rock solid signal strength readings while the Premiere's slip back and forth all over the map (as much as 23 on the signal meter and 7db SNR). I think they might have redone the signal meter algorithm in the Premiere. The HD's numbers don't seem to be comparable.

Average across all 7 affiliates:
HD: 84.7 Signal Strength, 27db SNR
Premiere: 56(low)-61.6(high) Signal Strength, 23.3-25.7db SNR
Roamio: 64 Signal Strength, 25.7db SNR

So the HD's tuner seems to be doing something that the Roamio's isn't in making the signal a bit better (1.3db avg SNR) but I don't really feel that's enough to justify 20.7 higher avg signal strength, especially in the upper 20's. The Roamio's readings are enough to encourage me to give it a shot, but not enough to be certain. I'm going to be watching my recordings closely over the next couple of weeks and see if the Roamio picks up blips and chirps that the HD doesn't.
Phane is offline  
post #28 of 29 Old 07-20-2014, 07:57 AM - Thread Starter
Member
 
Phane's Avatar
 
Join Date: Jan 2008
Location: Rochester, MN
Posts: 28
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 10
Quote:
Originally Posted by Phane View Post
Curious. The Roamio isn't reporting much better signal than the Premiere (and a bit lower than the HD) but the key seems to be that the Roamio and HD have rock solid signal strength readings while the Premiere's slip back and forth all over the map (as much as 23 on the signal meter and 7db SNR). I think they might have redone the signal meter algorithm in the Premiere. The HD's numbers don't seem to be comparable.

Average across all 7 affiliates:
HD: 84.7 Signal Strength, 27db SNR
Premiere: 56(low)-61.6(high) Signal Strength, 23.3-25.7db SNR
Roamio: 64 Signal Strength, 25.7db SNR

So the HD's tuner seems to be doing something that the Roamio's isn't in making the signal a bit better (1.3db avg SNR) but I don't really feel that's enough to justify 20.7 higher avg signal strength, especially in the upper 20's. The Roamio's readings are enough to encourage me to give it a shot, but not enough to be certain. I'm going to be watching my recordings closely over the next couple of weeks and see if the Roamio picks up blips and chirps that the HD doesn't.
After a couple of weeks, I'm happy with the Roamio. The tuner seems to work fine, despite the lower readings. I've only seen two cases of macro-blocking over the last two weeks and they were both pretty minor blips that were less than 5% of the picture for a split second. I'm seeing 0 RS Corrected errors across my favorite channels and the RS Uncorrected will be in the 200-500 range after leaving channels tuned in for a few hours. That's pretty good. KYIN still cuts out completely on occasions (inversion layers, maybe) but I can live with that since I mostly just have the DVR grab kids shows off of it when it can and it still works >90% of the time.

The two iView's still have an issue where KIMT's 3.1 cuts out a few times within about 10 seconds of tuning to it, but then ends up rock solid. It only happens on that one channel. I don't know what's causing it, I'm guessing it might have to do with the iView downloading the EPG and KIMT has some weird configuration that's giving it trouble.
Phane is offline  
post #29 of 29 Old 07-21-2014, 08:28 AM
AVS Special Member
 
Calaveras's Avatar
 
Join Date: Jan 2007
Location: Mountain Ranch
Posts: 3,014
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 94
Quote:
Originally Posted by Phane View Post
Curious. The Roamio isn't reporting much better signal than the Premiere (and a bit lower than the HD) but the key seems to be that the Roamio and HD have rock solid signal strength readings while the Premiere's slip back and forth all over the map (as much as 23 on the signal meter and 7db SNR). I think they might have redone the signal meter algorithm in the Premiere. The HD's numbers don't seem to be comparable.

Average across all 7 affiliates:
HD: 84.7 Signal Strength, 27db SNR
Premiere: 56(low)-61.6(high) Signal Strength, 23.3-25.7db SNR
Roamio: 64 Signal Strength, 25.7db SNR

So the HD's tuner seems to be doing something that the Roamio's isn't in making the signal a bit better (1.3db avg SNR) but I don't really feel that's enough to justify 20.7 higher avg signal strength, especially in the upper 20's. The Roamio's readings are enough to encourage me to give it a shot, but not enough to be certain. I'm going to be watching my recordings closely over the next couple of weeks and see if the Roamio picks up blips and chirps that the HD doesn't.

I wouldn't be concerned at all about the signal strength difference reported from different devices. That is entirely dependent on how the software scales the reading. One box's 20 might be another box's 50. I wouldn't be too concerned about the SNR readings in the mid 20's. Anything in the mid 20's is solid. The level to be concerned about is what happens near the digital cliff, barely above SNR 15.2 dB. Compared to my Sony TV, I found the Roamio to require about 0.5 db more signal to decode the weakest signal, implying the noise figure of the TV is about 0.5 dB lower, and the Roamio to be about 0.5 db more tolerant of multipath, at least the multipath I have.
Calaveras is offline  
Reply HDTV Technical

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off