Need more coax runs

[jziggity]

I haven't tried using a splitter, yet, but I suppose I could. I know right now my signal strength levels are good, and my SNR is good as well. There are just a couple of channels that are below the preferred 35dB SNR value, but only some times. Last night I was watching a channel that, per the HDHR GUI utility, was showing 95%-98% (low end of 34.4dB) for signal quality. Today, that same channel is pegged at 100% no problems (36-37.5dB). I do have two 2-way splitters that I could try, using one for the modems and one for the TA's/Primes. I guess the best method is to try and see what happens. Sometimes I guess you don't think about the easiest solution first. I'll give it a try and report back.

[barnabas1969]

I'm not saying to get rid of the 8-port amplifier he already has. I'm saying to use a 2-way splitter to connect the 2nd Prime/TA.

[hmmurdock]

I'm not saying to get rid of the 8-port amplifier he already has. I'm saying to use a 2-way splitter to connect the 2nd Prime/TA.

And that's certainly worth a try.

But if he has 2 Primes, 2 TAs and 2 Modems in that location, and only 2 drops going down there, that's 6 devices which means you're looking at:

-3.5db
-3.5db
-7db
-7db
-7db
-7db

...after you split everything. (and even if you just use a pair of 2 way splitters and rely on the pass-through on the TA's that still nets the same signal loss, unless the TA's are amplified)

So what you would be doing in that case is running the dirtiest signal to your most sensitive equipment, and sending the best signal (right off your drop amp) to the least used, and least sensitive equipment.

Yes, it might work, and yes it might save a few bucks, and yes, its worth a try, but its far from ideal from a signal quality standpoint.

If he's going to use passive splitters he's way better of relocating his existing amp to the wiring closet (as per my diagram) and just replacing the UG Amp with a passive splitter in the "old" location of the drop amp. At least that way the "good" signal would be going to the equipment that actually matters.

No matter what the OP decides to do as far as purchasing the UG Amp, or going with passive splitters, I would strongly recommend moving the current amp to the same wiring closet as the other equipment so you're feeding that stuff the cleanest signal possible.

[barnabas1969]

He currently has two feeds in that room. He is feeding two modems, a TA, and a Prime now. He is adding another TA and Prime. If feeding the Prime off of the TA works OK for him, then all he needs is a single 2-way splitter installed before the TA's.

[hmmurdock]

He currently has two feeds in that room. He is feeding two modems, a TA, and a Prime now. He is adding another TA and Prime. If feeding the Prime off of the TA works OK for him, then all he needs is a single 2-way splitter installed before the TA's.

Yes, I'm well aware, that you can add a device, by adding a splitter. I think we've covered that already. You can add even more devices by adding even more splitters. That doesn't mean you should actually do it that way though.

I'm not saying that it can't be done. I'm saying it isn't the best solution, and explaining why it might not work.

So far you've given questionable advice, backed by even more questionable logic. That's fine. That's your prerogative. But I'll kindly ask you to stop crapping all over my advice without at least bringing some sound reasoning to back it up. Thank you.

[barnabas1969]

All I can tell you is that I have a unity gain 9-port amplifier connected outside where the feed from the utility pole connects to my house. It has a zero-loss return path. There are two coax cables running from that amp to my wiring closet. One goes directly to my cable modem. The other goes to a 2-way splitter. The outputs of the 2-way splitter feed two TA's. The outputs of the TA's feed my HDHRP tuners. This is the way it worked best for me (better than using a 4-way splitter and connecting the TA's and tuners to the 4-way splitter). It may work for the OP too. It may not. He already owns a 2-way splitter, so it will cost him exactly nothing to try it. If it works, then his problem is solved. If not, then he can explore other options. My advice is to try what's cheap/free/easy first. How, exactly, is that questionable advice?

Oh... and if the OP has a Scientific Atlanta/Cisco TA, then it does have an amplifier in it. If he has a Motorola TA, it does not.

[barnabas1969]

But if he has 2 Primes, 2 TAs and 2 Modems in that location, and only 2 drops going down there, that's 6 devices which means you're looking at:

-3.5db
-3.5db
-7db
-7db
-7db
-7db

...after you split everything. (and even if you just use a pair of 2 way splitters and rely on the pass-through on the TA's that still nets the same signal loss, unless the TA's are amplified)

I don't know how you came up with those numbers. In case I'm not being clear, the picture below explains what I'm suggesting. There will be a 3.5dB drop at the output of each 2-way splitter. No big deal, unless the output of his 8-port amplifier is less than -3.5dB on any given channel.

This is exactly how I have my TA's and HDHRP's connected. It works very well for me.

[barnabas1969]

Another alternative, in case feeding the tuners from the outputs of the TA's causes a problem, is to use two balanced 3-way splitters. Balanced, meaning that all three output have the same attenuation (usually around 5.5dB for a balanced 3-way). This is still pretty cheap, and 5.5dB of attenuation shouldn't be much of a problem, again unless the output from the 8-port amp is low to begin with.

Here's another picture to show what I'm talking about:

[barnabas1969]

The OP's EDA-FT08300 amplifier has a 3dB forward gain. So, if the signal he's getting from the utility feed is close to 0dB on all channels, the amplifier should easily overcome the 3.5 or 5.5dB insertion loss of 2-way or 3-way splitters.

Now, would you care to explain why my advice is "questionable"?

[barnabas1969]

In case you don't know what I'm talking about when I refer to a "balanced" 3-way splitter, here is an example ($3.98 w/free shipping):
http://www.ebay.com/itm/NEW-Antronix-CM ... 4d108190b7

And, here's a picture:

[barnabas1969]

Here's the math:

( utility_input + 3dB_amp - splitter_insertion_loss)
2-way splitter: 0 + 3 - 3.5 = -0.5
3-way splitter: 0 + 3 - 5.5 = -2.5

Both are OK, unless the utility feed is bad. In order to be a problem, the utility feed would need to be lower than -6.5 (2-way) or -4.5 (3-way).

[jziggity]

Cool - spirited debate. Thanks guys for the suggestions and drawings. I will start with the simple and move toward the more complex and see what happens. I don't even have the 2nd Prime yet, but hopefully soon.

[hmmurdock]

I don't know how you came up with those numbers. In case I'm not being clear, the picture below explains what I'm suggesting. There will be a 3.5dB drop at the output of each 2-way splitter. No big deal, unless the output of his 8-port amplifier is less than -3.5dB on any given channel.

Simple math.

You drop 3.5db at each two way split.

Cable modem and VOIP modem each get a signal that's down 3.5db (one split)

Assuming unamplified TA's, you split the signal that's already down 3.5db at the splitter, so both TA's and both Primes see -7db. (two splits)

If the TA's are amplified, that's even worse for the signal. You' taking a signal that's been amplified, then attenuated, then amplified, then attenuated again before it gets to the Primes. That's a SNR nightmare.

This is exactly how I have my TA's and HDHRP's connected. It works very well for me.

Yes, I'm sure it's fine. And it may work fine for the OP as well. But there is certainly no guarantee. It is certainly not an ideal setup.

I've shown how he can modify his current setup for almost no cost (assuming he doesn't have a spare F/F coupler.. or it'll cost him less than 2 bucks if he doesn't) and a minimal amount of effort.

Moving the drop amp to the wiring closest is the ideal solution.

1) It consolidates all of his active equipment in one place
2) It provides each piece of equipment in his wiring closet with it's own feed from the amp, without any additional attenuation, or amplification beyond what is done in the amp to start with. If he setup was working before, this is absolutely going to ensure it will work when he adds the 2nd Prime.
3) It doesn't require him to make any additional runs to his wiring closet.

The only thing that is left for the OP to decide is if he wants to add a splitter in the amp's previous location to drive his old hardware, or if he wants to add the unity gain amp instead which will insure that his old devices see the same signal levels they did before moving equipment around.

As far as your questionable advice...

You were flat out wrong about what a unity gain amp does.
You continue to advise the OP to mangle the signal as much as humanly possible for the most sensitive equipment in his setup, (evidently) so as not to compromise signal levels for his infrequently used and less fickle equipment.
You seem to have a misunderstanding when it comes to SNR and how to keep it at a good level.
You suggest a setup that adds unnecessary complexity.

My solution is:
Simple. (run all of your sensitive equipment directly off of your drop amp)
Easy. (Move the amp. Unless it's welded to some structural steel it should take less than ten minutes)
Cheap. (buy a coupler for a couple of bucks or take one out of an unused coax wallplate)

I would understand your objections to it if it weren't all of the above, but right now you seem to insist that the OP attempt to do it every jerry-rigged way you can imagine, just so he can avoid doing it the right way. Why go through all of that trouble, when the best way to do it is cheap, simple and easy?

And a few other things for the OP to keep in mind when dealing with CATV...

1) Attenuation (outside of a signal that is too hot to start with) is bad. Given two signals running through the same conditions, the one with the higher level will always have a better SNR. The noise will be the same, but the signal on the attenuated run will be lower. Lower Signal + Same Noise = Worse SNR.

2) Junctions are bad. Properly shielded RG-6 has pretty good noise rejection. Most of your interference comes at your cable ends. The greater the number of junctions, the more noise you will introduce into your signal.

3) Splitters are bad. Spltters bring all the badness of attenuation to the party, and they invite their friend, Junctions, too. Using multiple splitters inline isn't bad. It's downright evil.

4) Simple is better. The less you molest your signal the better it will be when it reaches its destination. Ideal setup (assuming perfect signal coming in from the CableCo) would be 1 unity gain amp with enough ports to feed each device on your system without needing to further split the signal. If your incoming signal is on the weak side, then same thing except with a standard drop amp instead of the unity gain amp.

As I've said before, you're free to experiment and try whatever setup you like. But I can give you a laundry list of reasons why its a good idea to relocate your drop amp to your wiring closet. I can't think of a single one, why it isn't.

[barnabas1969]

Simple math.

Wasn't my math, posted above, simple enough?

You drop 3.5db at each two way split.
Cable modem and VOIP modem each get a signal that's down 3.5db (one split)

Assuming unamplified TA's, you split the signal that's already down 3.5db at the splitter, so both TA's and both Primes see -7db. (two splits)

You seem to think that I'm suggesting to daisy-chain splitters, thus adding up to 7dB attenuation. That's not what I'm suggesting at all. See my pictures above.

Assuming unamplified TA's, you split the signal that's already down 3.5db at the splitter, so both TA's and both Primes see -7db. (two splits)

If the OP has a Cisco/SA TA, it won't be attenuated. In fact, it gets a little boost. If he has a Motorola TA, it will be attenuated some.

I've shown how he can modify his current setup for almost no cost (assuming he doesn't have a spare F/F coupler.. or it'll cost him less than 2 bucks if he doesn't) and a minimal amount of effort.

You've shown him a way to daisy-chain two amplifiers. My advice is to try to avoid doing that if possible.

You were flat out wrong about what a unity gain amp does.

I actually have a degree in electronics technologies, and have worked extensively with both analog and digital high speed circuits. A unity gain cable TV amplifier takes the input signal, amplifies it, and then splits it. The amplifier section amplifies it exactly as much as the splitter section attenuates it. It is not a "zero-loss splitter" as you suggested. It is an amplifier. All amplifiers inject some noise. A good quality amplifier doesn't inject much noise, but some.

You seem to have a misunderstanding when it comes to SNR and how to keep it at a good level.

See my comment above regarding my degree. I actually understand SNR quite well.

You suggest a setup that adds unnecessary complexity.

Splitters are far less complex than amplifiers.

Attenuation (outside of a signal that is too hot to start with) is bad. Given two signals running through the same conditions, the one with the higher level will always have a better SNR. The noise will be the same, but the signal on the attenuated run will be lower. Lower Signal + Same Noise = Worse SNR.

^This is where you go very, very wrong. When you pass the signal through a good quality splitter, the signal AND the noise will be attenuated by exactly the same amount (well, almost exactly... of course there will be a variance depending on the frequency of the noise vs. the frequency of the signal... because every component in the system reacts differently depending on frequency), resulting in almost exactly the same SNR. SNR is Signal level divided by Noise level, and expressed in dB. When you say that the SNR is 35dB, you are saying that the signal level is (approximately) 60 times as strong as the noise level (in micro-volts).

Junctions are bad. Properly shielded RG-6 has pretty good noise rejection. Most of your interference comes at your cable ends. The greater the number of junctions, the more noise you will introduce into your signal.

It's true that each connection has the potential to degrade the signal. However, if you use good connectors that are properly attached to the cable, and you tighten each connection well enough, the degradation is very minimal. Also... my suggestions don't have any more connections than yours.

Splitters are bad. Spltters bring all the badness of attenuation to the party, and they invite their friend, Junctions, too. Using multiple splitters inline isn't bad. It's downright evil.

Wow. Evil? No. If the signal coming in to the OP's house is strong enough on all channels, the amount of attenuation I am talking about won't cause any problems. But, it all depends on the signal coming into his house. As I previously stated, if his incoming utility feed is too low on any given channel, then my suggestion won't work. However, if that's the case... he needs to get his cable company to fix the problem at the utility feed.

Simple is better.

I agree. Splitters are much simpler than amplifiers.

One of the problems with daisy-chaining amplifiers is that no amplifier has a perfectly flat response at all frequencies. They tend to amplify more in the middle of the range for which they were designed (approx 54-1000MHz for CATV). When you daisy-chain them, the problem is compounded. Splitters, on the other hand, react almost the same as the coax cable itself.

[mdavej]

...If the OP has a Cisco/SA TA, it won't be attenuated. In fact, it gets a little boost. If he has a Motorola TA, it will be attenuated some.

I don't want to get in the middle of this, but just wanted to add the Motorola TAs attenuate a lot. I've tried splitting before the TA and I've tried using the TA as a splitter. The latter gives you a huge loss, about double what it theoretically should. So always split before the TA if you have a Motorola.

FWIW, I used to have amps in my system before but found that I got much better results removing them altogether. The only thing that really needs a good strong signal is your tuner. Modems and TAs still work fine even after a big loss. So I would advise one split going direct to your tuner(s) and the other going to everything else. You can amplify and split however you want on the second leg going to everything else, but leave the leg going to the tuner as clean and untouched as possible. It's going to take a big hit when it gets inside your tuner and splits again.

[barnabas1969]

Yeah, I've read that the Motorolas attenuate the signal. I can't remember how much. That's why I posted the picture showing how to use two 3-way splitters in the OP's situation.

Depending on the OP's signal level at his utility feed, your suggestion might work. But, in that case, he'd need at least a 4-way splitter at the point of entry (POE), which would attenuate it 7dB. It's unlikely that his incoming signal is strong enough for that. The amp that the OP already owns is a very good quality amp.

[hmmurdock]

Wasn't my math, posted above, simple enough?

Simple? Yes. Correct? No. You forgot to take into account losses for daisy chaining the Primes and TA's with the 2-way splitters.

You seem to think that I'm suggesting to daisy-chain splitters, thus adding up to 7dB attenuation. That's not what I'm suggesting at all. See my pictures above.

That's exactly what your first diagram suggests. Just because there isn't a physical splitter dangling around in a wiring closet, doesn't mean you aren't splitting the signal at the TA when you daisy chain it. Just because you chose to ignore it doesn't mean it doesn't still split there.

If the OP has a Cisco/SA TA, it won't be attenuated. In fact, it gets a little boost. If he has a Motorola TA, it will be attenuated some.

"Some?" How about 3.5db at the absolute minimum? You can't split a signal without it dropping at least that much. Can't. Be. Done. Even with a Cisco TA. The Cisco might amplify it again but the signal still gets attenuated. Pop quiz... What signal will have better signal quality, assuming the exact same input quality

A) Signal connected directly to amp output
B) Signal connected to amp output then split once
C) Signal connected to amp output then split once, then once more
D) Signal connected to amp output then split once, then once more then amplified

Times up! The answer is A. It will always be A, because every component, and every junction, and every modification of the signal is going to add more noise. Every single one, every time. There are no exceptions. This is like accounting for heat loss in physics, it will always happen.

You've shown him a way to daisy-chain two amplifiers. My advice is to try to avoid doing that if possible

Yes, I showed him the optimal way to connect the two pieces of equipment that has asked about connecting. I also showed him a way to do it without the unity gain amp. You might also note that he doesn't have an ideal setup where every device can get a drop to single location, thus the need for amps (or an amp and a splitter) in two locations. I'm just trying to show him the best way to accomplish his goals. You're trying to show him a way to do it.

I actually have a degree in electronics technologies, and have worked extensively with both analog and digital high speed circuits. A unity gain cable TV amplifier takes the input signal, amplifies it, and then splits it. The amplifier section amplifies it exactly as much as the splitter section attenuates it. It is not a "zero-loss splitter" as you suggested. It is an amplifier. All amplifiers inject some noise. A good quality amplifier doesn't inject much noise, but some.

If you knew what a unity gain amp was to start you wouldn't have said he ran the risk of overdriving his signal with it. And I'm aware it isn't a zero loss splitter. I said it behaves as a zero loss splitter

See my comment above regarding my degree. I actually understand SNR quite well.

with all due respect to you and your associate degree (I have the same one, for the record), no you don't understand it quite well. You are correct, that at the output of the splitter, the attenuated signal will have the same SNR as the input (ignoring any noise added at the splitter) But at any point past the splitter the signal will be weaker, but any new noise added will not. thus any new noise added to the weaker signal will have a higher SNR as compared to the same signal that wasn't attenuated. Once you attenuate a signal it doesn't magically stop picking up more noise. Any noise added will affect a weaker signal more than a stronger one. You can ignore that all you like but it is also true, without exception. You will have a worse signal to nose ratio on an attenuated signal. Always.

Splitters are far less complex than amplifiers.

So you're suggesting he not use any amplifiers at all? Because it looks like me you're suggesting he amplify it, then split it, then split it again.

I'm suggesting moving his current amplifier. That's it. My main concern is getting a good signal to his sensitive equipment in the wiring closet. He can do whatever he wants for the old legacy equipment that he doesn't use frequently. Put a splitter back there (and if that works, good for him) or buy the unity gain amp if the levels for that are too low, but my focus is getting his equipment in the wiring closet a good signal and the cheapest most simple way to do that is to just move the drop amp. I'm not sure why you're so vehemently against doing that.

^This is where you go very, very wrong. When you pass the signal through a good quality splitter, the signal AND the noise will be attenuated by exactly the same amount (well, almost exactly... of course there will be a variance depending on the frequency of the noise vs. the frequency of the signal... because every component in the system reacts differently depending on frequency), resulting in almost exactly the same SNR. SNR is Signal level divided by Noise level, and expressed in dB. When you say that the SNR is 35dB, you are saying that the signal level is (approximately) 60 times as strong as the noise level (in micro-volts).

No, this is where you ignore real world conditions and want to stay in lossless physics experimentation-land. The noise at the output of the attenuation will have the same SNR as the input. (assuming no additional noise added, which is an incorrect assumption, but I'll give it to you for the sake of this discussion because you're still wrong about it) However, once the signal continues down the signal path, more noise added is inevitable. That noise that finds its way into the signal after the attenuation will have a higher SNR compare to a signal that wasn't attenuated. If you have a signal at 60db, and it picks up some noise at 1db, then compare that to a signal that is at 60db, then attenuated to 56.5db, then it picks up that same 1db noise. Which has a better SNR? 60:1 or 56.5:1 ? The lower signal will always have a worse SNR in the real world compared to a higher level signal, because of noise added after the attenuation.

It's true that each connection has the potential to degrade the signal. However, if you use good connectors that are properly attached to the cable, and you tighten each connection well enough, the degradation is very minimal. Also... my suggestions don't have any more connections than yours.

All other things being equal "minimal" degredation is still more than no degredation. What's better? Minimal, or none?

And yes, you are suggesting more connections.
Even if you include the coupler...
Mine: coupler -> Amp -> device
Yours: Amp -> Splitter -> device (splitter) -> device

Wow. Evil? No. If the signal coming in to the OP's house is strong enough on all channels, the amount of attenuation I am talking about won't cause any problems. But, it all depends on the signal coming into his house. As I previously stated, if his incoming utility feed is too low on any given channel, then my suggestion won't work. However, if that's the case... he needs to get his cable company to fix the problem at the utility feed.

If the OP's setup is currently working on all channels then what I'm suggesting will continue to work for him. Yours? Maybe. Maybe not.

I agree. Splitters are much simpler than amplifiers.

You're suggesting splitters and amplifiers. I'm just suggesting an amplifier.
"Amplifier" is simpler than "Splitters and Amplifiers."

One of the problems with daisy-chaining amplifiers is that no amplifier has a perfectly flat response at all frequencies. They tend to amplify more in the middle of the range for which they were designed (approx 54-1000MHz for CATV). When you daisy-chain them, the problem is compounded. Splitters, on the other hand, react almost the same as the coax cable itself.

Yes. That is a problem with daisy chaining amplifiers.

But given the choice of:
A) Slap in a splitter where the amp used to be (after moving it)
B) add a unity gain amp where the old amp used to be
C) rewire the whole house so that everything terminates in the wiring closet, and buy a drop amp large enough to feed every device (thus avoiding multiple amps)

I would choose, A then B then C.
I only addressed the 2nd amp because the OP brought it up.

But keep in mind, my suggestion for (possibly) daisy chaining the UG amp to the drop amp (if a splitter doesn't work there) would only affect the legacy devices. The important stuff in the wiring closet still gets the best quality signal.

No matter how you try to spin it, you're sending the best quality signal to the old legacy devices in the house, and butchering (to varying degrees depending on if you go with multiple splitters or balanced splitters) the signal going to his equipment in the wiring closet.

Now I may have read the comments from the OP completely wrong, but I'm under the impression that his focus is making sure the stuff in the closet is working right, which is what my focus is. You're priority seems to be less about making it work (welllllll this might work. iiiiiffffff, the signal from the pole is strong enough...) and more with making sure there is phisically an available F connector to connect the new Prime (and TA) to.

As I said, you're suggestions might work, and considering the minimal amount of effort it would take to move the drop amp, I honestly can't fathom why you would suggest a less-than-ideal setup, when it is so simple and easy to get an ideal one. I really don't. And I know you're going to come back and say "but this is good enough!" and I still say, I'd rather do it right than good enough every time.

[barnabas1969]

You forgot to take into account losses for daisy chaining the Primes and TA's with the 2-way splitters.

It depends on the TA. I'm not 100% sure, but I think I've read that the Motorola TA attenuates the signal 6dB.

The Cisco might amplify it again but the signal still gets attenuated.

Yes, that would be correct. But I still get 35-37dB SNR on all channels. So, who cares?

Pop quiz... What signal will have better signal quality, assuming the exact same input quality

A) Signal connected directly to amp output
B) Signal connected to amp output then split once
C) Signal connected to amp output then split once, then once more
D) Signal connected to amp output then split once, then once more then amplified

Well, A, B, and C would all have nearly equal "quality". Signal quality refers to the SNR. Now, which will have the highest signal strength? Yes, that would be A. But, if the resulting signal strength of B and C is negative 7dB (-7dB) or higher, it really doesn't matter (provided that the SNR at the input source is at least 33dB or so).

because every component, and every junction, and every modification of the signal is going to add more noise.

First, connectors (you call them "junctions") do not "modify" the signal. The resistance of the connection will cause a very, very small reduction in signal strength, but not enough to even be concerned about... unless you have a problem with water infiltration or some other cause of corrosion at the connection. Second, splitters and connections, when using good quality components and proper installation techniques DO NOT affect SNR (quality) enough to matter. The do reduce signal strength, but as long as the signal strength at the input to the tuner is within the tuner's spec's... it's OK.

This is like accounting for heat loss in physics, it will always happen.

It's nothing like thermal dynamics. You're just making yourself sound stupid now.

How about 3.5db at the absolute minimum? You can't split a signal without it dropping at least that much. Can't. Be. Done.

Yes. It. Can. Are you aware that there are unbalanced 2-way splitters with a 1dB attenuation on the 1st output, and 12dB attenuation on the 2nd output?

Look, I'm not going to bother replying to, or even reading the rest of your post. I'm trying to provide the OP with a solution that won't cost him much. In his first post, he stated that he was considering the purchase of an additional amplifier. Your suggested configuration (see your picture in post #9) is telling him to buy the 2nd amp, and connect them in series. I'm recommending that he instead spend $8.00 (at most, possibly nothing at all). Sure, it's possible that my suggestions won't work if the signal being delivered from his cable company sucks... but if that's the case, he should ask them to fix it! The simple facts of the matter are:
1) The OP doesn't currently own a 2nd amplifier. Buying one of the ones he asked about would cost him at least $70.00. My suggested configurations will save him money, and will work just fine provided that the signal being delivered from the utility pole is good.
2) Daisy-chaining amplifiers degrades SNR (signal quality). In order to get the best signal quality, don't daisy-chain amplifiers. In fact, as stated earlier by mdavej, you'll achieve the best signal "quality" by not using an amplifier at all.
3) Splitters reduce signal strength (power). But, they do not affect SNR (quality) in this scenario enough to matter.

Enough said. I'm not going to respond to you anymore.

[barnabas1969]

Just in case anybody cares, here are some pictures showing my signal strength and quality at my lowest frequency channel, a middle channel, and my highest frequency channel. My configuration, which hmmurdock thinks is so terrible, works very well. As you can see, they are all within +/- 7dB, and the SNR is 34.6 (pretty darn close to 35) or higher. But I'm stupid, so what do I know?

Freq: 189MHz, Power: 6.5dB, SNR: 36.8dB

Capture1.JPG (28.98 KiB) Viewed 1373 times

Freq: 549MHz, Power: -0.9dB, SNR: 34.8dB

Capture2.JPG (29.67 KiB) Viewed 1373 times

Freq: 777MHz, Power: -3.1dB, SNR: 34.6dB

Capture3.JPG (29.98 KiB) Viewed 1373 times

[hmmurdock]

Oh goody! I get the last word. How magnanimous of you!

Thanks for acknowledging I'm right about the TA's attenuating the signal. I knew you'd come around eventually. But I still like that you speak of the TA's that split and amplify the signal as if they are the bee's knees but when talking about unity gain amps, you act as if splitting and amplifying the signal is an affront to God. You might want to pick a side of the fence and stay on it next time (not here... since you already said you won't be posting here anymore)

And I'm happy for you that you get 35-37db SNR on all channels. That doesn't mean the OP will see the same results, and you're still suggesting a solution that will drop his SNR (even if it is a marginal loss) I find it odd you seem so dead set against trying to get the best quality signal to the equipment that typically requires the best quality signal to operate well.

And A,B and C would all have lower quality. It might not be a huge loss, but it will absolutely unequivocally have more noise than D.

And perhaps "modify" was a poor choice of words. "Alter" would have been better, but my point still stands. Every connection is going to alter the signal in some way, and it is another place for interference to creep into the signal.

And signal loss isn't the same thing as heat loss, but I wasn't saying they were the same thing. I was making an analogy. You might want to look into those. They can be helpful in understanding how things work.

And I will concede one point though... You're right. You can end up with a split where one output sees less than a 3.5db drop. I should have been more clear and said that the average drop between the two must be at least 3.5db And while you are correct with your example, I would have been a little more impressed if you could have come up with an example that might have been useful in the context of this discussion. I don't think the OP would benefit much from a 2 way splitter with a 12db drop on one output.

As far as every my recommendation vs yours:

I showed him the best way to set everything up, using the equipment he mentioned. I also acknowledged that the 2nd amp might not be necessary (a point that you seemed to have missed several times)

My approach is to change the signal as little as possible to as to present the best possible signal to the equipment that the OP is asking about. My plan requires nothing more than buying a splitter to feed the devices in the old amp location, and a F/F coupler. Should be less than $10 and 10 minutes of labor, and uses fewer components than either of your suggestions.

You basically hope that everything is ideal, from the pole to the wiring closet, and then assuming it is perfect, decide to cut corners and do it the "yeah, yours will present a better signal but mine is probably good enough" way.

Call me crazy, I'd rather make it good, instead of good enough.

And I don't think your setup is terrible. Just not as good as it should be. I think it's great that you stumbled upon something that works, but if you're in the process of changing your setup, why wouldn't you just try to do it right, especially if you already have the bulk of what you need on-hand?