Coax splitter to whole house

Split before the modem, run one directly the to modem, the other to an amplifier and splitter (aka distribution amp). The amp should be a bidirectional for VOD/PPV usage, etc.

Not personally familiar with that PCT unit, but it looks like what you need...

Jeff

This one from amazon was recommended to me although I haven't needed to use it yet. Lots of good reviews...

8-Port Bi-Directional Cable TV HDTV Amplifier Splitter Signal Booster

As jautor says, split the incoming signal to the the modem and amplified splitter.

They are both excellent drop amps, although I would go with the bypass amp (PCT-VC-9U) for 2 reasons: 1- The splitter that you would place (to connect your eMTA or cable modem) at the input of most drop amps is integrated into the PCT-VC-9U; and 2- it has a return path amplifier, which means that the loss on the return path (where the STBs talk to the cable company for VOD, PPV, etc.) is ~0dB instead of ~12dB through a regular drop amp.

Only 8 ports? I'd try it with just an 8-way passive splitter first. At 10 dB loss per port you'll probably be fine, unless your drop to the road is a couple hundred feet or more.

You must live in some wonderful corner of the world where the cable company does its job and provides a nice adequate signal that would handle that...

(certainly doesn't hurt to try - keep your receipt)

Nothing really special. Most cable systems I've rebuilt or swept in (including the one I'm connected to) call for around 15 dB (minimum) at the tap. That gives you 5 to 10 dB at the ground-block. Take away 10 dB for the splitter, and another 2 or 3 dB for the internal wire and you're still well above the -10 dB that digital tuners (including the ones in the modem) work great at. Most tuners will still work fine down to -15 dB or so.

PS: I just now checked, and my modem is seeing -7 dB with 10 dB of splitter loss (a 2 way and a 4 way) in the line.
Edited by olyteddy - 7/3/12 at 11:48pm

Gee, Ted, to read your description, you have a pretty poor distribution in your house. I mean, +10dBmV becomes -7 at your cable modem? Sure, if you had it connected to an 8-way splitter which is after a 4-way splitter. But, I know you're not going to do that. Let me say that in no way do I challenge your credentials or what you have in your house.

Perhaps you ought to mention that the signal levels of 5, 10, and 15 dBmV are analog while the -7dBmV that your cable modem sees is QAM, which is 12dB below analog.

I've always said, "Do the math."

For our example, let's say that the bandwidth of the system is 750MHz (116 NTSC channels), and that the cable modem is at channel 115). If the signal level at the tap is +15dBmV (analog), the level of the DOCSIS stream is +3dBmV. If we go 50' to the groundblock (Series 6 cable throughout) and then another 25' to the distribution point, the signal will lose ~4.2dB, which puts it at -1.2dBmV. Drop 3.8 for the 2-way splitter and another 1.4 for the 25' of cable to get to the cable modem, and there's -6.4dBmV hitting the cable modem. Nothing at all wrong with that. (Personally, I'd probably put a DC-9 or a DC-6 in place of that 2-way splitter with the TAP leg feeding the cable modem. It would depend on the transmit level of the cable modem.)

Back at the 2-way splitter, there's -5dBmV at the out port to feed the TV distribution splitter. A 4-way splitter loses 7dB, which leaves -12dBmV to head to the outlets. Sure, runs of ~50' to the outlets will keep the signal level above -15dBmV, but that's a bit too close for my comfort. If a sizable portion of the plant that feeds the tap is overhead, that 15 (3)dBmV is going to drop in the summertime, as will the signal along the drop, which may result in levels at the outlets being lower than the specs call for.

Not really. I have about 100' of drop and 35' from the groundblock. At 7 dB per hundred (at 600 MHz, my modems frequency) that's 10dB cable loss. 10dB splitter loss (a 4-way at the GB, a 2-way here at my desk) leaves -7dB for the modem. FWIW my transmit level is only 46dB (I'm fed by an 8-port 11dB end of line tap). My HDHomerun (connected to the other side of the 2-way at my desk) reports -10dB on channel 117, the highest used in this system (753MHz). That amounts to 8-way distribution without an amp. And that is the point of my original post.

olyteddy, only way that you are going to tell if there is any loss, is to have a tech come on site, and connect to the coax that is attached to your modem, and read what it is from the headend to where they are standing. Regardless what is printed on the label of the splitter or on the packaging for the barrel connector, or gas discharge, you are just assuming.

Uh, no. I can read the modem level from it's internal tuner. I could also hook up my signal level meter, but I have no need to because the Modem reports the input levels. Oh, and by the way, I did have a tech out the other day because we upgraded our cable and I changed my HSD account. Showed him the modem setup page and it agreed with his meter. Hey, you too can check the signal level at your modem! JUST CLICK THIS LINK IF YOU DARE!!!! (the preceding link assumes a Motorola or compatible modem...)

Um no you can not. You have no way of knowing true signal levels, etc, without the proper equipment. Just because the box shows something, does not mean that the algorithm that is used for the couple of thousand dollar test equipment the tech uses is going to show the same. What they may show is going to be completely different from what you show on the STB, DVR, or modem.

And btw, your link does nothing, since a lot of us do not use the same IP, nor have the same service as you do. Sorry, two strikes, try again.

Are you some kind of troll? I mean really, just what do you consider 'proper equipment'? I used to do this stuff for a living. Splicing, activating, balancing and sweep. I believe the cable tech's Tri-Lithic 860 meter would qualify as 'proper equipment' (it was, after all issued by the cable company). As would the set-up page of the modem. That's what the link leads to (and as the disclaimer states, that's the page for Motorola modems) and anyone connected to a cable modem can check their levels simply by clicking it. Two independent devices measuring the same commodity, and achieving the same result, to me seems believable. I really don't think a third device would differ significantly. Or a fourth...


PS: A good used 860 can be had for less than a kilo-buck.

Um no, that link does nothing on my system, since again, I do not have the same service, nor the same equipment. And if you were a tech, which I doubt you ever would, you would not be bragging that the STB, DVR, Modem is better than the equipment that techs use on site. Also, those techs that do know how to use the equipment properly, will tell you that the readings in the STB, DVR, modem are garbage and do not use the same algorithms that are used for the test gear, that the manufacturers of said test gear use.

Again, believe what you want when you pull up the page from your modem, stb, or dvr, but it will be way off what the true readings are that those who actually know how to use the test gear will actually see.

As I've said several times, if you have a Motorola Cable Modem it will show you the levels. Other brands may use something other than 192.168.100.1. I don't know, but if I needed to know there's always Google.

Where did I say the built in is better? I only contend that if that reading is the same, then it's likely right.

What algorithms? It's not rocket science; it's just a simple voltage reading.

Using a meter is not that difficult. Turn it on, pick a channel, push a button, read the signal strength off the LCD. Done it thousands of times. 'Sweeping' is only slightly more complicated. You have to look at a line instead of numbers.

Even when I did have a Motorola Modem, again, it was not the same as when I and a tech that I knew for 20 years stood side by side and chuckled as to how off the motorola algorithms are. Also to give you a clue, my Brother In Law is a engineer for Motorola's networking division, and he even confirmed that the algorithms that are used with the Motorola and other manufacturer modems are no where near as accurate as the calibrated equipment that is used by techs in their labs, and those shops that actually take the time to get them properly calibrated.

But what am I to know, even though I grew up around this stuff and worked with calibrating electronic equipment, have a better understanding than you do on how test equipment works, along with the algorithms that are used by said test equipment. But hey, this IS the Internet, so anyone can make stuff up, so why keep playing with you kid.

Anyway, back on topic: in a properly designed and operating cable system, you should have enough signal for a passive 8-way split. Assuming a normal drop length and normal distribution length. If you need an amp, then something ain't 'normal'.

I'm sorry you've had such faulty equipment in your life. I am however glad you and your tech friend were able to find humor in it. Congratulations to your brother in-law. Sounds like a nice job. While it is true that one would expect more accuracy from a $10,000 spectrum analyzer than from a $100 modem, that kind of precision isn't really needed if the readings from both devices agree. I really can't fathom how you could possibly know that you know more about test gear than I do as I have never met you, so I'll just have to take your word for it. Meanwhile, I'll stand by original suggestion that the OP try it without an amp.

That link, http://192.168.100.1 (generic address configured by the manufacturer, just like routers or any other piece of network gear), gets you into the cable modem (most manufacturers) where you can see exactly what is going on with it -- status, signal level, configuration, MAC addresses, and logs. It is as accurate as any $1000 or $5000 SLM. If you ever called tech support for a cable modem issue, the first thing they do is log into your cable modem to see what is going on with it.

If a tech comes out, of course he's going to throw on his SLM. Standard Operating Procedure. It's usually quicker and easier than turning on a computer, loading a browser and typing in a URL. And if there's no signal hitting the cable modem, that page probably won't come up, so he's just wasted time. Plus, with a meter, he can take a look at other frequencies on the system, such as adjacent channels, low end of the spectrum, and pilots to determine if what he's seeing is proper. I mean, in a 256QAM system, the DOCSIS carrier is -12dBc. That's 12dB lower than an adjacent analog carrier. Let's say that the bandwidth of a system is 750MHz (116 NTSC channels), with an anaolog channel (pilot) at channel 116 and that the cable modem is at channel 115. If, according to a meter, the signal level of channel 116 is +10dBmV, the signal level of channel 115 will be -2dBmV. There may be a slight variance, but very little. Load up the diagnostics page on the cable modem and he will see -2dBmV. The only limitation is that channel 115 is the only channel the cable modem can see. It doesn't care about anything else -- just channel 115.

It also tells you what the SNR (Signal to Noise Ratio) is, as well as the transmit level. Not just estimates, but actual levels. All on one page. To see the SNR with a meter, you have to change functions on the meter and hit a few more buttons, and to see what it's putting out, you have to disconnect the cable from the meter and connect the meter to the back of the cable modem and hit some more buttons on the meter to get to the right channel. If it's a DOCSIS 3.0 system, you have to go through 4 different channels to see what's going on.

And since the modem relies on telemetry from the headend to set the return level the diagnostic page is the simplest way to determine the transmit level, as the modem won't even transmit until it's connected to the cable. It beats putting a DC-10 in the line and having to add ten (the tap leg loss) and then subtract a dB (the through leg insertion loss).

Good point.
Edited by egnlsn - 7/8/12 at 9:22am

Is that why you post?