>If you don't know your engineered desired input, ask the person in your system who dictates these things. Be sure you understand if the number you're given is the actual downstream input of the data carrier, or the adjacent video carrier level. Sometimes it's easier to use an adjacent video level, since the received level of the data carrier will be relative, and some installers might not have a digital carrier compatible meter. The data carrier can always be looked at separately, if you expirience problems, anyway. If you are given the data carrier level, ask how many dB down from which adjacent channel, and do the math, if you will be using the adjacent video method. You can always take the measurements yourself at a trusted point within the system and use the offset as a relative figure, and apply a manufacturers suggested level to your formula, if you can't get the answer from someone.....You get the picture...... Anyway, I would use a DC before the splitter, utilizing the tap leg to achieve the desired level at the modem input, using the method described above. Many systems are engineered well enough that a given DC value may be pre-determinable, given that tap output levels are designed to take drop length into consideration, and give somewhat consistant levels at the end of the drop. The desired input to most modems is a fairly large window. Using a DC only robs the tv's in the house of a dB and a half or so at worst, and helps isolate somewhat the modem line from the rest of the house. It gives consistant results, and uses relative numbers to ensure correct return path levels, making your job easier. Correct input should calculate to correct return in relative terms, remember most house amps don't have return anplification, though. If it turns out the drop length will only squeek out enough to hit the modem correctly, you can always use the DC through leg to the modem, and amplify the tap side going to the tv's for levels, if needed. Departures will usually be exceptions though, not rules. A little up front prognostication will save a lot of trial and error futility and repeat trouble calls. Get yourself a data book from SA or Blonder-Tongue, or both, it will give you loss figures for DC's and a whole lot more info that helps answer problems like these and will make you a better technician.
>
>>
>>
>>>I install a lot of cable modems. Have installed several kinds but the ones I'm havinf trouble with is the GI Surfboards. The cable modem is external and connects to an ethernet card.etc,etc,etc,
>>>the problem is that when it is hooked to a four way splitter or eight way the modem becomes unstable or works intermittingly. I use primarily Regal Gold 120db EMI or Tru Spec 120dB EMI. The modem would work fine thru a 2way or three way but 4's and 8's would kill it. The RF being sent back was 19MHz and it measured 45db before the 4 ways and 45 dB out of the four way input. This was very strange and have no explanation for it.Has anyone come across this before or know why this happens. It is easily gone around to get it going again but the WHY is peckin at me. help if ya can. Thanks
>>>
(-REDACTED-)
>>
>>I do not think it's a RFI problem. I think it is a problem of too much attenuation on the reverse path AND
>>a return loss problem when the 4 and 8 port splitters are connected. 45 dB out of the modem is plenty
>>hot, however after a 4 and 8 way split you maybe at below threshold to hit the reverse laser with correct
>>input. Your possible "fix" is a seperate drop to the modem or a drop amplifier with an active reverse. Thats
>>my take from the Teleprompter Cable TV class of '69.
