While updating and streamlining the hardware setup improved the situation, why it achieved this welcome outcome was less clear. And then there was the truly “shocking” discovery…
Last time, in part 1 of this series, I gave a historical overview of my longstanding broadband and television relationship with Comcast (aka Xfinity, the company’s brand for consumer products and services), focusing on the service in my most recent (and current) Colorado residence.
Specifically, I discussed the increasing frequency and severity of service “drop” issues my wife and I experienced subsequent to inadvertent cabling damage consecutively done by our community’s water and sanitation service provider last October and our power company earlier this summer.
At last week’s writeup’s conclusion, I was awaiting the arrival of yet another Comcast technician, subsequent to a recent service issues-escalation to a near-daily cadence (a particular problem given that my wife and I both work from home), an on-site visit which I’d insisted on sticking with in spite of the service’s as-usual temporary resurrection later that same evening. The on-site technician visit wasn’t scheduled until the next afternoon, but he called me mid-morning that (next) day and asked if he could arrive early, a rare deviation from the “late arrival” norm.
Competence is the best
Appreciative of his promptness and hopeful that I could still persuade him into making the debug session gratis for us, I asked him to delay his arrival only 45 minutes until my wife and I had both wrapped up a few work to-dos. He happily obliged, subsequently pulling up on the street in front of our house right on time. I’m not sure whether he was an “official” Comcast employee or a contractor, not that it really matters; his vehicle was a minivan with no corporate markings on it, albeit filled with Xfinity-branded tools, cabling, and other equipment.
I identified myself as a “techie” and asked if I could follow him around, periodically picking his brain and more generally absorbing through observation at least a bit of his conceptual plus experience-accumulated knowledge. He was happy to oblige: “this’ll be fun” were his exact words. And indeed he was smart and nimble of mind, not to mention enthusiastic; a delightful experience, all in all. But he also was in a hurry, so I strove to practice appropriate question-cadence restraint.
The first notable thing I learned from him was that the strand of coax coming out of the ground around the back of the house was completely unrelated to the plastic box labeled “Comcast” attached to the wall on the other (front corner) end of the house, first mentioned and pictorially shown in part 1 of this series.
This cable ran straight to the tap near the street. I assumed it was the mysterious “other line of service that the prior owner had installed” mentioned by the realtor more than a decade earlier.
Speaking of taps, not to mention the RF amplifiers I mentioned last time, along with other constituent building blocks of a hybrid fiber-coax system such as Comcast’s, check out this block diagram from Wikipedia’s as-always thorough, accurate and otherwise excellent entry on the topic.
I don’t know, by the way, where the fiber-to-coax conversion node(s) is/are located in our community, either in general or to what degree of proximity to my residence location.
MIA (not to mention obligatory)
The second thing the technician immediately noticed and I learned, to our mutual great dismay, was that this length of coax was nowhere properly earth-grounded nearby where it entered my home. The plastic box, which I was now realizing was likely associated with the legacy Comcast service likely stretching back to the original owners, was grounded.
As was the tap near the street.
But here? Nothing. Thereby representing not only potentially serious fire and electrocution danger for the home and its occupants but also suggestive of why I’ve endured multiple bouts of electrical shock-induced equipment destruction in my so-far time here (ironically, by the way, there’s a summer-monsoon lightning storm underway as I type this).
Then he saw the three-way splitter attached to the outer wall, which predated (by how long I have no idea) the initiation of my home ownership. Attached to the input end was the PoE (which stands for point-of-entry in this particular case, not power over Ethernet) filter I’d subsequently added for network security purposes back when I was experimenting with MoCA, and which I’d neglected to remove afterwards.
And attached to the splitter outputs were short spans of coax that ended up at wall connectors in two downstairs bedrooms (one each), along with a longer third coax span that also ended up inside, this time in the furnace room, where it split one more time and fed both the CableCARD receiver and cable modem.
After shaking his head and muttering “I don’t know how you even have any broadband service at all” under his breath, he fired up his smartphone, on which was installed a Comcast-proprietary (presumably) app that enabled him to live-monitor my cable modem’s statistics over its WAN connection. Once again, this time louder and more emphatically, he said (and this time also gestured), “I don’t know how you even have any broadband service at all” and invited me to look at the phone display for myself.
I noticed that the upstream transmission power ratings for the various in-use channels were all in the upper 50 dBmV range. I didn’t know much (and still don’t know as much as I’d like) about cable systems, but I knew enough to realize that this wasn’t a good thing.
Swaps and simplifications
Again, I realized he was in a hurry, so I refrained from asking a bunch more questions. He indicated that he thought the existing splitter was left over from a long-past satellite television system installation. I don’t quite buy that theory; although the DC power pass-through support and broader top-end frequency afforded by such a splitter (2.4 GHz vs 1 GHz, with a common 5-MHz bottom end of the range), the latter theoretically allowing for neighbors’ latest-generation MoCA signals to “leak” into my setup, might destructively interfere with my DOCSIS 3.1 modem. So, the PoE filter I’d inadvertently left installed would (or at least should) have alternatively blocked neighborhood noise.
That said, the splitter and filter were admittedly ancient, both also containing passive RF circuitry. So, perhaps something inside either or both had gone awry with advancing age and longstanding lightning, moisture, temperature, and other ambient environmental exposure.
He asked if I was currently using the downstairs bedrooms’ splitter outputs. When I replied in the negative, he wholesale-replaced the splitter with a grounding block (and wire) containing an integrated PoE filter, leaving the other two previous splitter-supplied coax feeds detached.
He also swapped out the connectors on both ends of the new grounding block-plus-PoE filter combo. Then he revisited his smartphone app and happily reported that the data he was seeing was “still not great, but much better now, and good enough”. Again, I got only a brief glimpse of the screen, but enough to confirm that the upstream channels’ power measurements were all now comfortably in the lower end of the 50s dBmV range.
I wish they were even lower than that, but again, we’re at the end of the neighborhood “loop” line. And it’s been more than three weeks (as I write this) since his visit, with not a single service drop, so in the spirit of “perfect is the enemy of good”…I’m good!
In case you were wondering, by the way, I immediately ordered, and subsequently installed as soon as it arrived, a grounding rod to attach to the other end of the grounding block-and-wire he’d generously given me.
The sticker on the side of the rod’s snug plastic packaging was awesome.
When I walked him back to his car (passing by the mysterious plastic box marked Comcast on the way, of which he had no knowledge and no spare time to further explore that day), he gifted me a bunch of extra hardware—MoCA filters, grounding blocks, 2- and 3-way splitters, and a 10’ span of high-quality coax cable with connectors on both ends—all of which I’ll be discussing more in next week’s finale (as planned, but who knows for sure) to this series.
In the upcoming concluding post(s), I’ll suggest some possible theories as to why (and to what degree, straight from my cable modem’s logs) his efforts bore fruit. I’ll also share the results of my subsequent sleuthing regarding the aforementioned mysterious plastic Comcast box and the various cables running into and out of it, as well as what was within it. Stay tuned for that and more; for now, please continue to share your thoughts in the comments!
—Brian Dipert is the associate editor, as well as a contributing editor, at EDN.
Related Content
- Debugging intermittent Comcast, part 1: Scenario-setting
- Will a location change improve my MoCA?
- The whole-house LAN: Achilles-heel alternatives, tradeoffs, and plans
- A quest for faster upstream bandwidth
- Lightning strikes…thrice???!!!
The post Debugging intermittent Comcast, part 2: Remediation details appeared first on EDN.