Perhaps the 3801 isn't the only equipment being used. A quick search found
this posting from three months ago.
It references the AT&T 5268AC router. It appears to be made by Arris. This
seems to be a voice gateway with wifi.
https://forums.att.com/t5/AT-T-Internet-Equipment/Radio-frequency-interference-from-U-verse-equipment-on-Amateur/td-p/4888880
Art NK8X
ᐧ
On Sun, Dec 18, 2016 at 6:53 PM, Guy Olinger K2AV <k2av.guy@gmail.com>
wrote:
> On Fri, Dec 16, 2016 at 2:38 PM, Art Snapper <art@nk8x.net> wrote:
> > There are numerous reports on the web from hams experiencing RFI from
> and to
> > Uverse. My main point is AT&T was complaining about receiving RFI,
> without
> > keeping their own house clean.
> > 73,
>
> I'm sorry, Art. Just can't go along with you here. Nothing personal.
>
> I spent some time looking up hams, RFI, Uverse, etc., based on your email.
>
> An awful lot of the reports I found are over five years old, They
> predate discontinuance of the 2-Wire (brand name) model 3800 gateway
> box that AT&T used heavily on early Uverse.
>
> That actually is a big clue, because it verifies what I've always
> thought, that the 2-Wire 3800 model gateway box was the source of a
> lot of their troubles. The changes between a 3800 and 3801 were
> extensive, including a massive increase of buffering memory (10 or 20
> to 1 increase?) to improve the built in error correction. There was a
> long list that our AT&T mentor told me about, and their experience was
> that the 3801 could soak up a lot of trouble out in the cables. The
> drop-off in troubles attributable to the gateway is testimony to the
> success of that project, as is my and my neighbor's experience with
> the upgrade.
>
> The 3800 was severely susceptible to strong ham signals on 160/80/40,
> with certain frequencies capable of making the gateway reboot with 20
> watts on the antenna, much less 1.5 kW. My old problem of rebooting my
> and my neighbor's 3800 gateway with QRO on 160 was finally eliminated
> with the 3801, which had just come out, and at that time was available
> in restricted quantities. This time they were not going to lead off
> with mass deployment. And they were keenly interested in my case,
> described below
>
> Several reports that looked current (2016 in email header) in
> references I saw, in the text were actually referring to much older
> events. One 2016 email was complaining about something the email
> itself said occurred in 2001. I thought that was interesting, still
> complaining about something that happened 15 years ago.
>
> The 2-Wire 3800 is no longer manufactured, the Uverse tech support
> "book" now has a list of customer complaints that automatically get an
> existing 3800 replaced, regardless. No clue as to how many have been
> deprecated. Nobody on my block has one anymore. Read on.
>
> They have gone to a practice of using coax from the interface where
> the cable comes up from underground to the gateway box. Twisted pair
> is no longer used for that. With certain repair activities, they will
> automatically replace older twisted pair from cable entrance to
> gateway box with coax.
>
> AT&T has a group, headquartered in New Jersey that is a national group
> specifically formed to deal with ham radio issues with Uverse. It's
> director is a well-known east coast ham, and they have the authority
> to force local repair groups to follow certain procedures in
> ham-related cases, and mark certain customers for automatic referral
> to their group.
>
> Anyone who has a current problem of this sort, should get in touch
> with the ARRL RFI task force, who will do an entry level triage to
> eliminate common issues, and then give cases that pass triage to the
> AT&T group I am talking about.
>
> They were good with my problems, really good.
>
> They simply do not fit your description, and I do not see current
> timeframe references on the internet to paint them with the colors you
> are using.
>
> What good does it do us to push and haggle to get a major company to
> do the right stuff, and when they do, keep on beating them up
> regardless, bringing up things that have been fixed for years?
>
> In my case, 2010-2011...
>
> 1) I had problems with temporary main telephone cable routing due to
> the construction of NC Toll 540 (extension of I-540) and its
> intersection with US 64. At some point ALL the old cable was replaced.
> Things were dicey with cables where they were moving roadways back and
> forth to maintain traffic while bridges were being built and old
> roadways dug up and replaced with heavier construction to interstate
> specifications.
>
> 2) The 650 foot buried drop cables from the pole out on US 64 service
> road to both my and my neighbor's house were replaced and reburied
> using a different route. Since 1980, when my house was first built,
> widening US 64 from 2 to divided 4 lane roadway, stuff related to the
> 540 highway, including some mistaken contractor digs, and a couple
> cable-melting lightning strikes, over time had put so many splices in
> the drop, that the cable itself was way out of spec and a problem.
>
> 3) Neighbor's in-house wiring was cleaned up and rerouted to reduce
> lengths and avoid some wear points.
>
> 4) The Gateway box in my house was moved to a spot recommended by AT&T
> after walking around the house describing the ethernet lines. This
> replaced a 120 foot ethernet run from cable entrance to gateway with a
> scant 8 feet, with the gateway in my wife's office. The cable entrance
> is directly underneath the window in her office. AT&T did most of the
> crawlspace crawling around under the house, even though technically it
> was my problem and my expense.
>
> 5) Telco Common mode current blocks designed for the frequencies and
> cables in use were added at strategic points on the ethernet around
> the house. They were better than my #31 material FT240 choke wraps
> made after K9YC specifications. (That is NOT a dump on K9YC.)
>
> 6) 3800 gateways at both houses were swapped out for 3801's
>
> 7) These items were done one at a time followed by a retest at 1.825
> watching changes in power levels necessary to reduce error correction
> counts and reboots. We did this in the wee hours so we wouldn't knock
> the neighbor off her IP-phone line and internet. She worked a lot
> logged in to VPN from home and had a lot of conference calls.
>
> This stretched out procedure kept me at low power until all fixed, but
> it DID allow evaluation of individual steps taken, which they were
> very happy to have.
>
> If I live to be 100 and keep Uverse until my last gasp, there is no
> way they will ever recoup the money they spent fixing my
> extraordinary, ridiculous case from the monthly Uverse bill. But they
> did get a bunch of proven facts to use on procedures in their "book"
> for these kinds of troubles.
>
> Again, anyone having RFI trouble either way with an AT&T Uverse
> device, getting confused looks from local tech people who don't know
> what to do, should go via the ARRL RFI task force.
>
> 73, and a great holiday season to all.
>
> Guy K2AV
>
_________________
Topband Reflector Archives - http://www.contesting.com/_topband
|