Interesting and major DSL performance problem in AT&T West

Interesting and major DSL performance problem in AT&T West http://www.dslreports.com/forum/r23284439 en Tue, 09 Feb 2010 23:57:09 EDT Tue, 09 Feb 2010 23:57:09 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23363529 daveinpoway : The Netopia 3300-Series datasheet (»www.netopia.com/equipment/produc···s_ds.pdf) says that the 3346N can handle ADSL2+.]]> http://www.dslreports.com/forum/remark,23363529 Thu, 19 Nov 2009 07:35:54 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23362131 jdir : On my ADSL2+ modem

it shows that ANNEX A/B/L/M is available

when connect to ADSL - it just display ANNEX_M]]> http://www.dslreports.com/forum/remark,23362131 Wed, 18 Nov 2009 21:12:06 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23361953 ejg1 : We were trying to figure out whether the 3346n-002 was a ADSL2+ modem. It looked to me like it was as I posted a little earlier. Nobody seems to be able to prove it here though.

Would you happen to know?]]> http://www.dslreports.com/forum/remark,23361953 Wed, 18 Nov 2009 20:39:20 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23357954 daveinpoway : I noticed on my line that a Netopia 3346N-002 (TI chip) will report a downstream SNR Margin that is about 7 dB higher than a SpeedStream 4100 B (also uses a TI chip) reports. On the other hand, a newer 3347-02 (Infenion chip) reports about the same downstream SNR Margin as the 4100 B. ]]> http://www.dslreports.com/forum/remark,23357954 Wed, 18 Nov 2009 07:22:36 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23357504 ejg1 : Netopia-3000/22893556> show dsl all

ADSL Line State:...........Up
ADSL Startup Attempts:..1
ADSL Modulation:..........DMT
Datapump Version:.........DSP 4.2.1.0, HAL 4.2.1.0

.............................Downstream Upstream
.............................------------ ----------
SNR Margin:...................18.00.......17.00 dB
Line Attenuation:............30.50.......16.00 dB
Output Power:................13.85.......11.90 dB
Errored Seconds:...............0...........0
Loss of Signal:...................0...........0
Loss of Frame:...................0...........0
CRC Errors:........................4...........1
Data Rate:.......................6016........768

Netopia-3000/22893556>]]> http://www.dslreports.com/forum/remark,23357504 Wed, 18 Nov 2009 01:23:30 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23356855 wayjac : Oh, well the netopia you have won't show stats in its cli interface
Would you mind posting the output of the show dsl all command]]> http://www.dslreports.com/forum/remark,23356855 Tue, 17 Nov 2009 22:08:20 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23356800 ejg1 : I went through every one of those telnet options and I see no loop length nor max rate numbers. I could not find anything more on the web than what I posted earlier either.

Al I know is the SNR stats look a lot better than the speedstream 4100 I was using earlier. THis is not proof of the Netopia being a adsl2 modem though.]]> http://www.dslreports.com/forum/remark,23356800 Tue, 17 Nov 2009 21:56:01 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23356529 wayjac : I use xp and couldn't understand no telnet :o
I didn't know vista had telnet disabled]]> http://www.dslreports.com/forum/remark,23356529 Tue, 17 Nov 2009 20:56:23 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23356505 wayjac : show all-info
should give you all the line stats

I don't see a command that looks like it would show the modulations the modem supports, and I don't want you to misconfigure something by mistake, given some time you may get more familiar with the cli

but if a maxsync rate is displayed and it's above 8128kbps that's proof that the modem supports adsl2/adsl2+]]> http://www.dslreports.com/forum/remark,23356505 Tue, 17 Nov 2009 20:51:23 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23356430 ejg1 : A google search says: If you are using Windows XP the telnet command is available by default. But if you use Windows Vista the telnet command is disabled by default.

Anyway, I followed a procedure to enable telnet on my system and I have logged in. Here are a list of commands:

arp......................to send ARP request
atmping................to send ATM OAM loopback
clear....................to erase all stored configuration information
clear_log...............to clear stored log data
configure...............to configure unit's options
diagnose................to run self-test
download...............to download config file
exit.......................to quit this shell
help......................to get more: "help all" or "help help"
install.....................to download and program an image into flash
license....................to enter an upgrade key to add a feature
log.........................to add a message to the diagnostic log
loglevel...................to report or change diagnostic log level
netstat...................to show IP information
nslookup..................to send DNS query for host
ping........................to send ICMP Echo request
quit.........................to quit this shell
reset.......................to reset subsystems
restart.....................to restart unit
show........................to show system information
start.........................to start subsystem
status.......................to show basic status of unit
telnet........................to telnet to a remote host
traceroute..................to send traceroute probes
upload........................to upload config file
view..........................to view configuration summary
who............................to show who is using the shell
hotspot.......................to set or show hotspot authentication info
clear_certificate............to clear stored SSL certificate
?.................................to get help: "help all" or "help help"

Netopia-3000/22893556>help show

Netopia-3000/22893556> help show
Use "show" to show system information.
Follow it with:
all-info.......................to display all system information at once
atm...........................to display ATM information (detail with "all")
bridge followed by:
....interfaces...............to display bridge interfaces (detail with "all")
....table......................to display bridge table
config.........................to display current configuration
crash..........................to display current crash-dump information
dsl..............................to display DSL statistics
diffserv........................to print out the diffserv stats
dhcp............................followed by:
....server......................followed by:
........leases..................to display DHCP server lease table *
........store....................to display DHCP server non-volatile storage
....agent.......................to display DHCP relay-agent leases
....client........................to display DHCP client leases
enet.............................to display ethernet statistics (detail with "all")
features.........................to display available features
group-mgmt....................to display IGMP Snooping Group Addresses
ip..................................followed by:
....interfaces...................to display IP interfaces
....routes........................to display IP route tables
....arp............................to display IP ARP cache
....igmp...........................to display IGMP Group Addresses
....ipsec..........................to display IPSec Tunnel statistics
....firewall........................to display Firewall statistics
....state-insp....................to display Stateful inspection statistics
....lan-discovery................to display LAN Discovery table
log..................................to display next segment of the log (or "all")
memory............................to display memory usage (detail with "all")
ppp..................................to display PPP information *
pppoe...............................to display PPPoE information *
security-log.......................to display security log
status...............................to show basic status of unit
summary............................to show summary of current configuration
wan-users..........................to show WAN users (detail with "all")

* More complete help is available for these commands.

Netopia-3000/22893556>]]> http://www.dslreports.com/forum/remark,23356430 Tue, 17 Nov 2009 20:37:03 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23356045 wayjac :

said by ejg1

:

I did a search on telnet from the C: drive. Nothing pops up... Should it?

I don't know what to say
If I search for telnet I get this
[att=1,l]

]]> http://www.dslreports.com/forum/remark,23356045 Tue, 17 Nov 2009 19:20:15 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23356008 wayjac : You should be able to open a command prompt window
command prompt window example

and type telnet 192.168.1.254 then login and use the command show dsl if that's not it type help and you'll get a list of commands]]> http://www.dslreports.com/forum/remark,23356008 Tue, 17 Nov 2009 19:11:37 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23355929 ejg1 : I did a search on telnet from the C: drive. Nothing pops up... Should it?]]> http://www.dslreports.com/forum/remark,23355929 Tue, 17 Nov 2009 18:48:44 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23355594 wayjac : That's odd windows and mac have a builtin telnet client]]> http://www.dslreports.com/forum/remark,23355594 Tue, 17 Nov 2009 17:38:45 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23355130 d_l : It should be #1 now unless you have problems with your drop line or nearby out on the cable bundle.]]> http://www.dslreports.com/forum/remark,23355130 Tue, 17 Nov 2009 16:07:18 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23355123 ejg1 : I never watched to see if the Hybrid number changed from 2 to possibly 1 after running off of the homerun line. Next time I slap the 4100 on I will give it a look.]]> http://www.dslreports.com/forum/remark,23355123 Tue, 17 Nov 2009 16:05:48 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23355099 ejg1 : I looked for loop length and max rate in the HTTP log dump interface. I dont see any references to these anywhere.

I dont know how to use the telnet interface. From the cmd window TELNET gives me a "telnet is not recognized" response. I am willing to be coached though.]]> http://www.dslreports.com/forum/remark,23355099 Tue, 17 Nov 2009 16:01:57 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23349851 d_l : I think it also shows something that is equivalent to no bridged tap present vs. a bridged tap present. Translated to what the 4100 shows, that would comparable to hybrid #1 vs. hybrid # 2, 3, or 4.]]> http://www.dslreports.com/forum/remark,23349851 Mon, 16 Nov 2009 16:17:46 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23349778 wayjac : Do you know if your netopia will show a estimated loop length and max rate in the telnet interface
Telnet should show all the modulations the modem supports]]> http://www.dslreports.com/forum/remark,23349778 Mon, 16 Nov 2009 16:02:22 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23349340 d_l : The DSL sync performance differences between the 4100 and the Netopia are probably similar to those between the 4100 and 4100b: »What are the differences between the 4100 and the 4100b?]]> http://www.dslreports.com/forum/remark,23349340 Mon, 16 Nov 2009 14:38:23 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23349226 ejg1 : The modem says:
Netopia Model 3346N-VGx DSL Ethernet Managed Switch
Running Netopia SOC OS version 7.5.1 (build r10)
Multimode ADSL Capable
(admin completed login: Full Read/Write access)
Serial number 22893556, CPU MIPS4KEC, Product ID 1217

So yes... this is a ADSL2+

Cut and past from a different thread:
Router has two main chips:
Kendin KS8995XA Ethernet switch
TI TNETD7300AGDW Controller
Two other chips are RAM and Flash

]]> http://www.dslreports.com/forum/remark,23349226 Mon, 16 Nov 2009 14:19:15 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23347430 d_l : Is the Netopia ADSL2+ compliant? If it is, then the noise margin on such a modem is reported differently than an ADSL-only modem such as the 4100 and the method used always reports a higher noise margin which isn't necessarily indicative of stronger signal (the ADSL2+ modem's are "grading on a curve"). The ADSL2+ modems also will report a lower attenuation due to a more restricted calculation method.

Also does the Netopia use a TI AR7 chipset? That is the best available modem chipset for most line conditions.]]> http://www.dslreports.com/forum/remark,23347430 Mon, 16 Nov 2009 08:32:59 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23347001 ejg1 : Well this is interesting...

I have two different modems that I can compare line stats from.
One is the Speedstream 4100, the other the Netopia 3346n-002. The statistics are vastly different from each other. Is it fair to say the Netopia would be better for me to use in my situation? I dont see why not...

----Speadstream 4100
.....................................To Modem........... To Internet
Max Allowed Speed (kbps).......6016..............768
SN Margin (dB)..................... 14.0 or 11.0......-
Line Attenuation (dB)..............33.5.................-
Loss of Signal...........................0..................-
Loss of Frame...........................0..................0
CRC Errors................................5..................2

----Netopia 3346n-002
........................Downstream Upstream
........................------------ ----------
SNR Margin:...................19.00.......17.00 dB
Line Attenuation:............30.50.......16.00 dB
Output Power:................12.50.......11.90 dB
Errored Seconds:..................0........0
Loss of Signal:......................0.......0
Loss of Frame:......................0........0
CRC Errors:..........................0........1
Data Rate:.......................6016.........768]]> http://www.dslreports.com/forum/remark,23347001 Mon, 16 Nov 2009 02:03:21 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23346370 wayjac :

said by ejg1

:

Can you tell me what needs to be watched?

Your stats are decent I think you would want them to stay as they are:
SN Margin in double digits
Attenuation will not vary much from what it is

Here are my stats
..................................to modem....to internet
Max Rate (kbps).................9000........1100
Max Allowed Speed (kbps)....6016..........768
SN Margin (dB)...................17.0..........18.0
Line Attenuation (dB)..........31.3...........17.5

I'll let someone else talk about what the stats mean]]> http://www.dslreports.com/forum/remark,23346370 Sun, 15 Nov 2009 22:24:29 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23346194 ejg1 : I have seen others post line stats but I really dont know what is important. I would like to learn about this.

A faq I just found for the reader is here: »AT&T Southeast Forum FAQ »How do I access the Westell line stats and logs? What do the numbers mean?

and these are the stats I think you are referring to.

...................................To Modem To Internet
Max Allowed Speed (kbps)....6016.......768
SN Margin (dB)...................14.0........-
Line Attenuation (dB)..........33.5........-
Loss of Signal.........................0........-
Loss of Frame.........................0.......0
CRC Errors.............................5........2

Thanks,
ejg1]]> http://www.dslreports.com/forum/remark,23346194 Sun, 15 Nov 2009 21:41:02 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23345680 wayjac : Ok, the home run goes to the splitter only

Nice job :D

Would you post your dsl line stats]]> http://www.dslreports.com/forum/remark,23345680 Sun, 15 Nov 2009 19:40:42 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23345614 ejg1 : I wanted to use the same CAT6E from the 1) nid to the splitter as I used running out from the 2) splitter HR line to the terminal jack.]]> http://www.dslreports.com/forum/remark,23345614 Sun, 15 Nov 2009 19:23:57 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23345348 wayjac : The unused test jacks are ok to be left as they are connected or disconnected

Why did you run two cables to the filter/splitter]]> http://www.dslreports.com/forum/remark,23345348 Sun, 15 Nov 2009 18:21:09 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23345165 ejg1 : Thanks d_l,

This is something ugly to think about… but whenever an antenna is involved which either does or doesn’t do a good job at absorbing RFI, the phenomena can be controlled to a high degree depending on controlling length of that antenna.

That is, the length of wire can be tuned (shortened or lengthened) to non-resonance lengths to remove these influences. For this problem, if the length of the wire is a multiple of the frequency-at-question-wavelength, the wire length nulls out the resonance properties. I have not sat down with a calculator to figure what the implications are across the whole frequency bandwidth. This means I did not think about applying this concept when I ran my HR line some days ago. Maybe if I every have issues down the line I will think about it.

In the mean time, maybe someone else less lazy than I can run the lambda numbers.]]> http://www.dslreports.com/forum/remark,23345165 Sun, 15 Nov 2009 17:29:27 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23345065 d_l : So only a slight improvement with the ground wire connected in your situation. Most of the improvement is in the upload maxrate too. OK, that is good info. Thanks!

Maybe your plasma TV is responsible for the sorta harmonic interference (it probably is harmonic, but it isn't as clear cut as some of the other examples we've seen) in the charts that you previously posted.

Well, we have another interference source culprit to ask about on the next problem thread that reports interference. ]]> http://www.dslreports.com/forum/remark,23345065 Sun, 15 Nov 2009 17:06:20 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23345026 ejg1 : Yes, the plasma was the loudest source by far... not even close.

Here is a pic how I did the ground wire. I ran the CAT ground into the peanut and tied it to two wires I twisted together to have more copper to work with. Then I ran that into the ground block. The same block has the fatter black copper wire I used to tie to the telco's nid ground. In this pic you can see how I pulled the wire off to do this comparison test.

After this pic are the ground comparison tests.

With groundwire attached:

Downstream Current Rate (Reported): 6016
Downstream Maxrate (Estimated): 8572
Downstream Bit Margin (Calculated): 476

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 1136
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:43 163.875 - 189.75 KHz
bit(s) 130 560.625 - 564.9375 KHz
bit(s) 155:161 668.4375 - 698.625 KHz
bit(s) 186:189 802.125 - 819.375 KHz
bit(s) 198:201 853.875 - 871.125 KHz
bit(s) 211 909.9375 - 914.25 KHz
bit(s) 222:223 957.375 - 966 KHz
bit(s) 234:235 1009.125 - 1017.75 KHz
bit(s) 241:247 1039.3125 - 1069.5 KHz
bit(s) 255 1099.6875 - 1104 KHz

Without groundwire attached:

Downstream Current Rate (Reported): 6016
Downstream Maxrate (Estimated): 8512
Downstream Bit Margin (Calculated): 476

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 1044
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:41 163.875 - 181.125 KHz
bit(s) 113 487.3125 - 491.625 KHz
bit(s) 130 560.625 - 564.9375 KHz
bit(s) 156:160 672.75 - 694.3125 KHz
bit(s) 172 741.75 - 746.0625 KHz
bit(s) 186:189 802.125 - 819.375 KHz
bit(s) 198:201 853.875 - 871.125 KHz
bit(s) 211 909.9375 - 914.25 KHz
bit(s) 222:223 957.375 - 966 KHz
bit(s) 225:226 970.3125 - 978.9375 KHz
bit(s) 234:235 1009.125 - 1017.75 KHz
bit(s) 240:247 1035 - 1069.5 KHz
bit(s) 255 1099.6875 - 1104 KHz]]> http://www.dslreports.com/forum/remark,23345026 Sun, 15 Nov 2009 16:51:52 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23344951 d_l : They are not connected in any way to your POTS line (lack of dial tone), so they would have no effect on the DSL signal]]> http://www.dslreports.com/forum/remark,23344951 Sun, 15 Nov 2009 16:30:28 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23344926 ejg1 : I plugged my phone into every one of those test jacks. Only the top one has a dial tone.

Is it ok if I keep the unused jacks unplugged? Can they act as antennas for RFI?]]> http://www.dslreports.com/forum/remark,23344926 Sun, 15 Nov 2009 16:23:35 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23339026 anon : If the other test jacks are hooked up ahead of the splitter they should be removed . If there is no dialtone they are OK and can be left alone.]]> http://www.dslreports.com/forum/remark,23339026 Fri, 13 Nov 2009 22:00:37 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23337896 wayjac : Only one of the test jacks has inside wiring connected to it
I seen more than a few nid's that are fully populated with test jacks and surge protectors

I doubt that the att cable from the "street" has more than 2 wire pairs connected]]> http://www.dslreports.com/forum/remark,23337896 Fri, 13 Nov 2009 17:04:25 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23337672 anon : You should only have one test jack per line. If you have more they act as antennas for RFI. Is there any dial tone on the other binding posts as seen in the picture?]]> http://www.dslreports.com/forum/remark,23337672 Fri, 13 Nov 2009 16:26:09 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23337269 d_l : My money would be on the plasma TV as the worst culprit.

Yeah, prospecting for RFI with a portable AM radio is always fun. ;) We have suggested it to quite a few people with problems and some have blown off the idea. Their loss.]]> http://www.dslreports.com/forum/remark,23337269 Fri, 13 Nov 2009 14:53:40 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23336568 ejg1 :

said by d_l

:

Was that final test with the shielding grounded or ungrounded?

I did not try to see if there was a difference by undoing the ground wire. I will do the test and post the results here. It should not take that long.]]> http://www.dslreports.com/forum/remark,23336568 Fri, 13 Nov 2009 12:47:52 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23336532 ejg1 : It was a combination of RFI sources and I verified all of them.

In the center of the garage there are three items all within 3ft of each other and I did not want to put time into distinguishing whether just one or all were contributors. The transistor radio test confirmed the emissions. I believe the the phone cable routing passes fairly close to the items after exiting the nid. Next time I climb up there I will make it a point to look.

The Garage items were:
1) the garage door opener,
2) the garage butler (Device that closes the garage door after a set time in case you have forgot)
3) florescent lighting for the garage.

Then in the house there are two HUGE RFI emitter sources. I isolated them to be:

1) Panasonic DMR-E20 DVD recorder
2) My brand new Panasonic 54in Vieta Plasma TV.

I sat there for an hour plugging things in and out to figure out where the issues were.

The transistor radio link you provided earlier really helped. Hopefully others don't read that link and intermediately blow it off because it does sound kind of hokey. But then you soon learn it works pretty good !!! You just rotate the radio around like a globe and if there is no static you are good to go. If you get static, move around until it gets louder and it will lead you to the source.]]> http://www.dslreports.com/forum/remark,23336532 Fri, 13 Nov 2009 12:42:29 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23335320 d_l : Was that final test with the shielding grounded or ungrounded?

What do you think is the source device(s) of all your previous interference? Any guesses?]]> http://www.dslreports.com/forum/remark,23335320 Fri, 13 Nov 2009 08:44:59 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23334375 ejg1 : Well… mission accomplished ! Everything operates as expected. Stick a fork in this. Here is what I did.

SETUP
The picture below shows the HR line fed directly from the nid into the garage. Remember I bought 250ft of CAT6E STP cable and I kept it all intact to validate everything was working right. After passing through 250ft of coiled cable, I terminated the other end with the RJ11 jack plate. It is hard to make out the jack plate in the photo because it is overexposed. From the jack into the Speedstream 4100 and then into the laptop.

BASELINE TEST:
1) Circuit breakers to the house turned off except for the garage breaker to power the modem and computer. This was done to eliminate all possible sources of RFI.

Speakeasy tests showed 5138/644 , 5139/643 and 5140/642. Modem synch details:

Downstream Current Rate (Reported): 6016
Downstream Maxrate (Estimated): 8540
Downstream Bit Margin (Calculated): 476

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 1074
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:41 163.875 - 181.125 KHz
bit(s) 125 539.0625 - 543.375 KHz
bit(s) 130 560.625 - 564.9375 KHz
bit(s) 141:142 608.0625 - 616.6875 KHz
bit(s) 154:162 664.125 - 702.9375 KHz
bit(s) 172 741.75 - 746.0625 KHz
bit(s) 186:191 802.125 - 828 KHz
bit(s) 196:201 845.25 - 871.125 KHz
bit(s) 211 909.9375 - 914.25 KHz
bit(s) 215:223 927.1875 - 966 KHz
bit(s) 234 1009.125 - 1013.4375 KHz
bit(s) 242:247 1043.625 - 1069.5 KHz
bit(s) 255 1099.6875 - 1104 KHz

FINAL TEST
2) Many tests were done in between this final test to make sure everything was stable. In summary, the HR line was rolled out to the final location and cut. Only about 20 feet of CAT6E was needed . The other 230 feet was laid to rest in the corner. The line was fed into the house, wired up at the outlet and everything buttoned up. I also want to mention everything in the house got turned on to create as much RFI as possible. Here are the test results:

Speakeasy tests showed 5138/646, 5138/649, and 5140/650. Modem sync details:

Downstream Current Rate (Reported): 6016
Downstream Maxrate (Estimated): 8508
Downstream Bit Margin (Calculated): 476

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 1136
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:42 163.875 - 185.4375 KHz
bit(s) 111:117 478.6875 - 508.875 KHz
bit(s) 130 560.625 - 564.9375 KHz
bit(s) 154:162 664.125 - 702.9375 KHz
bit(s) 172 741.75 - 746.0625 KHz
bit(s) 186:190 802.125 - 823.6875 KHz
bit(s) 198:201 853.875 - 871.125 KHz
bit(s) 211 909.9375 - 914.25 KHz
bit(s) 222:223 957.375 - 966 KHz
bit(s) 234 1009.125 - 1013.4375 KHz
bit(s) 241:248 1039.3125 - 1073.8125 KHz
bit(s) 255 1099.6875 - 1104 KHz

Thanks for the help of our esteemed poster D_L who was patient enough to get me through this.

G’Night and best wishes to everyone!]]> http://www.dslreports.com/forum/remark,23334375 Thu, 12 Nov 2009 23:10:42 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23334297 ejg1 : I dont really know.

I have 6 jacks in the house and noticed they were all wired (tapped connection) in parallel to each other using the same twisted pair. Why I have 5 other test jacks that were installed is a wonder to me too.]]> http://www.dslreports.com/forum/remark,23334297 Thu, 12 Nov 2009 22:51:30 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23330597 anon : Why are there so many test plugs in the picture? You only have one line at your house-or are you at an apartment nid?]]> http://www.dslreports.com/forum/remark,23330597 Thu, 12 Nov 2009 11:39:33 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23329873 d_l : I'm not sure how critical the correct colors are to the operation, but if you wire it correctly to the standard convention, someone else checking your wiring later won't have a problem.

»AT&T Southeast Forum FAQ »Telco Wiring Color Code.

I think you will have to splice an wire extension to the ground drain wire with a splice peanut and then "tie' the wire and splice to the cable to protect the fragile drain wire from flexure. ]]> http://www.dslreports.com/forum/remark,23329873 Thu, 12 Nov 2009 09:07:12 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23329325 ejg1 : Finally the cable got delivered to my door.

I wired everything up at the nid. The only difficult thing was grounding the shielded CAT6 as discussed earlier. The ground line is somewhat thin and fragile and if you move it around frequently it will break like the pop top ring on a soda can.

So the home run line is now hot and just sitting there all coiled up. I know I did something right because the phone in the house works! :-)

Time for a break since there is something I am not so sure of.

The nid is wired up exactly as described before but I have a question about polarity of the HR line.

I am only using a single twisted pair so does it matter if white goes to one terminal of the jack and blue the other? or vise verse?

]]> http://www.dslreports.com/forum/remark,23329325 Thu, 12 Nov 2009 02:46:03 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23312153 d_l : Isn't there a bare wire, a drain wire, inside the foil of the shielding? That wire is what should be grounded.]]> http://www.dslreports.com/forum/remark,23312153 Sun, 08 Nov 2009 21:41:56 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23312097 ejg1 : I validated there is a heavy duty ground in the Telco nid side.

I could tie the ground to that bar you referenced if needed. Right now it is not unused for anything.

I just dont know what to do with my shielded (STP) CAT homerun line and how/what way that should be grounded? ]]> http://www.dslreports.com/forum/remark,23312097 Sun, 08 Nov 2009 21:28:56 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23311877 mhetterm :

said by ejg1

:

Most likely there is a ground in there I can use. Except, I don't know what to do ground my system at the moment. When the time comes, one of you guys will poke me I hope.... :-)

That metal "strip" with 8 or so screws in it to the right of your NID, near the bottom, looks like a ground bus bar to me ...]]> http://www.dslreports.com/forum/remark,23311877 Sun, 08 Nov 2009 20:37:18 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23311632 wayjac : I don't know what's behind the telco cover on your nid......I think you should have a look]]> http://www.dslreports.com/forum/remark,23311632 Sun, 08 Nov 2009 19:40:31 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23311601 ejg1 : There is a proper ground in the forbidden land. :-)]]> http://www.dslreports.com/forum/remark,23311601 Sun, 08 Nov 2009 19:31:55 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23311512 wayjac :

said by ejg1

:

To open the door of the telco side of the nid you need a special hex bit that is hollow in the middle as there is a protrusion in the exact center of the screw to keep just anyone from using just any old hex wrench from opening the thing. Think I will run to the HW store to see if I can find one.

Most all nid's I've seen have that tamper-proof fastner but a pocket screwdriver will work just as well as the "special hex bit" and it won't damage the fastner

[att=1,l]

]]> http://www.dslreports.com/forum/remark,23311512 Sun, 08 Nov 2009 19:12:22 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23311459 ejg1 : Thanks D_L, It seems like we are on the same page with the hook-up then. I can't wait till my cable shows up !

hey would you happen to know if the telco could ever be persuaded to swap out their CAT3 for my provided CAT6e from the street to their nid?. I just love over killing !]]> http://www.dslreports.com/forum/remark,23311459 Sun, 08 Nov 2009 19:00:28 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23311433 ejg1 : To open the door of the telco side of the nid you need a special hex bit that is hollow in the middle as there is a protrusion in the exact center of the screw to keep just anyone from using just any old hex wrench from opening the thing. Think I will run to the HW store to see if I can find one.

Most likely there is a ground in there I can use. Except, I don't know what to do ground my system at the moment. When the time comes, one of you guys will poke me I hope.... :-)]]> http://www.dslreports.com/forum/remark,23311433 Sun, 08 Nov 2009 18:55:33 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23310128 wayjac : You can use the red/green binding post on one or two of the unused test jacks for connection points, just remove the short rj11 from the ones you're going to use

There may be a ground connection behind the cover on the telco side of the nid. A pocket screwdriver will unscrew the fastner]]> http://www.dslreports.com/forum/remark,23310128 Sun, 08 Nov 2009 13:12:33 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23309617 d_l : I'm not sure if it was is necessary, but I wired my splitter more like this: »Pacbell FAQ »How do I install a splitter? which is a little different arrangement than the way wiring on the Keptels is usually run.

I used wiring "peanuts" to splice a short run of Cat5e from the splitter to the existing inside wiring because that IW couldn't be moved to connect to the lugs on the splitter. So my wiring essentially ended up like this:

T9 half of pair 1 in the short run of new Cat5e to T1
T8 half of pair 1 in the short run of new Cat5e to T2
W1 spliced to half of pair 2 in the short run of new Cat5e to T3
W2 spliced to half of pair 2 in the short run of new Cat5e to T4
Half of pair A in home run to T5
Half of pair A in home run to T6

Because that short run of Cat5e was exposed outside of the NID and the Corning Seicor box, I put a drip loop in it and covered it with a split wiring loom to protect it from UV rays.]]> http://www.dslreports.com/forum/remark,23309617 Sun, 08 Nov 2009 10:44:16 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23307938 ejg1 : My splitter arrived and after studying the links posted earlier I am ready to start wiring things up.

This is my understanding of the hook-up. Can you confirm whether it is correct?
T9 wires to T1
T8 wires to T2
W1 wires to T3
W2 wires to T4
Home Run LineA to T5
Home Run LineB to T6

I am also not clear about what we may need to do to ground the dedicated-shielded line.

]]> http://www.dslreports.com/forum/remark,23307938 Sat, 07 Nov 2009 20:41:20 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23297876 ejg1 : I am thinking the last module space at the bottom is going to be much too small to put the splitter. Since the splitter can be mounted outside the NIB and has its own weather proof box, I thought I would mount it on the same plywood the NIB is mounted on. Both the NIB and splitter are encased and away from the weather anyway by a screw-off door shield on the house.

I don't know what I should be using as a ground point... I would imagine something around the NIB has to be a ground point in case of lightning or something.]]> http://www.dslreports.com/forum/remark,23297876 Thu, 05 Nov 2009 17:23:02 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23296944 d_l : Is there room in that inset to mount the Corning Seicor box?

Edit: Also is there a grounding point available to ground the sheilding on the ST wire? Normally the shield is grounded, but in your case that might make it a big RFI antenna so you will have to try it with and wit out grounding.]]> http://www.dslreports.com/forum/remark,23296944 Thu, 05 Nov 2009 14:34:31 EDT Parts Ordered for Home Run line http://www.dslreports.com/forum/remark,23296752 ejg1 : OK I fixed the order... canceled.

I found another vendor who is shipping me out 250ft of Bulk CAT6E STP for $89 shipped. I figure since this is fairly hard to find in smaller quantities I can sell the leftovers on Ebay or some other poor guy that wants to do the same thing as I. »www.broadbandutopia.com/bubralcashca1.html

I could have gone a little cheaper with another vendor on CAT6 except money is not an issue. I like buying the best stuff available. One of my character flaws.

You mentioned that I should start studying the existing wiring from my NIB. So here are photos of how the NIB is currently wired.

I see two wires going into the telco side of the NIB. One is a big fat black wire, and the other a skinny grey wire.

Then on my side of the NIB, there are 6 testing modules (room for a seventh) of which only the top one has a bundle of wires going to it. From that bundle, a blue wire connects to the GREEN and a white with blue striped wire goes to the RED terminals. The other wires hang free and are not used. That whole bundle of wires leads up into the attic of the house. So can I assume all the jacks in the house are connected in parallel ???

By the way, the type of wire used is CRM or MPR verified UL CAT3

So… I am not sure how things are going to be wired up. I need to look at those pages you referenced again.

]]> http://www.dslreports.com/forum/remark,23296752 Thu, 05 Nov 2009 13:57:04 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23294178 d_l : Is there any way you can change the order for the patch cable? You really just want bulk cable for a home run as the RJ45 plugs would just have to be cut off. In addition, patch cable has stranded wires and you want solid cable to connect to your NID and splitter. Trying to handle those tiny wire strands on to screw connectors is maddening.]]> http://www.dslreports.com/forum/remark,23294178 Wed, 04 Nov 2009 23:49:37 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23293875 ejg1 : I bought a Corning/Seicor SPS-H70-SR1 instead from Hometech.com. at a pricy $73 shipped. (I did not want a knock-off)

The cable was easy although you have to watch the product descriptions like a hawk to be sure what you are getting. Cat6 Shielded 25ft PiMF STP Patch Cable 550MHz – White from iofast.com. $15 shipped

Both should arrive Friday/Saturday. I will start studying the existing wiring in the NID... let you know if I have questions. THANKS!

]]> http://www.dslreports.com/forum/remark,23293875 Wed, 04 Nov 2009 22:42:38 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23292704 d_l : You will probably have to order your splitter and cable on line:
»www.hometech.com/hts/products/ph···dex.html

Suttle makes a knock off of the Corning Seicor splitter, but I can't find a for sale link. It's much less expensive.

This guy sells shielded cable: »www.cat5ecableguy.com/inc/sdetail/14224 and I've bought it from him.]]> http://www.dslreports.com/forum/remark,23292704 Wed, 04 Nov 2009 19:21:12 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23292551 ejg1 : OK thanks

I just got back from various stores:

Target – Only Belkin CAT5e @ less than $1.00/ft
BestBuy – CAT6 only @ slightly more than $1.00/ft
Lowes – generic brand CAT5 @ $0.30 cents/foot
Central Computers – has tons of CATX in all lengths. CAT5, CAT5E, CAT6 at about $0.20 cents/ft

Problem is NONE of any of the products are shielded. The guy at CCS told me he could get me shielded by special order but there is a 50 ft minimum order. He should be calling me back to verify.

I want to buy a Corning/Seicor SPS-H50-SR1 but I can only find one place that sells them. The kicker is I have to buy a minimum of 5@$125. Got any ideas on where else to get one? ]]> http://www.dslreports.com/forum/remark,23292551 Wed, 04 Nov 2009 18:55:30 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23290681 d_l : Any grounding of spare pairs would be done at your NID so examine the inside wiring into it to see there the ends are. You need to do this before you start your home run any way in order to understand your wiring.

RJ45 is for ethernet plugs.]]> http://www.dslreports.com/forum/remark,23290681 Wed, 04 Nov 2009 13:34:20 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23290595 ejg1 : I don't know what wiring is inside the walls. I am guessing when the house was built at the end of 1999 they selected the cheapest stuff that would work. Most likely that is what is in my wall... the cheapest stuff. I should pull out an existing jack and the wiring behind and look.

I will get an RJ12 jack...]]> http://www.dslreports.com/forum/remark,23290595 Wed, 04 Nov 2009 13:21:54 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23290323 d_l : Cat5e should be more than adequate and is what you mostly see sold in hardware stores. They do have cat6, but I think most contractors use cat5e for POTS wiring now.

I'm guessing you have cat3 in use now. Is it 2 pair or more? I'm wondering if the unused pair(s) might be grounded which could make them act as pretty good antennas for any RFI. They would then couple with the used pair to put that noise in your line.

You would wire your home run to an RJ11 or RJ12 jack.]]> http://www.dslreports.com/forum/remark,23290323 Wed, 04 Nov 2009 12:39:40 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23290173 ejg1 : Topic 1
Your links have kept me quite busy. Lots of great stuff there. I just may be able to do this. There is one slot available for the POTS splitter as can be seen in the photo I posted earlier.

Good to go... I think.

Topic 2
I am reading a lot about what kind of cabling to use. It seems many people are looking to the future. Fewer are using CAT 3 cable and now going to CAT6.

Now if I understood things properly, the higher the CAT rating, the better the shielding and twisting properties of the copper wires inside. The length of the cabling I need is piddly so the costs for that are nothing.

I understand I will only be using a single copper pair from the CAT6 cable. Some might say it is overkill and a waste. But if the shielding properties are so much better, that is what I am after. I don’t care about unused lines of wire. I only care the wires are shielded/screened from any electromagnetics as possible.

Am I thinking about this properly?

I would think the CAT6 cable would be selectively wired to a RJ-45 jack. From there with the proper cable you have the short run into the modem.

Whatcha think?
]]> http://www.dslreports.com/forum/remark,23290173 Wed, 04 Nov 2009 12:12:53 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23288213 d_l : Telco doesn't mind. You will be working on your side of the NID.

»AT&T Southeast Forum FAQ »I don’t have a splitter. Can I use an in-line filter for a home run?
»Ameritech - SBC FAQ »NID Splitters and Installation
»AT&T Southeast Forum FAQ »What equipment and tools are required for a homerun? Where can I get them?
»Inside Wiring FAQ »Detailed Homerun Instructions

It is possible that the home run isn't the only solution. That's why you check with the radio first.]]> http://www.dslreports.com/forum/remark,23288213 Wed, 04 Nov 2009 00:49:03 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287865 ejg1 : It will take me a year to fumble around with a portable radio.

Is the pots splitter solution something a crafty guy like me can negotiate? Where do I find a POTS splitter and instructions how to go about this?

If I do it myself will I piss off the telco?]]> http://www.dslreports.com/forum/remark,23287865 Tue, 03 Nov 2009 23:09:48 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287830 d_l : That second line is a home run and you will need a POTS splitter (or in a pinch a microfilter can be used ) at the NID where you tested. The home run would need to be strung to avoid running too close (within several feet) to the house power lines in the walls or else you will duplicate your current noise problems on it.

You need to walk around with a battery powered radio to listen for interference noise.]]> http://www.dslreports.com/forum/remark,23287830 Tue, 03 Nov 2009 23:01:21 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287800 ejg1 : oh... I have no portable radios that are plugged in and especially not anything that can be working with the circuit breakers flipped off for the house.

I will have a look at that FAQ. Time to flip on the TV too...]]> http://www.dslreports.com/forum/remark,23287800 Tue, 03 Nov 2009 22:54:40 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287772 ejg1 : I think what you are saying is… if the max rate is as high as possible it implies my signal to noise will be as high as possible making it difficult for the same level of RFI to bleed into and degrade my DSL signal.

Is it possible to run a second line from my box to serve just the DSL signal inside the house??? I am not clear how to do that and not affect the same circuitry the house already has which knocks out some frequency bins.]]> http://www.dslreports.com/forum/remark,23287772 Tue, 03 Nov 2009 22:49:56 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287732 d_l : Forget about the speed tests. They are going to be about the same as long as you have full sync speeds, 6016, and a "decent" maxrate margin over the 6016 speeds, say at least 10% more. The decent maxrate margin implies that you have a reasonably high enough noise margin that your line isn't going to be plagued by CRC errors which would drive up packet loss and drive down your thoroughput speeds.

Did that explanation make sense to you? Keep the maxrate way up or another way to look at the same thing is to keep the noise margin high and you won't have problem.

Edit: Do you have a portable radio? »Ameritech - SBC FAQ »My DSL loses sync regularly. How can I detect interference with an AM Radio? Instead of following the FAQ to tune low on the AM dial, try tuning to to the mid-point frequencies of some of those notches. They might be AM radio stations, and your phone wiring is picking them up very well for some reason. ]]> http://www.dslreports.com/forum/remark,23287732 Tue, 03 Nov 2009 22:37:55 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287658 ejg1 : I am really bothered at these results. The electricity circuit breaker is off for most all the house. except the garage. Nothing can be running to be generating any interference.

But on the other hand... the speed tests are *really* close. The inside results might open the door for the TV RFI to make an impact though.

Speed tests at the outside box: 5144/657, 5138/654, 5142/651
Speed tests inside the house...: 5140/645, 5139/654, 5139/654

OUTSIDE

Downstream Current Rate (Reported): 6016
Downstream Maxrate (Estimated): 8584
Downstream Bit Margin (Calculated): 476

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 1074
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:41 163.875 - 181.125 KHz
bit(s) 130 560.625 - 564.9375 KHz
bit(s) 141:142 608.0625 - 616.6875 KHz
bit(s) 156:160 672.75 - 694.3125 KHz
bit(s) 186:190 802.125 - 823.6875 KHz
bit(s) 198:201 853.875 - 871.125 KHz
bit(s) 211 909.9375 - 914.25 KHz
bit(s) 222:223 957.375 - 966 KHz
bit(s) 241:247 1039.3125 - 1069.5 KHz
bit(s) 255 1099.6875 - 1104 KHz

INSIDE

Downstream Current Rate (Reported): 6016
Downstream Maxrate (Estimated): 6732
Downstream Bit Margin (Calculated): 424

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 1044
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:42 163.875 - 185.4375 KHz
bit(s) 128:134 552 - 582.1875 KHz
bit(s) 141:142 608.0625 - 616.6875 KHz
bit(s) 153 659.8125 - 664.125 KHz
bit(s) 155:166 668.4375 - 720.1875 KHz
bit(s) 169:174 728.8125 - 754.6875 KHz
bit(s) 176 759 - 763.3125 KHz
bit(s) 179 771.9375 - 776.25 KHz
bit(s) 181 780.5625 - 784.875 KHz
bit(s) 185:190 797.8125 - 823.6875 KHz
bit(s) 195:202 840.9375 - 875.4375 KHz
bit(s) 204 879.75 - 884.0625 KHz
bit(s) 206:208 888.375 - 901.3125 KHz
bit(s) 210:212 905.625 - 918.5625 KHz
bit(s) 220:227 948.75 - 983.25 KHz
bit(s) 229 987.5625 - 991.875 KHz
bit(s) 234 1009.125 - 1013.4375 KHz
bit(s) 236 1017.75 - 1022.0625 KHz
bit(s) 239:255 1030.6875 - 1104 KHz]]> http://www.dslreports.com/forum/remark,23287658 Tue, 03 Nov 2009 22:21:58 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287418 ejg1 : I am going to redo the tests because my laptop battery was running out plus I needed to turn on the lights in the garage to keep working.

I took a step even further. I flipped off the power breakers to the house. Only the small breaker for the garage is on. Now I feel more confident.

The goal is to use the same power chord plugged in at the same place for both tests. The first test will be at the outside box and the next at my standard outlet. With everything off they just have to be the same… I mean they just have to be!!! It should not take that long to do this and to post..]]> http://www.dslreports.com/forum/remark,23287418 Tue, 03 Nov 2009 21:34:15 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287370 d_l : Are there any outside lights on such as a sodium or mercury vapor lamp in the backyard?]]> http://www.dslreports.com/forum/remark,23287370 Tue, 03 Nov 2009 21:25:38 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287289 ejg1 : I am going to do this again. I forgot to flip the power switch on the desktop computer and monitor. The only things that will be on are the power supply for this laptop, the modem power supply, and the two 3 foot long florescents for the garage light.

Absolutely nothing else on.]]> http://www.dslreports.com/forum/remark,23287289 Tue, 03 Nov 2009 21:14:37 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287171 d_l : Look at the difference in that maxrate as compared to the two I posted before. There is something in your house that is injecting noise into your phone lines and the interference frequencies seem like they are harmonic (evenly spaced).

This might be something like a motor or some other triac-controlled device or perhaps an outside CFL light bulb controlled by a photocell.]]> http://www.dslreports.com/forum/remark,23287171 Tue, 03 Nov 2009 20:55:07 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23287078 ejg1 : ok here is my setup to show you what I am doing. I have abandoned the desktop computer and am using my laptop. It is easier to carry around. This means the power plugs were also pulled for the desktop and any peripherals. I cant think or find of a single item in the house that is plugged in.

One thing I wanted to mention is a month ago the power company installed a Smart Meter right next to my phone box. The gas company did the same thing but that is out of the picture and about 6 feet away. The Smart Meters supposedly communicate remote data back to the company. I don’t know how they are doing it.

You can see how I have run my own line off of the box and around the corner to the garage. Only the modem is plugged in and the laptop is off battery.

After booting up the modem, the three speakeasy tests gave: 5142/655, 5139/653, 5137/654. What I really like about this is the speedometer needle pegs to look rock solid. There is no wavering at all. A good sign.

Then I logged into the modem, downloaded the technician readout, and plugged it into the graph tool:

Downstream Current Rate (Reported): 6016
Downstream Maxrate (Estimated): 8564
Downstream Bit Margin (Calculated): 476

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 1136
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:41 163.875 - 181.125 KHz
bit(s) 130 560.625 - 564.9375 KHz
bit(s) 141:142 608.0625 - 616.6875 KHz
bit(s) 155:163 668.4375 - 707.25 KHz
bit(s) 186:189 802.125 - 819.375 KHz
bit(s) 198:201 853.875 - 871.125 KHz
bit(s) 211 909.9375 - 914.25 KHz
bit(s) 222:223 957.375 - 966 KHz
bit(s) 234 1009.125 - 1013.4375 KHz
bit(s) 242:246 1043.625 - 1065.1875 KHz
bit(s) 255 1099.6875 - 1104 KHz

Time to digest this I guess...]]> http://www.dslreports.com/forum/remark,23287078 Tue, 03 Nov 2009 20:35:20 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23286986 ejg1 : my home was built in Oct 1999. It was given provisions for an alarm system that I would not pay for to turn on, then the only people who monkeyed with the lines were the phone company

I am putting data together for you now...]]> http://www.dslreports.com/forum/remark,23286986 Tue, 03 Nov 2009 20:16:38 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23286954 d_l : I don't work for AT&T.

Probably some IW bridge taps: »SBC DSL FAQ »What do the hybrid numbers mean and what do they tell you about your line?

How old is your home phone wiring? Newer homes have Cat5e runs back to a punch down block and these seems to have the highest chances of hybrid 2 or 3 readings in the modem. The older daisy chained wiring might only have hybrid 1 or 2 readings if the total length of all the stubs is fairly short (say less than 100').

You might be able to disconnect the wire run to that jack off of a punch down block if you have a newer home.]]> http://www.dslreports.com/forum/remark,23286954 Tue, 03 Nov 2009 20:09:41 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23286689 ejg1 : I am not sure what a NID is but what I am refurring to is the box on the outside of the housewhere the phone comes in off of the street.

I thought I might take some aluminum foil and wall paper over the empty phone jack behind the TV to try and shield it some way. But first I really need a baseline reading so outside I go to gather data at the box.

What time are you off work???

Here is the modem log:
2009/11/03 21:52:23 GMT E |System |=============== SYSTEM UP ===============
2009/11/03 21:52:23 GMT E |System |Current Mode: PPP on the modem (Public IP for LAN device)
2009/11/03 21:52:23 GMT E |DSL |DataPump Version - 01.01.00.00
2009/11/03 21:52:23 GMT E |DSL |State: WAITING
2009/11/03 21:52:25 GMT E |Ethernet |Link 1 Up - 100Base-TX Full Duplex
2009/11/03 21:52:43 GMT E |DSL |State: INITIALIZING
2009/11/03 21:52:50 GMT E |DSL |HYBRID 2
2009/11/03 21:52:50 GMT E |DSL |Link up 1 US 768 DS 5728 (FAST:G.dmt)
2009/11/03 21:53:00 GMT E |PPPoE |tx PADI, id: 0000, ac: (NULL), sn: (NULL)
2009/11/03 21:53:00 GMT E |PPPoE |rx AC Name: 90084080500387-rback7.pltnca
2009/11/03 21:53:00 GMT E |PPPoE |tx PADR, id: 0000, ac: (NULL), sn: (NULL)
2009/11/03 21:53:00 GMT E |PPPoE |rx PADS id: 9161
2009/11/03 21:53:00 GMT E |PPP |LCP neg PAP
2009/11/03 21:53:00 GMT E |PPP |LCP up
2009/11/03 21:53:01 GMT E |PPP |IPCP nak option: 3
2009/11/03 21:53:01 GMT E |PPP |IPCP nak option: 129
2009/11/03 21:53:01 GMT E |PPP |IPCP nak option: 131
2009/11/03 21:53:01 GMT E |PPP |IPCP up ip: 71.146.21.153, gw: 71.146.31.254, dns: 68.94.156.1, 68.94.157.1
2009/11/03 21:53:04 GMT E |SNTP Client |Updated time from Primary server 132.163.4.103]]> http://www.dslreports.com/forum/remark,23286689 Tue, 03 Nov 2009 19:20:51 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23286631 d_l : If "go back to the phone box" means the NID, then yes absolutely. I was going to suggest that.

It would seem that your TV is amplifying the interference from some other source. Maybe the TV is broadcasting the the line noise it picks up into that phone jack as you suggested earlier.

What hybrid number is the 4100 reporting in the modem log?: »192.168.0.1/log.htm at the phone jack you normally use? What about at the NID? I'm wondering if your inside wiring has bridged taps that might act as antennas for noise. For example that jack behind the TV is probably an unterminated stub (aka bridged tap).]]> http://www.dslreports.com/forum/remark,23286631 Tue, 03 Nov 2009 19:10:21 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23286525 ejg1 : Everything in the house was unplugged except for the Oven's digital clock. (I was not going to pull the stove out :-/) The fridge was unplugged... everything! Only the modem, desktop, monitor, and power supply to the computer speakers were on. This is how I gathered the OFF data.

Then when I gathered the TV data I only plugged it back in to the 110V, rebooted the router, and ran the speed tests. Then I extracted the tech readout data.

I wonder if it makes sense to go back to the phone box and take some data there?]]> http://www.dslreports.com/forum/remark,23286525 Tue, 03 Nov 2009 18:53:36 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23286336 d_l : NO TV

»/users/deblin/···aph=true

Downstream Current Rate (Reported): 6016
Downstream Maxrate (Estimated): 6024
Downstream Bit Margin (Calculated): 412

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 982
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:47 163.875 - 207 KHz
bit(s) 118 508.875 - 513.1875 KHz
bit(s) 120 517.5 - 521.8125 KHz
bit(s) 127:132 547.6875 - 573.5625 KHz
bit(s) 141:142 608.0625 - 616.6875 KHz
bit(s) 153 659.8125 - 664.125 KHz
bit(s) 155:161 668.4375 - 698.625 KHz
bit(s) 171:172 737.4375 - 746.0625 KHz
bit(s) 178:179 767.625 - 776.25 KHz
bit(s) 186:190 802.125 - 823.6875 KHz
bit(s) 195:205 840.9375 - 888.375 KHz
bit(s) 208:212 897 - 918.5625 KHz
bit(s) 222:224 957.375 - 970.3125 KHz
bit(s) 229 987.5625 - 991.875 KHz
bit(s) 232:255 1000.5 - 1104 KHz

WITH TV

»/users/deblin/···aph=true

Downstream Current Rate (Reported): 5728
Downstream Maxrate (Estimated): 5728
Downstream Bit Margin (Calculated): 388

Upstream Current Rate (Reported): 768
Upstream Maxrate (Estimated): 1136
Upstream Bit Margin (Reported): 96

Bit tones with notches/missing bits:

bit(s) 38:42 163.875 - 185.4375 KHz
bit(s) 97 418.3125 - 422.625 KHz
bit(s) 109:115 470.0625 - 500.25 KHz
bit(s) 128:135 552 - 586.5 KHz
bit(s) 139:143 599.4375 - 621 KHz
bit(s) 153:165 659.8125 - 715.875 KHz
bit(s) 170:173 733.125 - 750.375 KHz
bit(s) 178:179 767.625 - 776.25 KHz
bit(s) 185:190 797.8125 - 823.6875 KHz
bit(s) 195:204 840.9375 - 884.0625 KHz
bit(s) 210:215 905.625 - 931.5 KHz
bit(s) 222:226 957.375 - 978.9375 KHz
bit(s) 231:255 996.1875 - 1104 KHz

From the looks of things you have a marginal Elite line to begin with. Those notches caused by the TV might be harmonic interference. Most of them seem to be somewhat evenly spaced. The only thing is that they are there with the TV off and powering on the TV only makes the interference stronger (wider, deeper notches).

I think your TV is emitting interference even when off. The stand by power may be causing your Elite to be marginal.]]> http://www.dslreports.com/forum/remark,23286336 Tue, 03 Nov 2009 18:21:26 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23286131 tonydi :

said by ejg1

:

I cant see how to produce the chart shown in the link you posted?

The tool is at the link I've highlighted in the FAQ.

]]> http://www.dslreports.com/forum/remark,23286131 Tue, 03 Nov 2009 17:41:22 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23285978 ejg1 : Well... I could not figure out how to find the modem ananysis tool.

This data is without the TV on. After rebooting the modem I did three speed tests at speakeasy all returning DL speeds of 5100+ . Then I saved the Technician readout seen in this link: »jiliz.fileave.com/techreadout_noTV.htm

This is with the TV on. After rebooting the modem I also did three speakeasys as above. They resulted in 3015, 3077 and 3084 respectively. Then I collected the Technician readout stats seen in this link: »jiliz.fileave.com/techreadout_after.htm

Whats next?]]> http://www.dslreports.com/forum/remark,23285978 Tue, 03 Nov 2009 17:12:25 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23285435 ejg1 : I cant see how to produce the chart shown in the link you posted?]]> http://www.dslreports.com/forum/remark,23285435 Tue, 03 Nov 2009 15:47:04 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23285267 d_l : We should be able to see the frequencies involved on the DMT chart from the 4100 or if the TV is simply spitting out pure impulse noise (I think this is unlikely). Interference frequencies can probably be filtered. It has been done before for a treadmill, but it was quite expensive. I'm not sure if pure impulse noise can be filtered the same way.]]> http://www.dslreports.com/forum/remark,23285267 Tue, 03 Nov 2009 15:16:51 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23285058 ejg1 : I found the firmware... installed and verified the modem is now at version 1.0.0.53]]> http://www.dslreports.com/forum/remark,23285058 Tue, 03 Nov 2009 14:43:58 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23284967 ejg1 : I have the 4100 online now but I see the firmware is not up to date. It is sitting at 1.0.0.48 and must be upgraded to 1.0.0.53

figuring out where to get the new firmware now...]]> http://www.dslreports.com/forum/remark,23284967 Tue, 03 Nov 2009 14:30:24 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23284836 ejg1 : I have a back-up Speedstream 4100 modem. Would this help? I could swap out if you see benifits.]]> http://www.dslreports.com/forum/remark,23284836 Tue, 03 Nov 2009 13:59:12 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23284663 d_l : Most likely your noise margin (SNR) is dropping severely and your CRC rates are very high while the TV is on, but is best to verify this guess. Unfortunately with the Netopia you can't run a DMT chart: »SBC DSL FAQ »How can I check for the maximum attainable sync speeds with a 5100b/4100 modem? to see if there might only be a specific range of frequencies (a notch) that needs to be filtered or maybe the TV is outputting a broad range of harmonic interference.]]> http://www.dslreports.com/forum/remark,23284663 Tue, 03 Nov 2009 13:24:21 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23284604 ejg1 : 1)Elite (6.0Mbps down)

2)Netopia 3346n-002

3) No additional router is used. (Router is built into the modem).

4) Modem synchs up fine (full speed synch-up with ATT PPPOE user passwords entered into it.)

5) Lights blink normally with data transfer. When there is no activity the lights stay solid green .

6) There are no items plugged into the phone jacks when testing. Just the dsl modem straight into the wall with no filter. No alarm exists for the house that uses any phone jack.

7) Power cycling the modem does not improve the situation.

8) No noises from the phone lines are detectable since nothing is plugged into any phone jacks except for the modem.

9) I will collect the before and after modem statistics and post them in my follow-up post.

Before and after TV data collecting might take me an hour. So please check back. Thanks!!!]]> http://www.dslreports.com/forum/remark,23284604 Tue, 03 Nov 2009 13:14:59 EDT Re: Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23284483 d_l : What modem model do you have? Please use the info in this FAQ to get stats from your modem with the TV both on and off and post them here so we can compare them: »Pacbell FAQ » What information do you need in order to help me with my DSL problem?.]]> http://www.dslreports.com/forum/remark,23284483 Tue, 03 Nov 2009 12:50:50 EDT Interesting and major DSL performance problem http://www.dslreports.com/forum/remark,23284439 ejg1 : I have just finished up with the DSL Reports Tweaking form and now think my issue is more appropriate here.

I have 6MB service the performance of which has degraded drastically since the purchase of a 54inch Plasma TV. To read about the whole story please look at my history in this DSLReports tweak thread: »Suffering with 60 sec page loads

To Summarize, I get full speed when the TV is turned off. When the TV is turned on the DSL line gets very noisy and my DL speeds degrade from speakeasy benchmarks of 5100 to 3600 to even 28K. The TV is located through a wall with a straight line distance of 10 paces to any computer equipment.

I have a surge protector on all the TV equipment. There is a phone outlet right behind the TV with nothing plugged into it. (Could that be acting as an antenna to my DSL line???) The TV has a power rating of 600 Watts. When doing the speed tests there is no phone plugged into any of the house outlets. No alarm system exists.

The problem is, one person can not use the Internet comfortably if another person wants to watch TV. I am not sure what to do at this stage to fix this.. Can somebody assist?]]> http://www.dslreports.com/forum/remark,23284439 Tue, 03 Nov 2009 12:43:07 EDT