how-to block ads
3 Technical Questions
With certification from CableLabs, manufacturers are able to produce cable modems for sale at retail so consumers no longer need to rely on their cable providers for leased cable modems.
See the official CableLabs DOCSIS website for more info:
Standard address for most modems: 192.168.100.1
DPX-1110 (3-color Com21 label on top): Modem's private (10.xxx.xxx.xxx) IP address.
Usually this can be obtained by tech support or those with access to the CMTS.
DPX-1110XB (White Com21 label on top): 192.168.100.1.
DCM-100: not available
DCM-200: not available
DCM-201: 192.168.100.1, username: admin, password: hitron
DCM-202: 192.168.100.1, username: dlink, password: dlink or with firmware 2.01; username: admin, password: hitron
Motorola or GI
All models: 192.168.100.1
DCM-105: Not Available.
DCM-215, 225, 235, 245, or 305: 192.168.100.1/moreinfo.html.
New WebStar DPX-110 (with MAC address beginning with 00407Bxxxxxx): 192.168.100.1
Only available when the modem is not "sync'd" due to changes made by Scientific Atlanta.
Old WebStar DPX-110 (with MAC address beginning with 009096xxxxxx): 192.168.100.1
SpeedStream 6101: 192.168.100.1, username: root, password: root
PCX-1100: Not Available
PCX-2200 or 2500: 192.168.100.1
TJ-110 or 210: Not Available.
TJ-615 (Firmware version 4 or higher): 192.168.100.1
Some pages request a password, leave it blank and continue.
TJ-715 and 715x: 192.168.100.1/diagnostics_page.html
It needs password, so search Google for it.
Sometimes it's necessary to power cycle the modem to access the diagnostic page after it has been online awhile.
Prestige 944: 192.168.1.1.
For modems without internal diagnostic pages, try the utility: DOCSDIAG. This has been rendered useless in most Adelphia systems due to increased security measures Adelphia has enacted in an attempt to prevent modem hacking.
Note: Many home routers will interfere with access to these internal webpage addresses unless the router's routing tables have been specifically configured for it. The easiest way to prevent this problem is directly connect the modem to the computer when you need to check the diagnostic page.
See this about signal levels listed on the diagnostic screens: link
This is the amount of signal received by the modem from the transmitter in the cable company head-end.
For all modems:
-15 dBmV to +15 dBmV maximum.
-12 dBmV to +12 dBmV recommended.
0 dBmV is the "optimal" level.
Upstream (Tx) Transmit Power (a.k.a. Return Signal) level:
This is the amount of signal transmitted by the modem to reach the receiver in the cable company head-end.
+8 dBmV to +58 dBmV maximum for QPSK. (DOCSIS 1.x)
+8 dBmV to +55 dBmV maximum for 8 QAM and 16 QAM. (DOCSIS 1.x)
+8 dBmV to +54 dBmV maximum for 32 QAM and 64 QAM. (A-TDMA DOCSIS 2.0)
+8 dBmV to +53 dBmV maximum for S-CDMA DOCSIS 2.0 modulation rates.
Recommended upstream signal levels are +35 dBmV to +52 dBmV.
A cable modem running a higher upstream modulation rate may downgrade itself to a lower modulation rate (i.e. 64 QAM to 16 QAM or 16 QAM to QPSK) if the upstream transmit level is higher than the maximum signal level allowed for the higher modulation rate and the CMTS is configured to allow such a change. This downgrade can cause slow speed, packet loss, and connection loss issues depending on the condition of the upstream channel.
A house or drop amplifier will NOT fix upstream signal problem because most house amplifiers don't amplify the upstream signals, they only pass the upstream signal through with some loss.
SNR (signal to noise ratio) levels:
This is how clear the signal is at either the modem receiver (downstream SNR) or the receiver in the cable company head-end (upstream SNR).
DOCSIS specifications list minimum CNR (carrier to noise ratio) levels not SNR levels. The SNR levels listed here are based on commonly recommended MER levels for digital cable signals. Not all QAM demodulator chipsets accurately calculate SNR levels that approximate actual MER levels, so these levels may vary depending on which chipset and/or firmware is used in the equipment.
QPSK: 12 dB minimum. 15 dB or higher recommended. (often used in upstream channels)
16 QAM: 18 dB minimum. 21 dB or higher recommended. (often used in upstream channels)
64 QAM: 24 dB minimum. 27 dB or higher recommended. (often used in downstream channels)
256 QAM: 30 dB minimum. 33 dB or higher recommended. (often used in downstream channels)
There is no upper SNR limit, although in practice 40 dB is about the highest seen.
Downstream SNR levels are read at the modem on the downstream data channel and can be viewed using the modem diagnostic screens.
Upstream SNR levels are read at the CMTS on the upstream data channel, not the modem or the modem diagnostic screens. The end-user cannot get the upstream SNR directly. Only the provider can read the upstream SNR level, directly from the CMTS. Also, the upstream SNR level provided by most CMTSs is not specific to any single modem, but is an averaged, aggregate level from all modems on that upstream channel on the upstream port.
Important notes concerning signal levels:
1. Signal levels not within the specifications listed above can cause slow speeds, connection problems, and connection loss due to packet errors, packet loss, and/or constant packet retransmission.
2. Its recommended to have the modem's signal levels at least 3dB away from the maximum/minimum levels listed above due to normal temperature related signal variation. If the modem's signal levels are at the maximum or minimum limits, they may be out of spec if the temperature changes significantly. Signal levels that vary more then 3 dB in a 24-hour period usually indicate a problem that should be looked into.
3. Most cable systems are designed is such a way that the optimal signal levels for the modem are achieved if the modem is connected to a dedicated "homerun" RG-6 grade cable line straight to a 2-way splitter at the demarcation or ground point. This is then connected to the "drop" line connected to the cable tap on the utility pole or at the street. Like this:
4. Excess splits, bad connectors, and/or poor quality cabling will certainly effect cable signal levels and will cause problems. Make the cable run to the modem as clean as possible, the less junk in the way the better.
5. Poor signal levels should not affect latency, so if your "ping" is 300ms with a downstream level of -14.0 dBmV, it is not going to get any better if your signal is increased to +1.0 dBmV. Latency is caused by a combination of physical network distance and network equipment "packet handling" time, not signal levels.
If your wondering how to get these levels from your modem, see this page: /faq/5861.
CableLabs DOCSIS Specifications
Spectrum Analyzer CNR Versus CMTS SNR. Ron Hranac. Communications Technology, Sept. 2003
More on CMTS SNR. Ron Hranac. Communications Technology, Oct. 2003
How to Increase Return Path Availability and Throughput. Cisco Systems.
Sunrise Telecom Online Learning Seminars
Feedback received on this FAQ entry:
The modem scans for a digital QAM channel (which can be digital TV or modem downstream channels) and "listens" for upstream channel descriptor (UCD) information on it (which is only found on modem downstream channels). If it doesn't get the UCD after a few seconds of receiving data on that channel, a T1 timeout occurs and the modem looks for another digital QAM channel.
Once the modem finds the correct downstream channel and receives the upstream channel descriptor, it then waits for its initial opportunity to "talk" to the CMTS. If it doesn't get that chance within 12 seconds, a T2 timeout occurs.
So during initial modem connections, T1 and T2 timeouts are perfectly normal to see in the modem logs.
After locating a downstream data carrier channel, the modem gets its chance to "talk" to the CMTS and sends data to negotiate the proper upstream signal levels and timing. This negotiation process takes several data exchanges between the cable modem and the CMTS. If at any point the modem doesn't receive a response from the CMTS within about 200 milliseconds, a T3 timeout occurs. Several T3s have to occur before a modem resets.
The modem then goes through the DHCP, TFTP, TOD processes and once all are completed the modem is online.
Once the modem is online, the CMTS sends "Periodic Ranging" or keep-alive requests to it every 30 seconds. If the modem doesn't receive one, a T4 timeout occurs. Once 16 of the T4 timeouts have to occurred, a modem resets.
If there is a downstream communication problem, the cable modem might not receive these "keep-alive requests". If the upstream utilization is too high, or too many modems are connected to the same upstream port, it is possible that some modems will not get the required bandwidth or transmit opportunities to respond to the keep-alive requests. Either of these issues can result in T3 or T4 timeouts occurring.
See these web pages for more info:
Cisco Document ID: 16510 - Troubleshooting uBR Cable Modems Not Coming Online
Cisco Document ID: 22543 - Cable Modems Dropping Offline in a 2-way Cable Network
•Internet Gateway from ViacomSoft (PC and Mac compatible)
•MidPoint from MidCore
»www.midcore.com/ (currently offline, but possibly available elsewhere)
Both are software router/firewall/NAT solutions that allow connection teaming, be it 2 or more cable modems, cable/dsl, cable/dial-up, or any other combo. They both allow parallel downloading and/or fail-back connections (where one connection takes over if the other fails), in addition to all the other features offered.
There are also a few different hardware options out there too, that do a similar thing for a high price.
•Nexland PRO 800 Turbo Internet Sharing Box
•Symantec Symantec Firewall/VPN Appliance (Models 200 or 200R)
This is by no means a complete list, its just to give an idea of what I have found in a relatively short time.
Here is a white paper that explains how this works:
Vicomsoft White Paper: HTML or PDF.
"What is not generally well-known is that the upstream cap can also affect the downstream speed -- if the upstream is saturated by uploading (e.g., sending a large PowerPoint file to the boss, or running a Napster or other public service), the downstream will drop to about the same speed. This is due to a weakness in the basic TCP Internet protocol, not Cable or DSL per se, and not the service provider.
Quoted from »cable-dsl.home.att.net/#Asymmetry
From a question in the forum by pianotech and a link provided by hobgoblin
For information on how to disable file sharing in many common P2P applications, see Adelphia's eSafety page on disabling P2P file-sharing:
ALL Internet connections share bandwidth at some point. DSL, dial-up, and most data services offered by the phone companies start to share bandwidth at the phone company central office. The Internet services offered by cable companies to most residential users start to share bandwidth earlier in the connection then that.
A node is a piece of equipment that connects the cable system fiber optic network of an area to the coax cable network distributed throughout the local neighborhood. Nodes are really just media converters, connected to similar equipment in the cable head-end, used to bring the signal from the cable head-end to a neighborhood without the long series of amplifiers previously needed to do so.
Nodes are important because due to the way cable systems are built, nodes are the common aggregation point for all the cable signal traffic in a particular neighborhood. This means all users connected to one node share the bandwidth available to that node.
The number of users per node depends on subscription rate, number of homes passed per node, and general system architecture. Homes passed per node can be anywhere from 100-2000 and subscription rate depends on competition, marketing, and system performance. Most cable companies design for 10-25% cable modem subscription rates for an area. This means anywhere from 10-500 active modem customers are connected per node in most cases, depending on local cable system design.
Bandwidth limits in cable modem networks are not per node. Bandwidth limits in cable systems are defined by the upstream or downstream channels assigned. Nodes aren't the limit because they can be configured to use larger data channels or multiple data channels for more bandwidth.
Often due to low bandwidth utilization, several nodes are combined together in the cable "headend" before they are connected to the ports on the CMTS, so another way a provider can increase available bandwidth to an area is to "decombine" a number of nodes using the same data channel.
The bandwidth available per CMTS data channel in DOCSIS 1.0, 1.1, and 2.0 CMTSs is about 30 mbps per downstream channel using QAM 64 modulation and about 42 mbps using QAM 256 modulation. Total upstream bandwidth is limited to about 5 mbps per upstream channel using QPSK modulation and 10 mbps using QAM 16 modulation for DOCSIS 1.0, 1.1, and 2.0 CMTSs Up to 30 mbps total upstream channel bandwidth can be made available on DOCSIS 2.0 CMTSs using DOCSIS 2.0 modulation rates. Actual usable data bandwidth is 10-15% lower due to the bandwidth overhead of DOCSIS protocols.
Every cable provider using DOCSIS cable modem equipment has these same bandwidth limitations.
This works because everybody connected in a neighborhood isn't using their allowed allotment of bandwidth every second of every day and cable modem systems are designed to take advantage of that. This is also why every single cable ISP prohibits customers running servers on computers connected to the cable modem because constant use by a single modem reduces overall capacity available for everyone else on those CMTS ports. If more than a few servers are active very little bandwidth would be left for possibly hundreds of other users.
Obviously more customers per node and CMTS port is more cost effective for the cable company, but pushing it too far can cause bandwidth capacity overload.
Try this article, its older but gives a good feel for the situation:
Here's a few good sites for an overall explanation:
For a technical description of what a modem does to get online:
Here's an ACLU comissioned report on open access which details how cable systems work in the appendix:
This was taken directly from a post (and edited) by MacLeech in the Adelphia forum here.
Adelphia CMTSs are frequently assigned multiple, non-contiguous blocks of IP addresses, so an IP address and/or IP subnet change isn't anything unusual after an outage or upgrade.
IP changes can happen during outages that are longer than your IP lease time or when a large number of people go offline, but it doesn't specifically mean any network upgrades or changes were done. Such address changes are totally dependent on how many modems are offline and length of the outage vs. size of the assigned subnets and the IP address lease time left. The larger and/or longer the outage, the higher the chances that your subnet and IP address will change.
Adelphia's change management window is during the "silent" off-peak hours. Normally 1:00am to 5:00am. An outage that starts at 10:00pm is an unscheduled one.
Question and answers originally seen in this thread, thanks to ILOVELA2 and Hobgoblin.
Adelphia does not use PPPoE. While commonly found in DSL networks, PPPoE is almost never found in cable modem networks.
Adelphia does not use static IP addressing, unless you are specifically paying and are configured for Static IP service. You should never have to enter your IP address, DHCP client ID, subnet mask, default gateway, or DNS server addresses. These are all configured automatically when subscribers' computers are properly setup using TCP/IP and DHCP.
Adelphia does not use proxy servers either.
Blaster/LoveSan worms, Welchia/Nachi worms, and their variants by dropping, filtering, and/or rate limiting certain outbound ICMP packets. These measures continue to this day and the effects of them do change when equipment or configurations are changed.
The Adelphia imposed ICMP packet restrictions cause inaccurate packet loss reports when using the standard Windows Tracert program, the Windows Pathping program, and the Line Quality test here at Broadband Reports. The standard Windows, Mac OS, and Linux Ping programs are unaffected, so if just using the standard Ping program results in no packet loss, there isn't really a problem. Linux and Mac OS X users using Traceroute also shouldn't be affected by the ICMP packet limitations as those trace route programs use outbound UDP packets by default.
•Here's an example of rate limiting as seen when using Windows tracert utility:Whether the ICMP packets are dropped, filtered, or rate limited depends on the severity of the effect those worms/viruses had on that part of the network, what particular equipment is currently installed in that part of the network, and how it is configured. Not all network equipment can be configured to deal with the offending packets the same way, so different techniques had to be used. In some areas, you may find combinations of techniques were used, such as ICMP packets being filtered at the CMTS, only to be totally dropped at a router further down the line.
To see a more accurate trace result which isn't affected by these Adelphia imposed ICMP packet limitations, use a different program that can trace using UDP packets.
Here's a short list of recommended trace route programs for Windows users:
•PingPlotter has a shareware version (not the freeware version) that can use ICMP, UDP, or TCP packets in traces. PingPlotter graphically shows latency, packet loss, and route history, so its excellent for beginners and advanced users alike. Make sure you turn on UDP packets in the Advanced Options menu, see this page for more setup instructions: PingPlotter Packet Options. It has a full-feature enabled 30-day trial period, so you can try it out long enough to run some tests before you have to pay.
•Ftrace is a program with output very similar to the standard Windows trace and ping utilities but it can utilize UDP packets. There is also a GUI enabled version that you can launch from your Windows desktop which also includes some very handy "copy" buttons for posting those traces here at BBR. To use UDP pings make sure you use the -u option (i.e. ftrace -u yahoo.com)
•VisualRoute is another Windows program that can do traces using UDP packets. Just don't believe the maps it generates as they are often very inaccurate. Read the manual for instructions on how to turn on UDP packets.
These programs don't use UDP packets by default, make sure you enable it.
Read this article for a bit of insight on the effects of the worms mentioned above: Internetnews: 'Friendly' Welchia Worm Wreaking Havoc or read this BBR thread for even more insight: Nachi the new champion bad boy. See this Adelphia page for more information on current virus activity: Adelphia E-safety page.
Global Crossing: »ipstats.globalcrossing.net/dotco···pe=trace
A bigger list of trace route sites can be found here:
The following is a list of modems that forum users have found specific issues with. The general issue is listed along with the thread in which it was discussed.
RCA 235 and other RCA models with the molex power plug:
Due to issues with the molex power-plug, it loosens and causes connectivity problems.
Due to a problem with the v1.06 firmware, connections using alot of bandwidth will lock-up the modem, forcing a reset to regain connection.
Motorola Surfboard 4100 and 4101:
Issues with firmware versions 4.0.11 and 4.0.12 cause slow upload speeds.
/forum/remark,6810235~mode=flat I'm searching for other links to this as well.
Motorola Surfboard 4200:
With certain firmware versions (v4.4.0 and possible a few earlier versions) this modem will lock up when passing certain types of traffic.
Motorola Surfboard 5100:
A firmware issue in version 220.127.116.11 can cause slow speeds when connected to certain CMTSs with particular configurations. Upgrading the firmware fixes this.
Some versions of the following modems seem to have issues with 256 QAM downstream channels. Slow speeds and packet loss are commonly seen symptoms:
Com21 1110 (3-color label on top)
Siemens SpeedStream 6101
Scientific Atlanta DPX-110
Linksys BEFUCM10 version 1 & 2:
This list is neither official, comprehensive, or complete.
Enter the proper information for your connection type (i.e. cable service, 4000 or 6000 for speed in kbps, operating system, and "normal" connection type) Configure your system as recommended in the results of the test.
For more "tweak" help visit the Tweak FAQ here:
If you want some personalized assistance, post in the Tweaks Forum with these answers.
Search Adelphia's results here:
Just enter your full zip code to see your specific area or just part of it to see results from larger area around you.
Mozilla and FireFox are apparently not supported at this time.
With whatever browser you are using, if you don't see "user has entered room", it isn't going to work.
Thanks to Johkal in this thread: /forum/remark,13169475
If you believe you are having problems related to DNS server issues, try using one of these DNS servers:
18.104.22.168 = resolver1.level3.net
22.214.171.124 = resolver2.level3.net
The above servers are popular because they are probably the fastest and closest non-Adelphia DNS servers on the internet. These servers are multi-homed and the IPs listed above will take you straight to the server at the nearest Level3 peering point, which is probably also the nearest Adelphia peering point due to network arrangements between the two companies: »www.level3.com/press/5357.html
126.96.36.199 = vnsc-pri.sys.gtei.net
188.8.131.52 = vnsc-bak.sys.gtei.net
The above servers are popular because the IP addresses are very easy to remember.
For Window's systems, the DNS server IP addresses can be changed the following way:
Go to Control Panel>Network Connections and select your local network.
Click Properties, then select Internet Protocol (TCP/IP).
You will see a window like the one below - this is the Internet Protocol window. Select "Use the following DNS server addresses" and enter the desired DNS server(s) in the space(s) provided.
If you have a router, changing the DNS servers listed in the router configuration pages will change it for all computers connected to the router.
Reselecting "Obtain DNS servers automatically" will let your computer use the DHCP configured DNS servers again.
Feedback received on this FAQ entry:
Servers of ANY type (including personal FTP, game hosting, email, and open WiFi networks) are prohibited on Adelphia's HSI service.
See the Adelphia HSI Service Agreements for complete details:
or see Adelphia's "common violations" page:
•Change the NIC card in the computer.
•Use the USB connection of the modem if you normally use the ethernet connection or ethernet if you normally use USB.
•If you use a router, using the MAC "cloning" feature to change the MAC address to another one.
•Change the MAC address of the NIC card. A few NIC cards allow this, see the manufactures' website for more details.
Unplugging the modem and/or computer for any length of time is NOT a guaranteed way to change your IP address.
• Phone support. Call (888) 683-1000, 24 hours a day, 7 days a week.
• On-line Chat Support. Click here.
Same levels of support as offered through the phone.
• Local office support. Check your bill for the closest location.
Most offices will swap cable boxes or modems over-the-counter if requested.
The vast majority of local offices are only available for walk-in service.
No customer phone support is usually offered.
• On-site Tech Support. Scheduled using one of the three methods listed above.
99% of the time they are cable techs, not computer techs.
Field techs are not authorized to repair your computer, remove viruses, or troubleshoot home networking setups.
The initial Technician sent will normally check and repair the physical cable installation and equipment from the cable distribution tap to the modem or cable box in the home. Problems they find at the cable distribution tap will be referred to the Maintenance department techs for further troubleshooting.
Maintenance Technicians are responsible for insuring proper operation of the cable distribution system in the local neighborhood. They normally will not need access into the home.