Belkin on a long and noisy line...

Hi All!

I am having a little bit of an issue wiht my line. Seems that the line is far from exchange and possibly very noisy. I have been fighting with my ISP over this (took them 1 month to sort the line and for me to get service), but I would like to ask the knowledgeable public if I could do anything to assist the situation.

I have a Belikin ADSL Modem/Router (F5D7632-4). I am still on the older firmware version and am about to upgrade it. I am currently (ever since I moved house) experiencing a lot of sync drops and reconnections. The synch speed downstream at 1720K (ADSL Max) seems to be quite stable and is with,
LocalSNRMargin = 20.0 dB
LocalLineAttn = 38.5 dB
Reported by the equipment.

From time to time, the line will loose synch and the modem will try to reconnect - the connection speeds are around 3000K (and because of the decreased SNR Margin -- down to 4dB, it drops after about 2, 3 seconds). I am very far from the Exchange and it does seem that the line is noisy.

My questions:
1) What can I do to stop the drops from happening?
2) Has anyone upgraded firmware on this type of Belkin modem/router? Would you recommend it?

Thank you!
Vlad

38.5db attenuation doesn't fit into a 'long' line really.

Not too sure what you mean, but I am at least a mile and half from my exchange. Would you please tell me if the problem can be somewhere else?

Vlad

It is below the attenuation limit for a fixed speed 2Mb line of 43db and with those figures I would expect a higher sync figure than you are getting. Though with your report of dropping to such a low SNR either there is something at your premises causing the figure to drop or there is something between you and the exchange causing the SNR drop. Have you tried connecting to the test socket behind the master socket lower faceplate to see if you get better figures. If you do it is possible to improve your internal wiring to eliminate the cause of at least some of this SNR drop.

Hi! I have taken the face plate and have tried the connection directly. The configuration of the router is as follows:
main BT socket, filter, modem, telephone. The telephone lives just next to the modem - a wireless dect base. both are on a power extension which is then shared with the freezer.

What puzzels me most is that I can have a week or so of stable line and then I get a lot of instability.

The problem is that now even with 1728 (which was a very stable rate) I am getting disconnects... Line is attenuated.

I do not know what to do. Looks like an intermittent problem, but BT is saying there is no problem on the line, ISP is saying there is no problem on the line. I cannot think of anything that is happeinign at home in a week which is not happening in another week... I simply cannot find the cause...

In reply to:

---

a wireless dect base. both are on a power extension which is then shared with the freezer.

---

Is this intended to act as a worst case RFI testbed, or did you happen upon it by accident ?

Try it with the phone unplugged, the freezer off and the router on its own in the test socket behind the master. How do the stats compare with normal.

Hi I realise how that sounds, and I did mention the configuration on purpose.

Since I have changed the config, made the tests even without the face plate, alas there is no change.

Here is the reported informatoin from my Router.

DriverVersion = 1.61
APIVersion = GS_API_491
FirmwareVersion = Y.1.28.186
DspVersion = 0x000000c1
Connected = true
OperationalMode = G.Dmt
State = Showtime
Watchdog = 0x00000053
OperationProgress = 0x000000ad
LastFailed = 0x00000000
TxBitRate = 448000
RxBitRate = 3232000
TxCellRate = 1056
RxCellRate = 7622
PhyTXCellCount = 329
PhyRXCellCount = 4152426516
PhyCellDropCount = 2048
OverallFailure = 0
LocalITUCountryCode = 0x0000b5ff
LocalSEF = 267
LocalEndLOS = 1003
LocalSNRMargin = 15.5 dB
LocalLineAttn = 40.0 dB
RawAttn = 40.0 dB
LocalTxPower = 10.9 dB
LocalFastChannelRxRate = 0
LocalFastChannelTxRate = 0
LocalFastChannelFEC = 0
LocalFastChannelCRC = 0
LocalFastChannelHEC = 0
LocalFastChannelNCD = 0
LocalFastChannelOCD = 0
LocalInterleavedChannelRxRate = 3232000
LocalInterleavedChannelTxRate = 448000
LocalInterleavedChannelFEC = 820
LocalInterleavedChannelCRC = 397
LocalInterleavedChannelHEC = 45
LocalInterleavedChannelNCD = 0
LocalInterleavedChannelOCD = 427
RemoteSEF = 2
RemoteLOS = 3
RemoteLineAttn = 27.0 dB
RemoteSNRMargin = 20 dB
RemoteFastChannelFEC = 0
RemoteFastChannelCRC = 0
RemoteFastChannelHEC = 0
RemoteFastChannelNCD = 0
RemoteInterleavedChannelFEC = 5
RemoteInterleavedChannelCRC = 3
RemoteInterleavedChannelHEC = 2
RemoteInterleavedChannelNCD = 6
ActivateLine = None
HostControl = Enable
AutoStart = true
failsafe = false
ShowtimeLed = 2
Whip = Disable
WhipActive = Inactive
Action = Startup
Profile = MAIN
Standard = Multimode
UtopiaInterface = Level1
EcFdmMode = FDM
MaxBitsPerBin = 15
TxStartBin = 6
TxEndBin = 31
RxStartBin = 32
RxEndBin = 255
RxAutoBinAdjust = Disable
TxAttenuation = 0
BitSwap = Enable
BitSwapUp = Disable
NTRMode = Disable
CodeType = AnnexA
MaxDownRate = 4095
PhysicalPort = 0
Retrain = Enable
DetectNoise = Disable
Capability = Disable
CodingGain = auto
FramerType = Type3
FastRetrain = Enable
ClockType = Crystal
ACPilotTone = Enable
DyingGasp = Disable
Defaults = None
resetDefaults = false
portSnmpIfIndex = 0
portSnmpIfType = 0


By reading the above, I think I should not have any problem... but I am not online... it does die after a while with these settings.

Seems that the PPP session gets terminated after a while, then the device does not get reconnected.

PS I did upgarde the Firmware of the device to version 1.00.9 from 1.00.8

V

Edited by batevladi (Mon 11-Sep-06 15:01:31)

how long had it been on ?

RemoteLOS = 3 - 3 losses of signal at the exchange
LocalEndLOS = 1003 a thousand losses of signal at your end ! One for every time the freezer thermostat cuts in, perhaps

Nice one, but the stats are from the modem being connected for less than 2 minutes. And the power (and modem) away from the room with the freezer. (it is still pluged into the main face plate). Actually in my previous house the connection configuration was similar plus the modem was living on top of the fridge. The connection was faultless. Same modem different house.

So, I guess the freezer theory falls through... anyone more suggestions?

Edited by batevladi (Mon 11-Sep-06 22:34:51)