Forum Discussion

Nednats's avatar
Nednats
Up to speed
13 days ago

Ongoing DOCSIS 3.1 instability – correlation between downstream power rise, error rates, and latency

I’m looking for a second opinion from those familiar with DOCSIS 3.1 behaviour, as I’m seeing a repeatable pattern that doesn’t look like a simple in-home issue.

Context

  • Virgin Media HFC (Hub 5, DOCSIS 3.1)
  • Long-running intermittent latency and packet loss
  • History of T2/T3 timeouts, upstream profile changes, and partial service events
  • Multiple engineer visits; most recent visit (26/11/2025) resulted in an equaliser being fitted due to limited downstream headroom

What I’m seeing
Over the last 12 hours (03/01/2026), the following appears clearly correlated:

  • As downstream power levels rise,
  • Pre-RS error rates increase significantly, and
  • Latency and packet loss increase on the ThinkBroadband BQM

This is visible when comparing:

Grafana modem stats (downstream power + Pre-RS errors)

 BQM (latency, jitter, packet loss)

The rise in errors and latency tracks the downstream power increase very closely. When downstream power stabilises or drops slightly, error rates and latency reduce accordingly.

Additional observations

  • Downstream SNR remains broadly stable
  • Post-RS error rates remain very low (errors are being corrected, but at scale)
  • Upstream power is already high and leaves little headroom
  • No corresponding increase in local traffic or load

Behaviour is time-dependent and not tied to in-home activity

Why I’m concerned
This doesn’t look like:

  • Home network congestion 
  • Local wiring or CPE fault
  • Random noise bursts

Instead, it looks more like:

  • A plant-level or cabinet-level issue
  • Possibly thermal drift, attenuation imbalance, or noise ingress affecting DOCSIS margins
  • A condition that won’t reliably present during short engineer visits unless the timing is right

Question
Am I right in thinking that:

  • Rising downstream power pushing channels closer to the top of acceptable range could be reducing error margin, leading to increased FEC activity and latency?
  • This points more towards a network-side balancing or plant issue rather than anything further an in-home visit can resolve?

Grateful for any technical insight or confirmation from others who’ve seen similar DOCSIS 3.1 behaviour.

13 Replies

  • Client62's avatar
    Client62
    Alessandro Volta
    11 days ago

    A power level change of approx +0.2 dBmV over a 9 hour period is quite negligible.

    • Nednats's avatar
      Nednats
      Up to speed
      9 days ago

      However, power level have still gone out of spec.

      The trend continues and power levels slowly rise. Now currently hitting 9.2 dBmV

       

  • Client62's avatar
    Client62
    Alessandro Volta
    9 days ago

    Share the actual Downstream and Upstream signal levels & errors et al from the VM Hub.

    Copy / Paste the text as screen shots often result in images that are unreadable.

  • Nednats's avatar
    Nednats
    Up to speed
    9 days ago

    Channel    Frequency (Hz)    Power (dBmV)    SNR (dB)    Modulation    Channel ID
    1    330000000    8.7    41    QAM 256    25
    2    154000000    8.9    41    QAM 256    3
    3    162000000    8.9    41    QAM 256    4
    4    170000000    8.8    41    QAM 256    5
    5    178000000    8.9    41    QAM 256    6
    6    186000000    8.9    41    QAM 256    7
    7    194000000    9.1    41    QAM 256    8
    8    202000000    9.1    41    QAM 256    9
    9    210000000    9.2    41    QAM 256    10
    10    218000000    9.1    41    QAM 256    11
    11    226000000    9.1    41    QAM 256    12
    12    234000000    9.1    41    QAM 256    13
    13    242000000    9    41    QAM 256    14
    14    250000000    8.8    41    QAM 256    15
    15    258000000    8.9    41    QAM 256    16
    16    266000000    8.9    41    QAM 256    17
    17    274000000    9    41    QAM 256    18
    18    282000000    9.1    41    QAM 256    19
    19    290000000    9    41    QAM 256    20
    20    298000000    9.1    41    QAM 256    21
    21    306000000    9    41    QAM 256    22
    22    314000000    8.8    41    QAM 256    23
    23    322000000    8.9    41    QAM 256    24
    24    338000000    8.8    41    QAM 256    26
    25    346000000    8.9    41    QAM 256    27
    26    354000000    9.1    41    QAM 256    28
    27    362000000    8.8    41    QAM 256    29
    28    370000000    8.8    41    QAM 256    30
    29    378000000    8.6    41    QAM 256    31
    30    386000000    8.3    41    QAM 256    32
    31    394000000    8.5    41    QAM 256    33
    3.0 Downstream channels
    Channel    Locked Status    RxMER (dB)    Pre RS Errors    Post RS Errors
    1    Locked    41    61    0
    2    Locked    41    233    0
    3    Locked    41    167    0
    4    Locked    41    99    0
    5    Locked    41    66    0
    6    Locked    41    107    0
    7    Locked    41    84    0
    8    Locked    41    109    0
    9    Locked    41    108    0
    10    Locked    41    131    0
    11    Locked    41    72    0
    12    Locked    41    62    0
    13    Locked    41    71    0
    14    Locked    41    86    0
    15    Locked    41    82    0
    16    Locked    41    101    0
    17    Locked    41    103    0
    18    Locked    41    168    0
    19    Locked    41    227    0
    20    Locked    41    205    0
    21    Locked    41    82    0
    22    Locked    41    72    0
    23    Locked    41    89    0
    24    Locked    41    49    0
    25    Locked    41    36    0
    26    Locked    41    35    0
    27    Locked    41    46    0
    28    Locked    41    54    0
    29    Locked    41    30    0
    30    Locked    41    1420    0
    31    Locked    41    1076    0

  • gitty's avatar
    gitty
    On our wavelength
    9 days ago

    Are you tracking the upstream? Power, modulation, T3, etc.

    You have some packet loss throughout the day, but at first sight, it doesn't look to be downstream related.

    • Nednats's avatar
      Nednats
      Up to speed
      7 days ago

      Yes. Here is the last 3 hours: 

       

      On 27 Dec BQM show minimal ponacket loss:

      on 28 Dec this happen:

      Another user posted a similar BQM and posted a social media link suggesting the more users expereincing similar issues at the same time: https://community.virginmedia.com/discussions/Speed/connection-issues-north-east--west-according-to-xtwitter/5689369

      Since then there is more packet loss daily - on the 2 Jan: 

      Then on 3 Jan this happens: 

      an since the 4th my BQM has looked liked this:

      On 3 Jan downstream power level became out of spec and have stayed out of spec. 

      Upstream is in spec but there is no alot of headroom

      • gitty's avatar
        gitty
        On our wavelength
        7 days ago

        How do you define out of spec? VM has 10 dBmV/8 MHz as an operational threshold, which is below specification (wiggle room). It is not the breaking point. I don't see any evidence that the modem's receiver is being overloaded. I think you can put the receive power aside for now in your quest.

  • Tudor's avatar
    Tudor
    Very Insightful Person
    9 days ago

    We need to see the upstream stats, they are just as important as the downstream when diagnosing problems. 

    • Client62's avatar
      Client62
      Alessandro Volta
      7 days ago

      ... and all the stats for the 3.1 DOCSIS channels too.