Gentlefolk,
I hope, stating the obvious, that some part of the BT Group responsible for placing sub contract orders has agreed a price with both major subcontractors against some form of task description. Amongst other things, I know that subcontractors are NOT allowed to draw any paper towels from BT stores. I mention this to illustrate that the contract is quite obviously constructed for the absolute minimum cost to the BT Group and there seems little or no concern over the practical results. At best the end user is oblivious of this situation and accepts the results as the optimum the line can support, especially as it takes a minimum (usually) of two days before DLM defines a lower operating speed. There are often varying degrees of even worse situations as are being described here and elsewhere.
In their defence, I do question how a usually untrained subcontractor can possibly be expected to react in a pro-active manner when (s)he is unable to monitor the end result comprehensively, other than using the 17070 facilities and a speed test NOT provided by the BT Group (as the uncompleted job prevents full access to the BT one). Compare that with the JDSU approach where the results are, I believe, uploaded into a BT Database. This can only produce gross anomalies between BT O installations and subcontract ones with the latter having no useful connection data recorded to help subsequent fault finding operations. (However I do suspect that ongoing modem performance data for all ccts. is recorded.)
I will add two local examples:-
1. A free (Unadopted) primary school obtained a VDSL service in mid January. Within a couple of weeks only the VDSL signal disappeared. It is now expected to be a) a faulty filter block, b) IDC block wiring fault or c) a faulty line card port. It took BT Openreach about six weeks, including the suggestion of total cable replacement, before a well-versed visiting BT O engineer spotted that the ports were not being counted properly as Huawei cabinets start counting from zero and not 1. This resulted in the engineers and their control staff confusing each other at length upon several occasions. (It also begs the question as to whether the PCPs will be rewired or spare intermediate ports properly recorded when eventually the line card(s) are replaced, hopefully testing every filter block too.)
2. A subcontract VDSL service was installed on 24 January on a longer line with a BT Wholesale estimate of 17.9 Mbps download. The initial sync speed was only 3.86 Mbps. A certain wheelbarrow, complete with the invaluable assistance from members of this society, happened by. The immediate diagnosis was a bridged tap due to star wiring in the attic, where of course the contractor had not ventured. After that was cured, the sync speed improved immediately to 4.69 Mbps. We then entered pantomime mode with India eventually resulting in a BT O visit on 1 February to discover a poor quality pair in the final u/g cable with a marginal increase of sync to 6.39 which crept up to 10.22 Mbps on 6 February, but fluctuated down to 8.70 Mbps. As we had an unlocked modem monitoring the line we observed that the Hlog graph still had a significant anomaly. A further BT Openreach visit was organised with difficulty and a crossed pair was discovered. Once that was corrected the sync speed hurtled up to 15.04 Mbps on 17 February. The sad situation though is that the line is still quite noisy so by 19 February the sync speed had dropped to 13.12 Mbps and 12.09 by 28 February. VERY SADLY the speed has returned to 6.40 Mbps by 22 March. BT’s line test facilities report no fault found and the end user is so totally disenchanted with the hideous difficulties faced to rectify the situation he has resigned himself to a quite unsatisfactory mediocre service still running today at 6.39 Mbps.
You might all wish to question the medium term maintenance prospects for current FTTC deployments.
Kind regards,
Walter
