MCDA / CONVEYANCE / INTERCEPTOR 7 / REGISTERS

Interceptor 7 — Deferred Terminology Register
and Schedule of Held Faults

Third revision. Unadopted. Circulated for comment and not withdrawn.

DOCUMENTI7-REG-03
SUPERSEDESI7-REG-02 (partially; see §4)
STATUSin service, unadopted
CUSTODIANDuty Engineer, Interceptor 7
LAST FIELD CHECK1994 infiltration survey (partial)
SECTION 1 — SCOPE

What this register is for

Interceptor 7 has been in continuous service since 1911 and has never been fully surveyed. Sections of it are known only through the documents that describe them, and several of those documents disagree. This register does not resolve the disagreements. It lists them, gives each one a number, and states the condition under which it could be resolved.

The Authority's general practice is to correct errors when they are found. Interceptor 7 is exempt from that practice by a standing directive of 1972, the text of which is reproduced at §5. The short version is that on a system this old, a correction made without a field check is indistinguishable from an error introduced, and the field check is usually not possible. So the wrong figure stays, and the fact that it is wrong stays next to it.

Staff transferring in find this uncomfortable. That is normal and it passes.

SECTION 2 — DEFERRED NOMENCLATURE

Terms retained in error

Three terms are currently deferred. Deferred means: the term is known to be inaccurate, a more accurate term exists or could be constructed, and the inaccurate term remains in service because replacing it would break something else. Each entry states what would have to happen for the term to be superseded.

Scour Gate
#term-scour-gate · deferred since 1968 · supersession: two field attestations
APPLIED TOassembly at Chamber 7-C
FUNCTION PERFORMEDnone; the assembly is welded
SCOURING PERFORMEDnone since 1968 reconstruction
CORRECT TERMknown to Maintenance, not deployed
WHY RETAINEDappears on 41 drawings and 3 valve tags

Maintenance holds the correct term and will not release it until two people who have been physically inside Chamber 7-C since 2010 say the same thing about the assembly independently. At present there is one such person. There has been one such person since 2010. The register notes this without comment.

Dry Weather Flow
#term-dry-weather · locally valid · supersession: full regauge
DEFINED ASbaseline flow absent precipitation
ACTUAL CONTENTbaseline plus groundwater infiltration
INFILTRATION FRACTIONestimated 0.3–0.6, unmeasured since 1994
WHY RETAINEDevery legacy gauge is calibrated against it

This is the most consequential deferral in the register and the one nobody argues about, because the alternative is regauging eleven interceptors against a figure that would itself need revising in a decade. The term is wrong. The term is also the only thing holding thirty years of flow data in a single comparable series. Both of those are true at once and the Authority has chosen the series.

Summit Run
#term-summit-run · deferred · supersession: none specified
APPLIED TOreach between Chambers 4 and 9
SUMMIT PRESENTno
GRADIENTcontinuous fall, 1:840
ORIGIN OF TERMunrecorded

Nobody knows where this one came from. It is not in the 1911 drawings, it is not in the 1968 reconstruction file, and it is in use by 1974. The most likely explanation is a transcription error from an adjacent scheme, but no adjacent scheme has a summit either. It is possible that someone simply liked it.

The register lists it because an unexplained term that everyone uses correctly is a different kind of object from a term that is wrong. Nobody is misled by Summit Run. There is nothing to be misled about. It is a name with no referent to be wrong about, and it works.

SECTION 3 — HELD FAULTS

Discrepancies retained without resolution

FAULT11 — Weir Crest Discrepancy
CLAIM Acrest at 12.40 m (1968 as-built)
CLAIM Bcrest at 11.85 m (all surveys, 1981–)
REFERENCE FRAME Aconstruction datum, superseded 1979
REFERENCE FRAME Bnational datum
DIFFERENCE0.55 m; datum shift accounts for 0.41 m
UNACCOUNTED0.14 m
RESOLUTIONnot authorised
INFO LOST BY RESOLUTIONunknown

The datum shift explains most of it. It does not explain all of it. The 0.14 m is the actual entry — the rest is arithmetic. Two readings of a physical object, taken thirteen years apart, differ by fourteen centimetres after every known correction has been applied, and neither survey team is available to ask.

Three explanations are live: settlement of the chamber, an unrecorded works, or one of the surveys being wrong. The register does not choose. Overflow modelling for Interceptor 7 currently runs both crest figures and reports the envelope, which is annoying, which is the point. If a single figure were adopted, the modelling would get tidier and the fourteen centimetres would disappear from every downstream document within two revision cycles. It has happened before on other assets. It is why this register exists.

FAULT14 — Chamber count
CLAIM Aeleven chambers (asset register)
CLAIM Btwelve chambers (1968 reconstruction file)
STATUSopen since 2003
RESOLUTIONnot authorised

The twelfth chamber, if it exists, is between 8 and 9 and has no surface access. The 1968 file references it twice, once in a schedule and once in a marginal note. No drawing shows it. A 2003 CCTV run was aborted at 61 m for reasons the operator recorded as "silt", and 61 m is short of where the chamber would be. The run has not been repeated. There is no operational need to repeat it, which is why it has not been repeated, which is why the fault is fourteen years old, which is a sentence that appears in some form in most of the Authority's long-standing registers.

SECTION 4 — PARTIAL SUPERSESSION

What this revision replaced, and what it did not

Revision 03 supersedes Revision 02 in respect of Sections 1, 2 and 3 only. Section 5 of Revision 02 was not reproduced here, for a reason that was recorded at the time and is not now legible on the file copy. Section 5 of the present document is therefore new text occupying an old number, which is bad practice, and is flagged as bad practice in the header of the file copy, and was done anyway because the alternative was renumbering a register that is cross-referenced from the Authority's flood consent.

Revision 02 remains in circulation. Both revisions are current. The Authority is aware.

INTERNAL — CIRCULATED 2019, NO REPLY RECORDED

Further to my note of March, can somebody confirm whether the comments box on the register intranet page goes anywhere. I have submitted four comments over two years and had no acknowledgement, and the page still says comments are reviewed monthly. If it is not monitored I would rather it was taken down than left up. If it is monitored I would like to know by whom.

I appreciate this is not a priority. It has been not a priority for a while now.

The comments box remains on the intranet page. It still says monthly.

SECTION 5 — STANDING DIRECTIVES

The 1972 directive and what has been added to it

No figure recorded against Interceptor 7 shall be amended other than following a physical inspection of the asset to which it refers. Where inspection is not practicable, the disputed figures shall both be retained and the dispute recorded. Recording the dispute discharges the obligation.

That is the whole of the 1972 text. Four further directives have accumulated since, none of them formally adopted, all of them observed:

D2 (c. 1981) A figure may not be promoted to authoritative on the strength of one source, however good the source. Two independent attestations, or it stays disputed.

D3 (c. 1990) Where a term is retained in error, the error shall be stated adjacent to the term, in the same document, not in a separate schedule. Separate schedules get lost.

D4 (2004) A cost stated later is a cost concealed. Where a decision has a consequence, the consequence goes in the same paragraph as the decision.

D5 (2011) Superseded text is not deleted. It is marked superseded and left in place, with a pointer to what replaced it. The file grows. This is acceptable.

D5 is why the register is 190 pages and why the useful part is the first forty. Various people have proposed a summary version for new starters. One was produced in 2013. It is not in circulation, because it was accurate, and because after eighteen months it was discovered that three teams had begun citing the summary in place of the register, and two of the held faults had quietly vanished from their working figures. Not falsified. Just absent. The summary had not included them, because a summary of a dispute is not a dispute, and the teams had no way to know that anything was missing.

The summary was withdrawn. The register was not shortened.

SECTION 6 — VACANCY

Custodianship

This register is maintained by the Duty Engineer, Interceptor 7.

POST NOT FILLED · SINCE 2016 · DUTIES DISCHARGED UNDER DELEGATION

The post has been vacant for nine years. The duties are discharged under a delegation that was intended to run for six weeks. Four people currently hold parts of the delegation and none of them holds all of it. The register continues to be revised. It is not clear, from the file, who revised it.

Two schools of thought exist internally. The first is that the vacancy is an administrative failure that has persisted through inattention and should be filled. The second is that the delegation works, that a single custodian would resolve disputes that four custodians are obliged to leave open, and that the register has been more honest since 2016 than before it. Nobody has written this second position down as a proposal, because writing it down would require someone with the authority to propose it, and that authority sits with the vacant post.

The register records both positions and takes neither.

SECTION 7 — SUMP LOG

Standing observations

The Sump Log is the continuous instrument record at Chamber 7-C. It is not a person and it does not have opinions, but by long convention its anomalies are transcribed into the register as numbered observations, in the flat voice used by the technician who started the practice in 1988 and did not stop.

SUMP LOG — TRANSCRIBED OBSERVATIONS
  1. Level at 7-C is always approaching the weir and rarely crossing it.
  2. This is the design condition. It is not a fault.
  3. A sump that has drained is a sump that has stopped telling you anything.
  4. The record is continuous. The record is not complete. These differ.
  5. Gaps in the record are recorded as gaps and not interpolated. Interpolation was attempted in 1996 and the interpolated values are still, occasionally, cited as measurements.
  6. The instrument at 7-C has drifted 0.02 m per decade since installation. The drift is known, uncorrected, and stated on every export.
  7. An uncorrected known drift is more useful than a corrected unknown one.
  8. Nothing at 7-C has ever happened quickly.
SECTION 8 — OPEN

Matters not carried forward

Filters for the 7-C instrument housing have not been ordered since the supplier changed part numbers. The old number is on the requisition template. The template has not been updated. Three requisitions have been rejected. Nobody has updated the template because the person who could is in Section 6.

The 2013 summary is held in the archive and is not to be circulated. The reason is recorded at §5. It is recorded there because D3 requires the reason to sit next to the thing, and putting it in an archive index would put it somewhere nobody looks.

Whether the twelfth chamber exists remains open. Whether the fourteen centimetres are settlement remains open. Whether the vacancy is a failure or a feature remains open. The register's function is not to close these. It is to make sure that when they are eventually closed, the closing is visible as an act, and someone's name is on it.