32Forms and templates with filled examples

32.0Scope of this chapter

This chapter collects the forms that the commissioning manager (CxM) and the contractors fill on a data-centre campus, each with a field table and a filled example that a site engineer can print, fill and check. The forms use the MSC-1 campus tags and numbers from the case ledger (four buildings B1 to B4, three data halls per building, 12.0 MW of IT load per hall).

1. How the forms are presented. Each form has a purpose of two sentences that names the defining chapter section and a field table (field, type, required, example value, who fills it). A filled example follows as a second table, then a numbered list of common errors on that form.
2. Document and query forms. Request for information (RFI), technical query (TQ), transmittal, submittal cover sheet, material approval request (MAR) and the cover of the method statement and risk assessment (MSRA). Chapters 04, 05 and 06 define them.
3. Quality forms. Inspection and test plan (ITP), inspection request (IR), nonconformance report (NCR), corrective action request (CAR) and punch list. Chapters 07 and 09 define them.
4. Change and decision forms. Change log and decision log. Chapters 08 and 31 define them.
5. Safety and energisation forms. Permit to work (PTW), lockout/tagout (LOTO) record with the isolation certificate, switching programme and energisation certificate. Chapter 10 defines them.
6. Meeting and report forms. Minutes of meeting (MOM), daily report and weekly commissioning (Cx) report. Chapter 11 defines them.
7. Test record forms. Level 3 (L3) checklist, Level 4 (L4) script page, point-to-point (P2P) sheet and meter verification sheet. Chapters 03, 22 and 24 define them.
8. Fibre test forms. Fibre optic (FO) Tier 1 test record and optical time-domain reflectometer (OTDR) record. Chapter 26 defines them.
9. Integrated test and issue forms. Integrated systems test (IST) scenario sheet and commissioning issue log. Chapters 03, 09 and 28 define them.
10. Handover forms. Turnover checklist and training record. Chapter 12 defines them.
11. Standards behind the forms. ISO 19650-1 and -2:2018 for container naming and status, ISO 9001:2015 for nonconformity and corrective action, and ASHRAE Guideline 0-2019 and ANSI/ASHRAE/IES Standard 202-2024 for the commissioning process and issues log. The safety, fibre and meter forms use the Occupational Safety and Health Administration (OSHA) rule 29 CFR (Code of Federal Regulations) 1910.147 and NFPA 70E-2024 for isolation, Permenaker No. 12 Tahun 2015 for K3 Listrik (electrical occupational safety and health) duties. The fibre and meter forms use ANSI/TIA-568.3-E (current edition; the loss values quoted are -D values) and ANSI/TIA-606-C for fibre records and IEC 62053-22 for meter accuracy classes.

Figures F32-1 to F32-5 show five of the forms as they print: the RFI form, the switching programme sheet, the P2P sheet, the meter verification sheet and the IST scenario sheet. Research note RF-32 lists the chapter sections and standards that each form is built from.

32.1Rules that apply to every form

32.1.1The common header block

Every form on the MSC-1 programme starts with the same header block. The header lets the common data environment (CDE) file the form, lets a reader find the system it belongs to, and shows whether the form is a draft or a signed record. The field names follow the container naming of chapter 04 (section 04.4) and the test record identifier of chapter 03 (section 03.16.2).

The template revision and the record revision are two different fields. A record filled on an old template revision is returned by the document controller, because the old template may lack a field that a later revision added.

32.1.2Signature, date and time rules

1. Sign with name and role. Every signature box holds the printed name, the role and the company. A signature without a printed name cannot be traced to the competence register.
2. Write dates in one format. The MSC-1 forms use "14 Sep 2026" or the ISO format "2026-09-14". A date such as 09/10/26 is returned, because it can be read as September or October.
3. Write times in 24-hour format. Switching programmes, permits and IST sheets record times to the minute ("09:47"). IST and sequence of events (SOE) records use the campus time server to the second.
4. Sign only what you saw. A witness signs a step or a test only when present at it. A record signed after the event by a person who was absent is a false record.
5. Correct with one line. An error on a paper form is struck through with one line, initialled and dated. Correction fluid and overwriting are not accepted.

32.1.3Required fields and the return rule

A field marked "Y" in the field tables of this chapter must be filled before the form leaves draft. A field that does not apply is filled with "not applicable" and a reason. A blank required field returns the form to the originator, and the response clock of the workflow starts only when the complete form is accepted. This rule comes from the RFI workflow in section 05.3.3 (status "returned") and is applied to every form in this chapter.

32.1.4Register of the forms in this chapter

The table lists the 30 forms with the section that defines each one and the register that keeps it. Figure F32-6 places each form at the process step where it is first raised.

32.2Document and query forms

32.2.1Request for information (RFI) form

Purpose: The RFI form asks the designer a written question about a conflict, an omission or an ambiguity in the contract documents, and records the signed answer. It is defined in sections 05.3.1 field list, 05.3.3 status values and 05.5.3 priority classes. It is the record that lets the contractor proceed on a documented answer, and it is the first link from a site question to a change, an NCR or a commissioning issue.

Worked example: RFI-B2-BMS-0231 raised by the controls subcontractor on the coolant distribution unit (CDU) 07 return-temperature sensor of hall B2-DH3. The response period for priority P2 is 5 working days; the RFI was opened on Monday 08 Mar 2027 and answered on Thursday 11 Mar 2027, 3 working days later. Figure F32-1 shows this RFI as it prints.

Common errors:

1. Two subjects in one RFI. The originator asks about the sensor location and about the pump speed signal in one form. The answer to one subject closes the RFI while the second stays unanswered. Each subject gets its own number (section 05.3.2).
2. Drawing reference without revision. "See P&ID 4410" does not say which revision the originator read. The responder may answer against a newer revision and the conflict is never seen. Every reference carries its revision code.
3. No proposed solution. An RFI without a proposed solution takes longer to answer, and the RFI manager on MSC-1 returns it. The originator always states the answer that the contractor would use.
4. Required-by date copied from the priority. The required-by date comes from the programme activity, and the priority comes from the required-by date. A P1 RFI must name the activity that is stopped and its planned date.
5. Closed by the responder. The responder answers; the RFI manager closes only after the originator confirms that the answer is usable. An RFI closed at "answered" hides answers that the site cannot use.

32.2.2Technical query (TQ) form

Purpose: A supplier or manufacturer uses the TQ form under a purchase order to ask about, or to propose a deviation from, the equipment specification before manufacture. It is defined in sections 05.11.1 use of the TQ and 05.11.2 fields in addition to the RFI fields. It carries the RFI fields plus six supply fields, so that the effect of the deviation on type tests, the factory acceptance test (FAT) and the spares list is decided before the equipment is built.

Worked example: TQ B2-MVS-TQ-017 from the 20 kV switchgear manufacturer (section 05.11.2). Figure F32-14 shows the TQ as printed, the rating check and the route of the TQ to the FAT dossier. In the MSC-1 worked example the switchgear class is 31.5 kA and the maximum 20 kV bus fault level with the bus tie closed is 25.0 kA.

A TQ that is accepted with a type test certificate still needs the certificate filed in the FAT dossier, because the FAT witness checks the rating plate against the accepted TQ.

Common errors:

1. TQ used after manufacture. The supplier sends a TQ when the panels are already built. The deviation is then a nonconformity, and an NCR with a concession is the correct record (section 09.6).
2. Effect on type tests left blank. A deviation in busbar rating, enclosure or arc classification may leave the design outside its type test. The designer cannot accept the TQ without the certificate number.
3. Spares effect ignored. A changed fan, relay or breaker part number enters the spares list (chapter 12) without anyone noting it. The spares register then lists the wrong part.
4. TQ answered by the procurement team. A technical deviation is decided by the designer of record. A procurement signature on a TQ has no technical authority.

32.2.3Transmittal form

Purpose: The transmittal records which documents, at which revision, were sent from one organisation to another, for what purpose and when the recipient acknowledged them. It is defined in sections 04.7.2 transmittal fields and 04.7.3 workflow. It is the proof that a superseded revision was withdrawn and that the site works from the current revision.

Worked example: Transmittal of the approved L4 chiller staging script to the contractors for use.

Common errors:

1. Superseded field left blank. The transmittal issues rev C and does not name rev B. The site copy of rev B stays in the folder and the test is run on it. The superseded field is required whenever the revision is above the first issue.
2. Purpose "for information" on a construction issue. A document issued for information cannot be built from. The purpose code decides what work the recipient may do.
3. Acknowledgement never chased. An unacknowledged transmittal gives no proof of receipt. The issuer lists acknowledgements outstanding beyond 2 working days in the weekly document report.
4. Documents listed without status code. The CDE suitability code (chapter 04, section 04.3.1) tells the recipient whether the document is authorised. A list without it forces the recipient to look it up in the CDE.

32.2.4Submittal cover sheet

Purpose: The submittal cover sheet heads every package of product data, shop drawings or samples that a contractor sends for review, and it states what is submitted, against which specification, with which deviations. It is defined in sections 06.3.2 cover sheet fields, 06.4.2 review periods and 06.5 review stamps. The reviewer's code (A, B, C or D, section 04.6.1) is written on the same sheet, so the cover is also the record of what work the code allows.

Worked example: Submittal B2-E-SUB-0044 rev 0 for the eight UPS systems of hall B2-DH3. The submittal was received on 03 Jul 2026. Typical hyperscale practice gives 15 working days for UPS packages (section 06.4.2), so the latest return date is 24 Jul 2026. The CxM comments were due on working day 7 after receipt (14 Jul 2026). Figure F32-7 shows the cover sheet as printed and the review timeline in working days.

Common errors:

1. Deviation hidden in the datasheet. The contractor marks "none" under deviations while the datasheet shows a different breaker frame. The reviewer reads the datasheet line by line against the specification and returns the package with code C when an undeclared deviation is found.
2. No contractor's review statement. A package that the contractor has not reviewed is returned unreviewed. The statement and signature are required before the review clock starts.
3. Commissioning review omitted. The CxM receives the package after the code is applied. A CxM comment received after the code is applied cannot be added to that review (section 06.4.2), so the CxM sets day 7 as the internal return date.
4. Code B treated as code C. Under code B the contractor proceeds and incorporates the comments; under code C fabrication of the affected items stops. The cover sheet states the code in words beside the letter.

32.2.5Material approval request (MAR) form

Purpose: The MAR asks the Engineer to approve a specific material or product, with its manufacturer, product code and compliance statement against the specification, before the material is ordered for site. It is defined in sections 06.7.2 MAR fields, 06.7.3 compliance statement and 06.7.4 samples. The approved sample is labelled with the MAR number, so the receiving inspection can compare every delivery with the approved product.

Worked example: MAR B2-FO-0011 for the building B2 backbone fibre. The specified attenuation limit is the ANSI/TIA-568.3-D maximum (check against ANSI/TIA-568.3-E) for indoor-outdoor single-mode cable, 0.5 dB/km at 1310 nm and 1550 nm (chapter 26, section 26.5); the offered values are example datasheet values.

Common errors:

1. Compliance statement copied from the specification. The contractor writes "complies" in every row without the offered value. The reviewer cannot check a statement that holds no number, and returns the MAR.
2. Sample not labelled. The approved sample stays in the room without a label. Six months later nobody can say which sample was approved. The label carries the MAR number, the review code and the date.
3. Delivery not compared with the MAR. The receiving inspection checks only the quantity. The ITP line for delivery inspection (section 07.9.2, line 2.1) names the MAR number as the reference.
4. One MAR for two products. A MAR that covers two cable types from two manufacturers gets one code for both. Each product gets its own MAR.

32.2.6Method statement and risk assessment (MSRA) cover sheet

Purpose: The MSRA cover sheet heads the method statement and risk assessment for one task and records its scope, its references, its approval chain and its risk scores before and after controls. It is defined in sections 06.9.2 MSRA structure, 06.9.3 risk rating method and 06.9.5 review procedure. The permit issuer checks the MSRA number and revision on this sheet before issuing a permit for the task (chapter 10, section 10.1.4).

Worked example: MSRA cover for the first energisation of 20 kV switchgear B2-MVS-01 (section 06.9.2 example and 06.9.3 worked example). The risk method is the 5 × 5 matrix: scores 15 to 25 are high and the task does not start, 8 to 12 are medium, 1 to 6 are low.

Common errors:

1. Only one risk score. The MSRA gives the score after controls and omits the score before controls. The reviewer cannot see whether the controls are the reason the task is acceptable. The MSRA records both scores and the controls.
2. Prerequisites written as intentions. "Tests will be done" is not a prerequisite. Each prerequisite names the record, for example "VLF record TS-02-02 signed".
3. Generic MSRA reused. An MSRA written for building B1 is reused for B2 with the B1 tags and room numbers. The permit issuer compares the tags on the cover sheet with the permit request and refuses a mismatch.
4. Revision on the permit differs from the approved revision. The permit lists rev 0 while rev 1 is approved. The permit issuer checks the revision against the document register.

32.3Quality forms

32.3.1Inspection and test plan (ITP) form

Purpose: The ITP lists, in construction order, every inspection and test of one work package, with the acceptance criterion, the record and the intervention code of each party. It is defined in sections 07.3.1 header block, 07.3.2 line columns and 07.2.1 intervention codes. It is the plan that the IR, the checklists and the test sheets execute, and it decides which activities stop the work until a written release.

The ITP has a header block and a line table. The field table covers both.

Worked example: Six lines of ITP-ELE-MV-002 rev 1 for the 20 kV cables of building B2 (lines from section 07.9.2). The party columns are GC (GC), Engineer (ENG) and commissioning (CX).

A hold point stops the work until a written release is received. A witness point lets the work continue after the notice period when the notified party does not attend, and the record then states "notified, not attended" (section 07.2.1).

Common errors:

1. Acceptance criterion without a value. "As per specification" gives the inspector nothing to measure. The criterion names the value or the condition and its source clause (section 07.3.3).
2. Hold point given to every line. An ITP with 40 hold points stops the work 40 times and the inspectors cannot attend them all. Hold points are kept for activities that are covered up or energised next, such as lines 3.4, 5.2 and 6.3.
3. CxM column missing. The ITP has columns for the contractor and the Engineer only. The CxM then has no witness right on the VLF test. The CxM reviews every ITP for the systems in the commissioning plan and asks for a CX column.
4. Line numbers changed between revisions. Rev 1 renumbers the lines and the IR log still quotes the rev 0 numbers. The IR records the ITP number, revision and line together.

32.3.2Inspection request (IR) form

Purpose: The IR gives formal notice that work at an ITP intervention point is ready, records the inspection and its result, and releases or rejects a hold point in writing. It is defined in sections 07.5.2 IR form fields, 07.5.3 notice periods and 07.5.4 response codes. It is the record that proves a covered-up or energised item was inspected before it was covered or energised.

Worked example: IR for the VLF withstand hold point on the three cores of feeder B2-MV-F07 (ITP line 5.2). A routine hold point needs 24 hours of notice; a hold point before energisation needs 48 hours, with the energisation permit request at the same time (section 07.5.3).

Common errors:

1. Notice counted from the time sent by email. Notice periods count from the time the IR is received in the document control system. An IR emailed at 17:30 on a Friday gives no working hours of notice until Monday.
2. Self-inspection skipped. The contractor raises the IR before its own QC engineer has checked the work. The inspector finds open items and the IR is rejected with code C, which counts as a failed IR in the key performance indicator (KPI).
3. Witness absence not recorded. At a witness point the party does not attend and the contractor proceeds without writing it down. The IR is closed with code NA and the notice given, so the record shows the party was notified.
4. Late cancellation not counted. A cancellation less than 4 working hours before the inspection counts as a failed IR in the KPI (section 07.5.4). The register records the cancellation time.

32.3.3Nonconformance report (NCR) form

Purpose: The NCR records work or material that does not meet a stated requirement, its containment, the approved disposition, the correction and the verification that closes it. It is defined in sections 09.3 NCR form fields, 09.4.1 severity, 09.5.1 workflow and 09.6.1 dispositions. It is the ISO 9001:2015 clause 8.7 record of the control of nonconforming outputs on the project. Figure F32-8 shows the order of questions that decides whether a finding is recorded as an NCR, a change, an RFI, a TQ or a punch item.

Worked example: NCR on a Modbus register map version found during the EPMS point check of UPS B2-DH3-UPS-A3 (an example in the pattern of section 09.13.3). The NCR is major because a wrong alarm mapping affects the function of the monitoring system.

Common errors:

1. Description without numbers. "Busway joints not tight" cannot be verified. The description states how many items, which items and by how much: "9 of 48 joints below the label torque".
2. Correction recorded as corrective action. Re-torquing the joints is the correction. Changing the method so that it does not happen again is corrective action, recorded on a CAR (section 09.1.2).
3. Use as is without the designer. Use as is and repair always need a concession and the designer's written agreement (section 09.6.1). An NCR closed "use as is" with only the QA/QC manager's signature is reopened.
4. Containment missing. The NCR describes the defect and the disposition but not the containment. Containment must be done within 1 working day, and for electrical work it means a lockout or withdrawal of the energisation permit.
5. Retest not named. The joints are re-torqued and the NCR is closed without the insulation resistance and thermographic retests. The retest field lists every test that the correction invalidated.

32.3.4Corrective action request (CAR) form

Purpose: The CAR asks the responsible organisation to find the root cause of a major NCR, a repeated minor NCR or a process failure, to remove the cause and to prove that the action worked. It is defined in sections 09.7.1 when a CAR is raised, 09.7.2 CAR fields and 09.7.7 effectiveness. It is the ISO 9001:2015 clause 10.2 record of corrective action.

Worked example: CAR-B2-ELE-0014 for the busway joint torque NCRs of building B2 (section 09.7.2).

Common errors:

1. Root cause stops at the person. "Installer error" is a symptom. The 5-why continues until it reaches a process that can be changed, such as the method statement review.
2. Effectiveness check with no sample size. "Checked, fine" gives no measure. The check names the sample (the next 3 runs, 144 joints) and the result as a number.
3. CAR closed on the action date. The CAR is closed only after the effectiveness result is recorded, which on MSC-1 was 4 weeks after the action.
4. No preventive action in other places. The same method applies to other bolted joints. The CAR names every other place where the cause can act.

32.3.5Punch list form

Purpose: The punch list records each item that is incomplete or incorrect at a punch walk, with a category that decides by when it must be closed, and records the verification of each correction. It is defined in sections 09.8.2 punch categories, 09.8.4 punch list fields and 09.8.5 punch walk procedure. On MSC-1 the case ledger sets the ageing limits: category A 1 working day, B 7 working days, C 30 working days.

Worked example: Four items from the pre-L4 punch walk PW-B2-DH3-L4-02 in hall B2-DH3 on 6 Oct 2026. Due dates are working days from 6 Oct 2026 (Tuesday).

Common errors:

1. Category chosen by convenience. A missing earth bond is entered as category B so that the test can start. Category A covers any item that affects safety, energisation, filling or the validity of the next test (section 09.8.3).
2. No requirement stated. "Looks wrong" is not a punch item. The requirement field names the drawing, manual or specification clause the item fails.
3. Closed without a functional check. A connected leak cable is closed without a leak simulation. For a sensor or an alarm the action taken includes the test that proves it.
4. Items transferred to an NCR and left open on both. When an item becomes an NCR, the punch status is set to "transferred to NCR" with the NCR number, and only the NCR stays open.

32.4Change and decision forms

32.4.1Change log form

Purpose: The change log is the register of every potential and approved change, from its origin through notice, proposal, valuation, decision and the commissioning test that closes it. It is defined in sections 08.10.1 register fields, 08.10.2 status codes and 08.9.2 commissioning impact. The CxM uses it to see which ITPs, scripts and test days each change adds before the change is approved. Figure F32-9 shows the status codes 0 to 8, the field that each code requires and the two example changes VO-B2-017 and VO-B2-023.

Worked example: Two rows of the B2 change log on 15 Mar 2027. The second row is the CT ratio change described in section 08.15.

Common errors:

1. Commissioning impact left blank at approval. A change approved without the test days and script revisions arrives at the CxM after the schedule is fixed. The CxM fills the impact field before status 6.
2. Change closed at "implemented". Status 7 means the work is built. Status 8 needs the commissioning record that shows the changed system tested.
3. Defect logged as a change. A finding that the contractor's work does not meet the contract is a defect. It is recorded as an NCR or a Cx issue, and section 08.12 gives the decision.
4. Notice date missing. A change without its contract notice date cannot be checked against the FIDIC 28-day or NEC4 8-week notice periods (section 08.16).

32.4.2Decision log form

Purpose: The decision log records each decision made by a manager on the programme with its owner, the options considered, the reason, the cost, the contract effect and the actions. It links each decision to the confirming email sent the same working day. It gives the CxM a dated record of who decided what, which protects the test programme when a verbal decision is later disputed.

Worked example: Decision DL-B2-044 on a deferred L4 retest for the CDU pump changeover in hall B2-DH3 (an example built on the fields above).

Common errors:

1. Decision written as a discussion. "Discussed IST start" is not a decision. The decision field is one sentence that someone can act on.
2. No confirming email. A verbal decision recorded only in the log is disputed later. The confirming email is sent the same working day and its link is entered.
3. Disagreement not recorded. When the CxM recommended a different option, the entry says so: "CxM recommended option A; programme director decided option B".
4. Superseded decisions left open. A later decision that replaces DL-B2-031 is entered with its own number, and DL-B2-031 is marked superseded with the new number.

32.5Safety and energisation forms

32.5.1Permit to work (PTW) form

Purpose: The PTW is the written authority to do one task in one place for one period under stated precautions. The permit issuer, the area authority and the permit receiver sign it, and it carries a briefing record and a hand-back. It is defined in sections 10.1.3 permit types, 10.1.4 fields of the permit form and 10.1.5 life cycle. On a partly live data-centre campus the PTW is the document that keeps a work party away from energised equipment and links the task to its isolation certificate.

Worked example: Electrical isolation permit for work on UPS system B2-DH3-UPS-A1 under isolation certificate IC-B2-LV-0388 (section 10.2.5). The request was submitted at least 2 working days before the work (section 10.1.5, step 1).

Common errors:

1. Permit issued without a joint walk-down. The issuer signs in the permit office. The issuer and the receiver walk the area together and check each precaution before either signs (section 10.1.5, step 5).
2. Validity past the shift. A permit valid from 07:30 to 07:30 the next day covers two shifts with one briefing. Electrical isolation permits are valid for one shift and renewed daily.
3. Isolation removed before cancellation. The isolating authority removes the locks when the receiver hands back. Only after cancellation can the isolating authority remove the isolation.
4. Briefing signed by the supervisor only. Each worker signs the briefing section and applies a personal lock. A permit with one briefing signature for six workers is suspended.

32.5.2Lockout/tagout (LOTO) record: isolation certificate and lock register

Purpose: The LOTO record has two parts. The isolation certificate lists every energy source, isolation point, lock and the proving-dead result of one isolation, and the lock and tag register tracks each lock from application to removal. It is defined in sections 10.2.5 isolation certificate, 10.2.6 group lockout and 10.2.7 lock and tag register. Together they are the record of the six steps of OSHA 29 CFR 1910.147 and of the electrically safe work condition of NFPA 70E-2024 for one item of equipment.

Worked example: Isolation certificate IC-B2-LV-0388 for UPS B2-DH3-UPS-A1 with its lock register rows (sections 10.2.5, 10.2.7 and the isolation programme in section 10.4). Before the isolation the load of 1.56 MW is transferred to the maintenance bypass breaker (MBB), which stays closed by key K1 and carries the load during the work. The proving-dead test uses the live-dead-live method: the detector is proved on a known live source before and after the test (section 10.3.2). Figure F32-10 shows the six isolation points and the times of the certificate.

Common errors:

1. One energy source missed. A UPS has a module input, a rectifier input, a static bypass input, a battery, an inverter output and a back-feed path from the live load bus through the output isolator. A certificate that lists only the input breaker leaves the battery, the bypass and the back-feed live. The energy sources table is copied from the as-built single-line diagram.
2. Detector not re-proved after the test. A detector that failed during the test reads 0 V on a live conductor. The certificate records both proving times and the instrument serial number.
3. Lock keys outside the key safe. The isolating authority keeps a lock key in a pocket. Keys go in the numbered key safe, and the key safe is locked by the group lock box.
4. Lock removed by another person. A lock is removed only by its owner, or under the absent-owner procedure of section 10.2.8 with the written authority it requires.

32.5.3Switching programme sheet

Purpose: The switching programme is the written, checked and approved sequence of switching operations for one network change, with one action per numbered step, the expected indication of each step, the hold points and the signatures. It is defined in sections 10.4.3 format, 10.4.4 approval timeline and 10.4.5 worked example. The senior authorised person (SAP) operates only from the approved programme, and no step is improvised on the day.

Figure F32-2 shows the sheet as it prints.

Worked example: Steps 9 to 16 of SP-B1-HV-007 rev 1 (section 10.4.5). In the MSC-1 worked example two 20 kV feeders of 1,334.6 A each supply B1-MVS-01, and ring R1 loops through RMU-01 to RMU-04 with the normal open point at RMU-02 S2. The expected charging current at step 11 for about 1.2 km of cable is near 1 A; a reading of tens of amperes means a fault or a closed tee-off.

The full programme has 24 steps and 3 hold points. With read-back each step takes 2 to 4 minutes, so the 24 steps take about 75 minutes plus the hold points, inside the 4-hour window.

Common errors:

1. Two operations in one step. "Open breaker P03 and close earth switch P03" is two steps. A combined step has one expected indication for two device states.
2. Programme written from the design drawing. The panel order on site differs after a change. The checker walks the route with the as-built mark-up 5 working days before the switching day.
3. Missing check steps. A programme with only "open" and "close" steps has no proof of the state between steps. Every operation is followed by an indication check.
4. Step changed on the day. The SAP adds a step in pen. A new or changed step needs a written revision approved by the control person; otherwise the SAP stops, makes the equipment safe in its current state and calls the control person.

32.5.4Energisation certificate

Purpose: The energisation certificate records that a defined scope of equipment was energised under an approved switching programme after the energisation readiness board (ERB) approved it, and it declares the new live boundary. It is defined in sections 10.5.3 gate criteria, 10.5.4 energisation certificate and 10.5.5 energised-zone declaration. It is the document that changes the permit rules in the named rooms from construction rules to live-equipment rules.

Worked example: Energisation certificate for the first block transformer of hall B2-DH3, energised after the 20 kV ring of building B2 was proved. The block transformer is 2.5 MVA (MSC-1 ledger), with an LV full-load current of 3,478 A at 415 V.

Common errors:

1. Scope written as "LV room". The certificate names every energised device and every excluded device by tag. A vague scope lets a contractor work on a feeder that became live.
2. Live boundary not posted. The certificate is signed but the doors of the new live rooms carry no sign. The energised-zone declaration includes the signs and the access control change.
3. ERB decision missing. The equipment is energised on the switching programme alone. The ERB decision number is a required field, because the ERB checks the gate criteria that the switching programme does not.
4. Distribution late. The PTW office receives the certificate the next day and issues a construction permit in a room that is now live. Distribution is part of the same-day post-switching report.

32.6Meeting and report forms

32.6.1Minutes of meeting (MOM) form

Purpose: The MOM records who attended a meeting and each item as information, decision or action, with an owner, a due date and the closure evidence for each action. It is defined in sections 11.3.1 minutes template, 11.3.2 rules and 11.4.1 action log. Actions in the minutes are copied to the action log the same day, and each action is tracked to closure in the action log.

Worked example: Four items of the weekly Cx coordination meeting of building B2, meeting 52, held on 19 Mar 2027 (items in the format of section 11.3.1).

Common errors:

1. Actions without a named person. "Controls to fix" names a company. The action owner is a named person who can be called.
2. Minutes written as a transcript. Minutes record decisions and actions in full sentences. A record of who said what makes the decisions hard to find.
3. Actions not copied to the action log. The action exists only in meeting 52 minutes and is forgotten by meeting 54. The minute-taker enters every action in the action log the same day with the source "52.4".
4. Distribution after a week. Minutes issued a week later reach the owners after their due dates. MSC-1 issues minutes within 1 working day.

32.6.2Daily commissioning report form

Purpose: The daily report records, for one building and one shift, the commissioning personnel on site and the checklists and tests completed, passed and failed. It also records the issues raised and closed, the blockers and the tests planned for the next day. It is defined in sections 11.5.1 purpose and content and 11.5.2 daily report fields. It is the source of the daily numbers that the weekly report and the S-curves add up.

Worked example: Daily report DR-B1-2026-11-05 for building B1 (section 11.5.2). The pump vibration limit of 4.5 mm/s is the project specification value quoted in that section; it comes from the pump vibration clause of the project mechanical specification.

Common errors:

1. Counts without the plan. "42 checklists signed" does not show whether the day was on plan. The report gives the planned number beside the actual number.
2. Failures without the value. "Pump failed" gives no information. The report states the measured value and the limit, for example 7.4 mm/s against 4.5 mm/s.
3. Blocker without an owner and date. A blocker that nobody owns stays a blocker. Each row names the owner and the date by which it is cleared.
4. Numbers that do not match the registers. The daily report says 11 L3 passes while the test register shows 10. The daily numbers are taken from the registers, never typed from memory.

32.6.3Weekly commissioning report form

Purpose: The weekly Cx report gives the owner and the programme managers the status of commissioning for one building in twelve fixed sections. The sections run from a five-line summary to the detailed appendix, and each number is taken from a named register. It is defined in sections 11.6.2 sections of the weekly report, 11.6.3 calculation rules and 11.6.4 worked example. It is the record that the programme reviews against the schedule milestones of energisation, L4 completion, IST and ready for service (RFS). Figure F32-11 shows the test and issue counts of sections 5 and 7 of report WR-B2-2027-W12.

Worked example: Sections 1, 5 and 7 of weekly report WR-B2-2027-W12 for building B2 (example values). The MSC-1 first-pass yield target is 90 % (case ledger), and the defect ageing limits are 1, 7 and 30 working days for categories A, B and C.

Common errors:

1. First-pass yield calculated on retests. A retest that passes is counted as a pass in the first-pass yield. First-pass yield counts only tests that passed at the first run, divided by tests run (section 11.6.3).
2. Summary longer than five lines. The readers of section 1 read nothing else. The summary is five lines and each line carries a number or a date.
3. Sections dropped when empty. A week with no safety events omits section 2. Every section appears every week, with "none" where it applies, so the report can be compared week to week.
4. Forecast dates without a reason. A milestone moves by two weeks with no explanation. The milestone table gives the reason and the decision or issue number behind each change.

32.7Test record forms

32.7.1Level 3 (L3) checklist form

Purpose: The L3 checklist records the start-up and the component-level checks of one item of equipment. It holds the prerequisites, the OEM start-up steps, the settings file version and the measured values with their acceptance values and sources. It also holds the point checks, the local alarms and the findings. It is defined in sections 03.8.1 structure, 03.8.2 filled L3 checklist for a UPS module and 03.16.2 record identifier. It is signed by the OEM, the contractor and the CxA, and its signature releases the L3 gate and the green tag of the item.

Worked example: Ten of the 28 rows of L3 checklist ELE-L3-UPS-B2-DH3-A1-M2-01 rev 0 for UPS module 2 of block B2-DH3-A1 (section 03.8.2). The block UPS system is 2.5 MW (MSC-1 ledger). The values are examples; the approved OEM start-up procedure and specification 26 33 53 govern.

All 28 rows passed and the checklist was signed by the OEM start-up engineer, the electrical contractor and the CxA on 2027-02-24.

Common errors:

1. Settings file version not recorded. A later setting change cannot be traced to the tested version. Row 12 is filled on every L3 checklist.
2. Alarm checked on the display only. The EPMS and BMS paths are then unproven. The alarm rows record the arrival time at the EPMS or BMS.
3. Acceptance column blank on the template. The performer writes "ok" because the template gives no value. Every acceptance cell carries a value and its source before the template is approved.
4. Ohmic values without the string average. The deviation rule needs the average, so the row records the average beside the maximum deviation.

32.7.2Level 4 (L4) functional test script page

Purpose: The L4 script page is one page of a functional performance test script. It holds the header that identifies the script and its revision, and the step table. Each step in the table has one action, the expected response from the sequence of operation, the acceptance value with its source, the measured value, the result and the witness initials. It is defined in sections 03.9.1 script anatomy, 03.9.2 step table columns and 03.9.3 filled 20-step chiller staging script. A step result is pass, fail or not tested, and a script status is not started, passed, failed or retest.

Worked example: Page 2 of script MEC-L4-CHW-B2-STG-01 rev C with steps 10 to 14 (section 03.9.3). The plant serves the B2 air halls on the cold loop and has five chillers of 2.5 MW each (four duty, one standby). This is an example value sized from the MSC-1 cold-loop duty of 36.5 MW for the campus. The warm-loop trim chillers of chapter 18 are a separate plant of 8 duty and 1 standby units of 2,500 kW. The chilled water supply setpoint is 18.0 °C and the return 24.0 °C (MSC-1 ledger). The test ran on 14 Apr 2027 with heater skids HS-01 to HS-04 (4 × 2.5 MW) as the load.

The full script passed 20 of 20 steps at the first attempt, which gives a first-pass yield of 100 % for the script. Figure F32-15 shows the heater load and the running chiller capacity for all 20 steps of the script (section 03.9.3).

Common errors:

1. Two actions in one step. "Trip CH-02 and observe CHWS" mixes an action and an observation. Steps 13 and 14 are separate, each with its own acceptance value.
2. Timers shortened for the test. The integrator shortens the 10 min stage-up timer to save time. Check the timer values in the controller before step 1 and record them on the page.
3. Fault simulated in the BMS. A fault forced in the BMS software does not test the chiller's own fault output. Step 13 opens the chiller control circuit at the chiller.
4. "In progress" written as a result. A step result is pass, fail or not tested. A step that cannot be completed is "not tested" with the reason in the measured column.

32.7.3Point-to-point (P2P) sheet

Purpose: The P2P sheet records the test of every BMS point of one piece of equipment along its whole path. The path runs from the physical quantity at the field device through the controller input or output and the BACnet object. It ends at the value shown on the graphic, in the alarm list and in the trend. It is defined in sections 22.6.1 definition and method, 22.6.2 procedure and 22.6.3 filled P2P sheet for CDU 07. The difference between the reference value and the graphic value is judged against the sensor tolerance of section 22.7, and every failed point becomes an issue with a number.

Figure F32-3 shows the sheet as it prints.

Worked example: Eight of the 30 rows of P2P sheet P2P-B2-DH3-CDU07 (section 22.6.3). Reference instruments: probe TC-REF-04 (certificate CAL-2027-118, expiry 2027-11-30), clamp-on flow meter FL-REF-02 (CAL-2027-090, expiry 2027-09-15), pressure calibrator PR-REF-03 (CAL-2027-101, expiry 2027-10-20). Permit PTW-B2-0457.

The TCS_RetT failure is issue ISS-B2-0812, category B. A lead-wire compensation of 1.3 Ω was entered and the point was retested on 2027-04-14 with a difference of +0.02 K; the retest row is added to the sheet and the sheet is re-signed.

Common errors:

1. Reading the controller instead of the graphic. The graphic can show a different object from the one tested. The BMS value column is read from the graphic.
2. Transmitter tested at one value. A 4–20 mA transmitter tested only at room conditions hides a span error. Test transmitters at 4, 12 and 20 mA with a loop calibrator, then compare with the reference.
3. OEM display used as the reference. The OEM display reads the same sensor as the BMS and proves only the communication path. Sensor accuracy needs an independent calibrated reference instrument.
4. Forced values left in place. A point forced for the test and not released reports a false value for months. The override summary row is filled "empty" before the sheet is signed.

32.7.4Meter verification sheet

Purpose: The meter verification sheet records the comparison of an EPMS meter with a calibrated reference power analyser at several load points. The error on kW, voltage and current is judged against the tolerance for the meter and CT class. It is defined in sections 24.6.1 acceptance tolerance, 24.6.2 procedure and 24.6.3 meter verification sheet. The error is calculated as (meter − reference) / reference × 100, and every error at every load point must be inside the tolerance.

Figure F32-4 shows the sheet as it prints.

Worked example: Meter verification for B2-DH3-LVMSB-A1 meter M-A1 (PM8000) during the L4 load bank test, at five points of the 2,048.7 kW load-bank test load (section 24.6.3; 3,000 A per phase on the LV main, a different quantity from the 1.5 MW block IT load).

Check of the 10 % row: (205.6 − 204.9) / 204.9 × 100 = +0.34 % from the rounded values on the sheet; the analyser computed +0.32 % from the unrounded means. Both are inside ±0.5 %. The worst error occurs at the lowest load, which is expected because meter error grows at low current. On the same board, feeder meter M-F07 (PM5000 on a class 0.5 CT, 300/5 A) read +1.8 % at 10 % load against ±1.0 % and failed. The 2.5 mm² CT leads over 18 m were replaced with 4 mm² leads, and the retest gave +0.6 %.

Common errors:

1. One load point only. A meter checked only at 50 % load hides the low-current error. The sheet has five load points, or a later repeat when the load cannot be set.
2. CT ratio missing from the reference. The reference reads secondary amperes while the meter reads primary amperes. The CT ratio is entered in the reference before the first reading (section 24.6.2, step 4).
3. Readings taken at different times. On a varying load, a meter reading and a reference reading 2 minutes apart differ by the load change. Readings are taken within 10 s of each other, three times, and the mean is used.
4. Test block links closed in the wrong order. Opening a CT circuit under load produces a dangerous voltage. The links are opened and closed in the order on the test block label, and the meter current is compared before and after.

32.8Fibre test forms

32.8.1Fibre optic (FO) Tier 1 test record

Purpose: The FO Tier 1 test record holds the optical loss test set (OLTS) result of one fibre link at each wavelength. It also records the reference method, the test cords, the loss budget limit, the margin and the polarity. It is defined in sections 26.5.2 loss budget formula, 26.6.2 polarity test and 26.8.5 Tier 1 test record fields. The limit is the ANSI/TIA-568.3-D budget, or the engineered budget where the specification sets one, programmed into the tester before the test.

Worked example: Tier 1 record for the OS2 campus link from the building B2 main distribution frame (MDF) to the campus core (section 26.5.3). Budget = 1.20 km × 0.5 dB/km + 2 × 0.75 dB + 2 × 0.3 dB = 0.60 + 1.50 + 0.60 = 2.70 dB at both wavelengths. The 10GBASE-LR channel insertion loss limit of 6.2 dB in IEEE 802.3 Clause 52 is higher, so the TIA budget governs. The 6.2 dB value is 10 km × 0.42 dB/km plus 2 dB of connection loss, as listed for Clause 52 in IEEE P802.3dj task force paper stassar_3dj_optx_01a (April 2023).

Common errors:

1. Two-jumper reference used with a one-jumper limit. The two-jumper reference removes one connector from the measured loss and gives an optimistic result. The reference method on the record matches the method assumed by the budget.
2. Limit typed after the test. The technician enters the limit in the spreadsheet after the tester has run. Program the limit in the tester before the test, so that the tester itself reports pass or fail.
3. Length from the drawing. The drawing length differs from the installed length. The length is taken from the sheath markings or the OTDR, and the drawing length is a check only.
4. No end-face inspection recorded. A dirty connector raises the loss and can damage the reference cord. The record shows the inspection result before the loss test.

32.8.2Optical time-domain reflectometer (OTDR) record

Purpose: The OTDR record holds the Tier 2 trace settings and the event table of one fibre in each direction. It also holds the bidirectional average of every splice and the pass or fail of each event against its limit. It is defined in sections 26.9.1 what the OTDR shows, 26.9 set-up table and 26.9.7 OTDR event table fields. The bidirectional average, (loss A to B + loss B to A) / 2, is the accepted splice loss on single-mode fibre, because it cancels the gainer effect at splices between fibres with different backscatter. Figure F32-12 shows the trace drawn from the event table of the worked example below.

The record shows these set-up values (section 26.9).

Worked example: Event table of link B2-MDF to CAMPUS-CORE, fibre 01, at 1550 nm from end A with a 500 m launch cord and a 500 m tail cord (section 26.9 worked example).

Total from the trace: 0.32 + 0.06 + 0.06 + 0.45 + 0.30 + 1.20 km × 0.19 dB/km = 1.42 dB at 1550 nm, which agrees with the Tier 1 result. After the tray at E4 was re-routed to the minimum bend radius, the link loss was 1.01 dB.

Common errors:

1. One direction only. A single-direction trace reports the gainer at E3 as −0.06 dB and hides the real loss. Splice loss is the average of the two directions, so the record holds both traces.
2. No launch cord. Without the 500 m launch cord the near-end connector sits in the OTDR dead zone and its loss cannot be measured.
3. Ghost logged as a fault. A spike beyond the fibre end with no loss step, at twice the distance of a strong reflection, is a ghost. It is recorded and the strong reflection is cleaned.
4. Wrong IOR. A default IOR shifts every event distance and the bend at E4 is searched for at the wrong place. The IOR is taken from the cable datasheet and written on the record.

32.9Integrated test and issue forms

32.9.1Integrated systems test (IST) scenario sheet

Purpose: The IST scenario sheet records one Level 5 failure scenario of a data hall against time from the initiating action. Each line gives the system, the expected response, the acceptance value and its source, the observed value with its time stamp and the result of each line. It is defined in sections 03.10.2 scenario sheet columns, 28.3.2 script header fields, 28.3.3 step format and 28.11 script S01. A line result is pass, fail or not observed, and a failed scenario is closed only by a retest record on the same script with the same acceptance values (section 28.15.3).

Figure F32-5 shows the sheet as it prints.

Worked example: The filled sheet of scenario S01 for hall DH3 of building B2 (section 28.11 and 28.11.1). T0 is the relay test trip of both 20 kV incomers B2-MV-IN-1 and IN-2. The generator bus load acceptance is 12.0 MW × power usage effectiveness (PUE) 1.22 = 14.6 MW (MSC-1 ledger).

The battery supplied the block for the 12.4 s from T0 to the generator bus closing. At a block load of 1.5 MW this is 1.5 × 12.4 / 3,600 = 0.0052 MWh = 5.2 kWh, about 2 % of the 301 kWh battery screen value (MSC-1 ledger).

Common errors:

1. Observed column filled from memory. The observed values are written after the scenario from what the leads remember. Each observed value comes from a logger, a PQ analyser file or an SOE record with its time stamp.
2. Acceptance without a source. "Temperature acceptable" cannot be judged. Each acceptance value names the OPR, the specification, the standard or the OEM data it comes from (section 28.7.1).
3. Conditional pass with no issue. An unexpected alarm is noted and forgotten. A conditional pass always carries an issue number, here IST-DH3-014.
4. Initial conditions not proven. The scenario starts with the load below 12.0 MW or the batteries below 100 %. Hold point 1 at T0 − 15 min confirms the load, the loggers and the breaker positions before T0.

32.9.2Commissioning issue log form

Purpose: The issue log is the ASHRAE Standard 202 issues and resolution log. It lists each issue found during commissioning with its source record, category, responsible party, due date, corrective action and the retest that closes it. It is defined in sections 03.11.1 purpose and fields, 09.9.1 issues log and 28.15.3 issue log fields for IST. The due date follows from the category ageing limit, and the issue is closed only with the record of the retest or verification.

Worked example: Four rows of the issue log of building B2 on 25 Mar 2027 (rows from sections 03.11.1, 09.9.1, 22.6.3 and 28.11.1). Category B has a limit of 7 working days.

The MSC-1 form uses one field, category A, B or C, with the definitions and ageing limits of section 09.8.2. The rule is stated in section 09.9.1, and sections 03.11.1, 28.15.3 and 30.2.4 use the same field. When a contractor's own system uses P1 to P3, the CxM states the mapping to A, B and C in the Cx plan.

Common errors:

1. Description without the measured value and the limit. "Changeover slow" cannot be retested against anything. The description gives the measured value and the acceptance value.
2. Closed on the contractor's word. An issue moves to "ready for verification" when the contractor reports the fix, and to "closed" only when the CxA records the retest record ID.
3. Two schemes for urgency in one log. A log that uses P1 to P3 for some rows and A to C for others cannot produce one ageing report. One field and one scheme are used.
4. Source record missing. An issue without its record and step cannot be retested on the same script. The source record field is required.

32.10Handover forms

32.10.1Turnover package checklist

Purpose: The turnover checklist records, for one system, the receipt and review status of each of the 13 sections of the turnover package, from the index and contacts to the open items list. Each section shows the items expected, received, accepted and rejected. It is defined in sections 12.3.1 what a turnover package is, 12.3.2 worked example and 12.3.3 review procedure. The document controller checks completeness within 2 working days, the reviewers check content within 10 working days, and the DCOps lead signs the acceptance sheet.

Worked example: First review of turnover package TP-B1-CHW-01 for the chilled-water plant of building B1 (section 12.3.2). Seven of the 13 sections are fully accepted.

Common errors:

1. Package assembled per subcontractor. The operations team works by system, so a package that follows the subcontract boundaries is returned. The checklist is per system.
2. Counts without the expected number. "46 drawings received" means nothing without the expected count from the drawing register. The expected column is filled before the package arrives.
3. Software backups omitted. Section 12 is left empty because controller backups are "with the integrator". Controller backups and licences are part of the package and are handed over under control.
4. Open items list accepted with category A items. A package with an open category A issue is not ready for acceptance. Section 13 shows the count of open items by category.

32.10.2Training record form

Purpose: The training record proves that a session of an approved training module was delivered to named members of the operations team. It records the attendance time, the assessment results, the follow-up for anyone who failed and the location of the recording. It is defined in sections 12.9.1 training plan, 12.9.2 procedure and 12.9.3 training record fields. The training plan sets a pass mark of 80 % (section 12.9.1), and the record of section 12.9.3 counts the attendees present for at least 90 % of the session.

Worked example: Session TRN-B1-CHW-02-S3 for shift team 3 of building B1 (section 12.9.3). The module is 6 hours: 3 hours in the classroom and 3 hours on the plant.

Common errors:

1. Attendance sheet without times. A signature at the start of the session proves only arrival. The record shows the time present for each attendee.
2. No assessment. A record of attendance alone does not prove that the engineer can operate the plant. Each critical module has a practical assessment with a pass mark.
3. Failed attendee left without follow-up. The person who scored 70 % is not booked again and later operates the plant alone. The follow-up field names the next session.
4. Recording not linked. The session was recorded but the file is not in the CDE. New shift engineers then have no access to it. The recording link is a required field.

32.11How the forms connect

32.11.1Record flow from a question to a closed test

The forms of this chapter link to each other through their number fields. The table shows the links that the CxM checks during the weekly document audit. Figure F32-13 draws six of these chains with their example numbers and the result of the week 12 audit.

32.11.2Weekly audit of the forms

Purpose: Find broken links and blank required fields before they reach a witness test or a handover review.

Owner: CxM, with the GC document controller.

Steps:

1. Sample ten records per form type. Each Monday the document controller draws ten closed records of each form type from the registers. The sample takes about 2 hours for the 30 forms.
2. Check the required fields. Every field marked "Y" in this chapter is filled, or filled "not applicable" with a reason.
3. Follow each link. Each linked number opens the linked record in the CDE, and the linked record points back.
4. Check the signatures. Each signature has a printed name and role, and each witness signature is from a person on the attendance list of the test.
5. Record the result. The audit result is entered in the weekly Cx report, section 10, with the number of records checked and the number of defects found.

Acceptance: Zero closed records with a blank required field; every link opens a record that links back.

Records: Weekly document audit sheet (chapter 04, section 04.14.1); weekly Cx report section 10.

Worked example: In week 12 of 2027 the B2 document controller sampled 300 closed records (10 of each of the 30 forms). Seven defects were found: 3 P2P sheets with no override summary entry, 2 IRs with no notice period, 1 NCR without the retest field and 1 issue log row closed with no evidence record. The defect rate was 7 / 300 = 2.3 %. All seven records were reopened, completed and re-closed within 2 working days.

32.12Checklist for site use

1. Use the current template. Check the form code and template revision against the document register before filling a form.
2. Fill the header block. Record number, CDE identifier, building, hall, asset tags, revision and status are filled on every form.
3. Leave no required field blank. Fill "not applicable" with a reason where a field does not apply.
4. Write values with units and sources. Every acceptance value names its source clause; every measured value carries its unit.
5. Quote revisions. Every drawing, script, MSRA and ITP reference carries its revision code.
6. Name a person. Every action, issue and punch item has a named owner and a due date.
7. Sign only what you witnessed. Witness signatures are given at the step, with printed name, role and time.
8. Link the records. Fill the linked-item fields so that an RFI, NCR, issue, change and retest can be followed in both directions.
9. Close with evidence. An NCR, CAR, punch item or issue is closed only with the record ID of the verification or retest.
10. Check permits and isolations together. A PTW number, its isolation certificate and the lock register rows agree before work starts.
11. Use the approved switching programme only. No step is added or changed on the day without a written revision.
12. Program limits before testing. Fibre budgets, meter tolerances and P2P tolerances are set before the first reading.
13. Remove test overrides. Every P2P and L4 sheet records that no forced value or override is left.
14. Issue minutes and reports on time. Minutes within 1 working day; the daily report the same day; the weekly report on its fixed day.
15. Audit ten records per form each week. Enter the defect rate in section 10 of the weekly Cx report.

32.13Sources

1. ISO 19650-1:2018, Organization and digitization of information about buildings and civil engineering works, including BIM, Part 1: Concepts and principles, ISO. https://www.iso.org/standard/68078.html
2. ISO 19650-2:2018, Part 2: Delivery phase of the assets, ISO, 2018. https://www.iso.org/standard/68080.html
3. ISO 9001:2015, Quality management systems, Requirements, clauses 8.7 and 10.2, ISO. https://www.iso.org/standard/62085.html
4. ISO 9000:2015, Quality management systems, Fundamentals and vocabulary, ISO. https://www.iso.org/standard/45481.html
5. ASHRAE Guideline 0-2019, The Commissioning Process, ASHRAE, 2019. https://www.ashrae.org/technical-resources/bookstore/commissioning
6. ANSI/ASHRAE/IES Standard 202-2024, Commissioning Process for Buildings and Systems, ASHRAE, 2024. https://www.ashrae.org/technical-resources/bookstore/commissioning
7. ASHRAE, "Principles of Building Commissioning: ASHRAE Guideline 0 and Standard 202", course slides, 2020, accessed 31 October 2026. https://www.utah.gov/pmn/files/1040363.pdf
8. AIA Document G716-2004, Request for Information (RFI), AIA Contract Documents. https://www.aiacontracts.com/
9. USACE, NAVFAC and AFCEC, UFGS 01 33 00 Submittal Procedures, Whole Building Design Guide, accessed 31 October 2026. https://www.wbdg.org/FFC/DOD/UFGS/UFGS%2001%2033%2000.pdf
10. FIDIC, Conditions of Contract for Construction (Red Book), second edition 2017. https://fidic.org/books/construction-contract-2nd-ed-2017-red-book
11. US Occupational Safety and Health Administration, 29 CFR 1910.147 The control of hazardous energy (lockout/tagout), accessed 31 October 2026. https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.147
12. National Fire Protection Association, NFPA 70E Standard for Electrical Safety in the Workplace, 2024 edition. https://www.nfpa.org/codes-and-standards/nfpa-70e-standard-development/70e
13. UK Health and Safety Executive, HSG85 Electricity at work: Safe working practices, third edition, 2013. https://www.hse.gov.uk/pubns/books/hsg85.htm
14. ANSI/NETA ATS-2025, Standard for Acceptance Testing Specifications for Electrical Power Equipment and Systems, InterNational Electrical Testing Association, 2025. https://www.netaworld.org/standards/ansi-neta-ats
15. IEC 62053-22:2020, Electricity metering equipment, Static meters for AC active energy (classes 0.1 S, 0.2 S and 0.5 S), IEC. https://webstore.iec.ch/en/publication/62001
16. Fluke Networks, "Fiber Optic Link Loss Budget Calculation - SimpliFiber Pro", knowledge base, accessed 31 October 2026. https://ru.flukenetworks.com/knowledge-base/simplifiber-pro/loss-budget-calculation-simplifiber-pro
17. Fluke Networks, "Cable Testing 101: There's No Gain with Gainers", accessed 31 October 2026. https://flukenetworks.com/blog/cabling-chronicles/cable-testing-101-theres-no-gain-gainers
18. IEC 61300-3-35:2022, Visual inspection of fibre optic connectors and fibre-stub transceivers, IEC. https://webstore.iec.ch/en/publication/64254
19. ASHRAE Guideline 13-2024, Specifying Building Automation Systems, ASHRAE, 2024. https://webstore.ansi.org/standards/ashrae/ashraeguideline132024
20. ASHRAE TC 9.9, Thermal Guidelines for Data Processing Environments, 5th edition, reference card, ASHRAE, 2021. https://www.ashrae.org/file%20library/technical%20resources/bookstore/supplemental%20files/therm-gdlns-5th-r-e-refcard.pdf
21. Schneider Electric, White Paper 148, Data Center Projects: Commissioning, Rev 1, accessed 31 October 2026. https://download.schneider-electric.com/doc/SPD_DBOY-6NJNK6_EN/DBOY-6NJNK6_R1_EN.pdf
22. NEC, NEC4 Engineering and Construction Contract, 2017 with amendments. https://www.neccontract.com/products/contracts/nec4-ecc

Educational field reference compiled from public standards, guides, manufacturer documents and published incident reports; sources close each chapter. The project specification, the Engineer of Record and the OEM documents govern; check the current edition of every standard before use. Worked examples use an example campus (MSC-1: 4 buildings, 12 halls of 12 MW), not project data. Independent research; not affiliated with any current or former employer. Privacy · Terms.