A method of procedure (MOP) is a step-by-step written script for one planned change in a data centre, such as replacing a UPS battery string. It lists the scope, risks, prerequisites, ordered steps with checks, go/no-go points and a rollback, and it is approved before anyone touches live equipment. Below is a filled example for an invented hall.
One fact about our own offer, so you can judge this fairly. The EDWartens data centre engineer job simulation is a training project at a simulated company, done day by day with project documents, emails and tickets. It is not employment, an internship or a work placement. The first week is free and every ticket is checked as you hand it in. A graded, verifiable certificate is optional, and the longer Premium version is paid; prices show in your own currency before you pay. It shows on your EDWartens profile under Work experience, marked as a job simulation.
SOP, MOP and EOP
Critical facilities run on three kinds of written procedure. They look similar on a page and are used in very different moments.
| Type | Use it for | Trigger | Example |
|---|---|---|---|
| SOP, standard operating procedure | Routine work done the same way each time | A schedule or a routine task | Weekly walk-round and log check |
| MOP, method of procedure | One planned change with specific risks | A change request approved for a window | Replace a battery string on a UPS |
| EOP, emergency operating procedure | Responding to a failure | An alarm or a fault | Load running on battery, generator not started |
Standards bodies and industry groups describe these documents in their guidance on operations and maintenance. The BICSI standards cover data centre operations, and the Uptime Institute explains how redundancy relates to site availability.
When does a change need a MOP?
Use a simple rule: if the work can drop load or reduce redundancy, it needs a MOP. Replacing a battery string, moving a UPS to bypass, switching a feed, working inside a live panel and isolating a chilled-water pump all qualify. Changing a light fitting in an empty aisle does not. The change request sets out why and when, an engineer who did not write the MOP reviews it, and a change approval group agrees the window. Safe working rules come from the site and the law; in Great Britain the HSE electricity at work pages are the starting point.
The invented hall
The example hall is Hall 3 at a made-up site, Northquay Data Centre. All numbers are invented. Power is 2N: an A side and a B side, each able to carry the whole hall. Each side has three UPS modules of 250 kW, so two are needed and one is spare (N+1). Each side has batteries in five strings with 71 kWh of usable energy.
| Item | Capacity | Load | Spare |
|---|---|---|---|
| UPS, one side, N+1 (two of three modules) | 500 kW | 410 kW if one side carries the hall | 90 kW |
| IT load, whole hall | n/a | 410 kW, split about 205 kW per side | n/a |
| Battery energy, one side | 71 kWh usable | n/a | n/a |
| Cooling, N+1 | Five units, four needed | Four running | One unit spare |
| Generator | Two sets, one per side | Standby | Start-up tested monthly |
The sums a reviewer rechecks
Before a MOP is approved, the reviewer reworks the numbers on their own. Four are worth knowing.
Battery runtime. Runtime in minutes = usable kWh x efficiency / load in kW x 60. With 71 kWh and an assumed inverter efficiency of 0.94:
| Case | Usable energy | Load on this side | Runtime |
|---|---|---|---|
| Normal, both sides up, five strings | 71 kWh | 205 kW | 19.5 minutes |
| Other side lost, five strings | 71 kWh | 410 kW | 9.8 minutes |
| Normal, one string out (four strings) | 56.8 kWh | 205 kW | 15.6 minutes |
| Other side lost, one string out | 56.8 kWh | 410 kW | 7.8 minutes |
The working for the third row: 56.8 x 0.94 / 205 x 60 = 15.6 minutes. This is a straight-line estimate. Real discharge curves are steeper at high loads, so the manufacturer's data decides the real figure.
Rack power. A 16 A three-phase 400 V feed gives 1.732 x 400 x 16 = 11.09 kVA. Design to 80 percent of that, which is 8.87 kVA per feed, so that a breaker does not run at its limit.
PUE. Suppose the hall uses 410 kW of IT load, 118 kW for cooling, 24 kW lost in UPS and distribution and 9 kW for lighting and other loads. The total is 561 kW, and PUE = 561 / 410 = 1.37. The measurement boundary and period must be stated, as in ISO/IEC 30134-2.
Margin of 2N. With one side down, the other carries 410 kW out of its 500 kW N+1 capacity. A change that reduces one side to two healthy modules leaves it exactly at N, with no spare module. That is allowed only if the MOP says so and the approval records it.
A filled MOP
The change is to replace battery string 3 on UPS A of Hall 3. These sections come first in the document.
| Section | Content |
|---|---|
| Title and number | MOP-0310 Rev A: Replace battery string 3, UPS A, Hall 3 |
| Purpose | Replace one battery string flagged by monitoring for high internal resistance |
| Scope | UPS A battery cabinet only. No work on the B side, the hall or the UPS modules |
| Risk | Runtime on the A side falls from 19.5 to 15.6 minutes while one string is out; hall stays on 2N throughout |
| Prerequisites | Change approved; new string delivered and inspected; B side healthy; generators tested this month; fuel above half; insulated tools checked |
| Window | Tuesday 02:00 to 05:00, a low-demand period agreed with customers |
| Roles | Change lead (calls go/no-go and stop), UPS technician (does the work), observer (reads the steps), control room (logs and alerts) |
| Stop rule | Anyone may call stop. Only the change lead may restart |
The step table is the heart of the document. Every step is one action, with a result the observer can see and a place to initial.
| Step | Action | Expected result | Initial |
|---|---|---|---|
| 1 | Hold the briefing; read the MOP aloud; confirm roles | Every role confirmed | ___ |
| 2 | Check both sides for alarms; confirm load sharing near 205 kW each | No alarms; both sides near 205 kW | ___ |
| 3 | Confirm generator ready and fuel above half | Both sets in auto, fuel confirmed | ___ |
| 4 | Read float voltage on all five strings | All within the vendor window; no alarms | ___ |
| GO/NO-GO 1 | Calculate runtime with four strings; minimum is 15 minutes | 15.6 minutes or more; B side healthy | ___ |
| 5 | Tell the control room and the customer duty contact; log the start time | Start time logged | ___ |
| 6 | Open the DC breaker labelled S3, and no other | UPS shows four strings; only “string 3 open” alarm | ___ |
| 7 | Prove the voltage tester on a known source, test S3 terminals, prove again | 0 V at S3; tester proved before and after | ___ |
| 8 | Apply lock and tag to the S3 breaker | Lock and tag fitted | ___ |
| 9 | Replace the string 3 modules to the vendor method; record torque values | New modules fitted; torque recorded | ___ |
| 10 | Measure the new string's open-circuit voltage before connecting | Within the window of the other strings | ___ |
| GO/NO-GO 2 | Is the new string within the window and the B side still healthy? | Yes to both, or go to rollback | ___ |
| 11 | Remove lock and tag; close the S3 breaker with the observer watching | UPS shows five strings; small charge current | ___ |
| 12 | Final verification: alarms, load sharing, runtime estimate, string readings | No alarms; both sides near 205 kW; runtime 19.5 minutes | ___ |
| 13 | Tell the control room; update the asset and capacity registers | Registers updated; end time logged | ___ |
| 14 | Sign off | Change lead and observer signatures | ___ |
Rollback. Before step 9: close the S3 breaker per step 11, confirm five strings and stop. After step 9, if the new string fails GO/NO-GO 2: leave S3 open and locked, record the four-string state and the 15.6 minute runtime, raise a defect and end the window. At any time, if load drops or an unexpected alarm appears on either side: stop, make the equipment safe, and switch to the emergency operating procedure for loss of redundancy.
A 14-question MOP review checklist
- Is the document numbered, versioned and approved, with the reviewer not the author?
- Is the scope stated, including what is not in scope?
- Is every step a single action?
- Does every step have an expected result and a place to record it?
- Is the redundancy state stated at each stage?
- Are the runtime, load and capacity sums shown and correct?
- Are there clear go/no-go points with a criterion, not a feeling?
- Does the rollback work from every point, not only from the start?
- Is the stop rule clear, and does anyone have the right to use it?
- Are roles named, with a person for each role on the day?
- Are customers and the control room told before and after?
- Are permits, isolation and lock and tag steps included?
- Are tools, spares and test instruments listed and checked?
- Is there a step to update the registers and close the change?
When the procedure meets reality
The procedure is never the whole story. Three things commonly happen, and the document should help with each.
A bad reading at the start. At step 4, string 5 shows a low float voltage. The runtime calculation was done with five healthy strings, so with one out and one weak, the real figure is below 15.6 minutes. That is a no-go. The change lead stops, logs the reading and re-plans. A MOP is a plan until the data agrees with it.
A request for extra work. A tenant's engineer asks if the team can also check a tag on a different cabinet while they are there. The answer is no. The scope is fixed, and extra work in a live window is how a planned change becomes an incident. Log it as a new request.
Honest reporting. If something goes wrong, the report says what happened, in order, with times, and does not guess at blame. A short sample: “02:41 string 3 opened. 02:43 voltage proved 0 V. 03:10 new string measured within window. 03:14 breaker closed. 03:25 final verification complete, no alarms.” A clear, boring log protects everyone.
Shift handover: a short sample
| Item | Status | Next shift |
|---|---|---|
| MOP-0310 battery string 3, UPS A | Complete 03:30, five strings, no alarms | Review readings at 12:00 |
| Cooling unit 4 fan alarm | Open, spare unit running | Vendor visit booked |
| Generator B fuel level | Above half | Top-up delivery Thursday |
| Capacity register | Updated after MOP-0310 | No action |
Practise it
The best way to learn a MOP is to review one and run one under time pressure. The data centre engineer job simulation gives you a hall, procedures to review, a capacity request and a shift handover at a simulated company, and each ticket is checked as you hand it in. The maintenance electrical engineer simulation, the substation and motor control centre engineer simulation and the project coordinator simulation cover the skills around it, and all sit on the job simulation hub.
In the lab, the data centre simulator lets you explore a hall, and the multimeter simulator lets you practise reading an instrument. For background, read our guide to data centre commissioning levels L1 to L5 and our safe isolation procedure. The free course on electrical fundamentals for technicians builds the basics.
A finished simulation shows on your EDWartens profile under Work experience, marked as a job simulation, and the optional certificate has a public page where anyone can verify it. The commissioning internship track is a free guided four-week option, and the jobs board lists current openings.
Last updated 19 October 2026. The site, hall and figures in this post are invented. Always follow your own site's procedures, vendor instructions and safe working rules.
Practise with us
Internship
Electrical Design
Free internship: 4 weeks, a free certificate and nothing to buy. Premium adds mentor calls and the graded certificate.
See the internshipRoles
Electrical design
Roles that ask for this skill, 72 open in this country now. Free to see and to apply. We are not a recruitment agency.
See rolesTake the free course
Questions
What does MOP mean in a data centre?
MOP means method of procedure. It is the written, step-by-step script for one specific planned change, such as moving a UPS to maintenance bypass or replacing a battery string. It is reviewed and approved before anyone touches live equipment, and followed exactly on the day.
What is the difference between a MOP, an SOP and an EOP?
An SOP, standard operating procedure, covers routine work done the same way every time. A MOP covers a specific planned change with its own risks and approvals. An EOP, emergency operating procedure, covers what to do when something goes wrong, such as the load running on battery. Each has a different trigger and a different reader.
Who writes a MOP?
It is often written by the vendor or the facilities engineer who knows the equipment. The site team then reviews it against the actual hall, checks redundancy and approves it through change control before the window. The person who runs it should have read it more than once beforehand.
What makes a good MOP?
One action per step, an expected result and a check for every step, clear go/no-go points, a stated redundancy state at each stage, named roles, and a rollback that works from any point. A step that says “check everything is fine” is not a step.
How do I calculate UPS battery runtime?
Multiply the usable battery energy in kilowatt-hours by the inverter efficiency, divide by the load in kilowatts and multiply by 60 to get minutes. That is a simple linear estimate. For a real decision, check the manufacturer's discharge curve at the actual load, because batteries give less at high rates.
What is PUE and how is it calculated?
Power usage effectiveness is the total energy used by the facility divided by the energy used by the IT equipment. A figure of 1.0 would mean all energy reaches the IT load, which no site reaches. Lower is better, and the measurement boundary and period must be stated or the figure means little.
What qualifications do I need for a critical facilities role?
Usually an electrical or mechanical background, authorisation to work safely on the equipment, and a site induction. Industry bodies such as BICSI and Uptime Institute offer courses and credentials, and employers differ on what they ask for. Never claim a credential you have not earned, and a simulation is not one.
Can I put a data centre simulation on my CV?
Yes, as practice work with a clear description, such as “data centre engineer job simulation (simulated company)”. It is training, not employment, so list it under projects or training and be ready to explain a decision you made in it.
Sources
- Uptime Institute, tier classification system (checked 19 October 2026)
- BICSI, standards including ANSI/BICSI 002 data centre design and implementation (checked 19 October 2026)
- ASHRAE, Datacom series on data centre thermal guidelines (checked 19 October 2026)
- ISO, ISO/IEC 30134-2 key performance indicators, power usage effectiveness (October 2026)
- HSE, Health and Safety Executive, electricity at work (checked 19 October 2026)
Written by the EDWartens engineering team for general education. Product names are trademarks of their owners; mentioning them does not imply endorsement. Prices and terms of other providers were checked on the date shown and can change.
Are Job Simulations Worth It? Certificates, CV, Employers
Data Analyst Portfolio Projects With Messy Data
Free Job Simulations for Engineers: The Honest List
RAID Log Example for Engineering Projects
Solar PV Commissioning Checklist: IEC 62446-1 Tests
Electrical Engineering Work Experience Without an Internship