How Should Buyers Specify a Custom Three-Phase Online UPS Before Production? 

Direct answer：A custom uninterruptible power supply RFQ should freeze the load schedule, required kW and kVA, input and output voltage, frequency, phase and neutral arrangement, bypass source, battery autonomy, environmental conditions, communication, protection, installation constraints, drawings, and acceptance tests. Buyers should define external performance and interfaces, while the supplier returns a compliant design, calculations, deviations, and controlled production documents.

What does “custom” need to mean in the contract?

Customization can refer to ratings, cabinet layout, input and output arrangement, battery configuration, protection, communication, monitoring, cable entry, color, labels, documentation, or factory tests. It should not remain an undefined promise. Each requested change needs a measurable requirement, an owner, and acceptance evidence.

The target industrial-frequency online three-phase UPS is listed for 10–600kVA, with three-phase input and three-phase output. This range identifies a product family, not the final project configuration. The RFQ must still establish the selected rating, actual load, voltage system, battery duty, bypass, environment, interfaces, and installation boundary.

Buyers should separate three document levels. The employer’s requirements define the business and site need. The supplier proposal identifies the offered design and deviations. Approved-for-production documents freeze what will be built. Confusing these levels creates disputes when an early budget sketch is treated as a final manufacturing drawing.

Which load data determines the UPS rating?

List every protected load with quantity, operating watts, apparent power where available, power factor, phase allocation, starting or inrush behavior, nonlinear characteristics, duty cycle, and criticality. Mark loads that operate simultaneously and loads that may be shed. State current load and planned expansion separately.

A custom UPS should not be selected from the sum of equipment nameplate kVA alone. Nameplates may overstate normal demand, while motor starts, transformers, imaging equipment, drives, or rectifier loads can introduce short-duration peaks and harmonic effects. Ask the supplier to show how the selected rating addresses steady-state load, transient demand, overload, and future margin.

Define the design margin as a project decision. Excessive oversizing can increase capital cost, footprint, battery quantity, and low-load operation. Insufficient margin can restrict expansion or create overload transfers. The offer should show both operating kW and kVA, the assumed load power factor, and the percentage loading for present and future cases.

Load schedule fields that prevent ambiguity

• Load name, quantity, kW, kVA, power factor, and phase connection
• Normal, start-up, regenerative, pulsed, or cyclic behavior
• Simultaneous operating group and load-shedding priority
• Required ride-through or battery autonomy for each critical group
• Present demand, committed expansion, and optional future demand
• Maximum acceptable interruption or transfer condition

How should input, generator, and bypass conditions be defined?

State nominal input voltage and frequency, permitted variation, phase sequence, neutral and protective-earth arrangement, upstream transformer data, available fault level, upstream protective device, and cable entry. If a generator supplies the UPS, provide generator rating, voltage regulation, frequency behavior, ramp or loading sequence, and other large loads sharing the source.

The bypass source requires equal attention. Specify whether the rectifier and bypass use a common or separate source, whether a neutral is available, the acceptable bypass voltage and frequency window, and the required behavior when bypass is outside limits. Identify whether an external maintenance bypass is in the supplier’s scope and how isolation will be achieved.

Do not copy electrical values from a reference design without checking the site. Three-phase UPS products from different manufacturers use different voltage windows, connections, bypass arrangements, short-circuit capabilities, and overload conditions. The bidder should return model-specific data for the proposed configuration rather than a generic industry table.

What output performance and load protection must be frozen?

Define nominal output voltage and frequency, phase and neutral arrangement, voltage regulation expectation, permitted waveform distortion for the specified load type, frequency behavior, overload duty, and transfer behavior. State whether galvanic isolation, an output transformer, or a particular earthing arrangement is required by the site design. Do not assume that “industrial frequency” alone answers those questions.

Short-circuit and protection coordination must be treated as a system issue. Provide downstream breaker or fuse information and the fault-clearing objective. Ask the supplier for the UPS output fault behavior and coordination data needed by the electrical designer. A UPS can limit current differently from a transformer-fed source, so downstream protection should not be selected from normal current alone.

For loads that cannot tolerate an interruption, define the allowed behavior during inverter overload, internal fault, bypass transfer, battery depletion, emergency power-off, and maintenance. The cause-and-effect schedule should show which alarms warn the operator before a shutdown and which conditions initiate an automatic transfer.

How should battery autonomy and recharge be specified?

State the required runtime at a defined load, end condition, ambient temperature, and battery age or design margin. Clarify whether runtime applies to the present load, future load, or a load-shed profile. Define whether batteries are internal or external, the available battery-room space, monitoring, isolation, and who supplies racks, cabinets, DC protection, and interconnecting cables.

Recharge performance can be as important as initial autonomy. State the maximum acceptable recharge time, available input capacity, and whether repeated outages are credible. Ask the supplier to identify charging current, assumptions, and any impact on upstream input demand. If the system must start from battery without mains, include that function and its test in the specification.

Do not reuse a battery calculation from a different UPS model or DC bus. The supplier should issue a calculation for the offered inverter efficiency, DC voltage, cell or battery type, number of units, discharge data, temperature, and aging basis. The wider USYTU UPS product category can help buyers identify available architectures, but each project calculation remains configuration-specific.

Which interfaces, controls, and site constraints belong in the RFQ?

Provide the required local display language, alarm list, dry contacts, emergency power-off, remote monitoring, communication protocol, network connection, time synchronization, event retention, and integration with a building or data-center management system. Define which signals are mandatory and who maps, tests, and documents them.

Cybersecurity requirements should be stated before a network card or remote-access feature is accepted. Identify permitted protocols, addressing responsibility, password policy, remote-access restrictions, and the party that approves connection to the plant network. A communication option is not complete until the point list and test method are agreed.

Send dimensioned site information: room size, door and corridor clearances, floor loading, ventilation or cooling conditions, altitude, ambient temperature, humidity, dust exposure, seismic or vibration requirements if applicable, front and rear service clearance, and top or bottom cable entry. A technically adequate UPS may still be unsuitable if it cannot be transported, positioned, ventilated, or maintained in the available space.

What drawings should the buyer provide and receive?

The buyer normally provides a site single-line diagram, load schedule, source and bypass data, room layout, cable-entry preference, control philosophy, communication requirements, and applicable project standards. This is the external design basis. Unless the procurement contract explicitly assigns it otherwise, the buyer should not need to create the manufacturer’s internal power-circuit design.

The supplier should return a proposed single-line diagram, general arrangement, terminal plan, cable-entry drawing, interface schedule, protection information, battery arrangement, heat-loss and ventilation data, foundation or floor-loading data, and bill of materials at the level agreed for approval. Documents should carry revision numbers and approval status.

An uninterruptible power supply drawing found in an old tender can be used as a reference, but it should not silently override the current load and site data. Mark reference documents as “for concept,” “for quotation,” or “for production approval.” If two documents conflict, the contract should state which has priority and how the discrepancy is resolved.

How should customization be controlled before production?

Create a compliance matrix with one row per requirement. The supplier should answer compliant, deviation, optional, or not included and reference the relevant document. A quotation that says “can be customized” without a marked response does not establish the final scope.

After technical clarification, issue a frozen requirement list and approved document register. Changes should identify the reason, affected drawings, price, delivery impact, retest requirement, and approval authority. Verbal requests made during a meeting should not reach production until they appear in controlled documents.

Separate mandatory changes from preferences. A different input voltage, fault rating, battery duty, or bypass architecture can affect the electrical design. Cabinet color or label language may affect documentation and production but not power performance. Treating all changes as equal makes review slow and hides the items that need engineering verification.

Where the project can use a standard configuration, compare it with the proposed custom design. A nearby 10–80kVA high-frequency three-phase online UPS represents a different product family that may be relevant to some ratings, but it should not be treated as technically interchangeable. Ask the supplier to explain differences in topology, footprint, input and output behavior, battery arrangement, service method, and project suitability.

Which FAT and handover evidence should be agreed?

The factory acceptance test should be written against the approved configuration. Define inspection of model and nameplate, visual workmanship, wiring and terminals, start-up, normal operation, battery operation, bypass transfer, alarm and communication points, emergency functions, output checks, load test conditions, and document review. State which tests are witnessed and which are supported by routine records.

Performance values and test durations should come from the approved specification, not be invented at the factory. If full-load testing is not available, the supplier should state the alternative method and the buyer should approve it before the test. Record instrument identification, measured values, software version, settings, deviations, and corrective actions.

Site acceptance should verify installation-specific items that the factory cannot prove: source and bypass connections, phase sequence, earthing, ventilation, cable terminations, upstream and downstream protection, generator interaction, plant communication, load transfer, battery arrangement, and operator procedures. Handover should include approved drawings, settings, test reports, manuals, spare-parts list, maintenance schedule, and training records.

Price comparison should use the same scope. The article on why same-kVA UPS projects have different costs highlights the importance of comparing batteries, bypass, installation, testing, and services rather than the cabinet rating alone.

A representative custom UPS project

Representative scenario — not a claimed customer case.

An industrial plant requests a custom 200kVA three-phase online UPS using an old single-line diagram. The drawing shows no separate bypass data, the load list mixes present and future motors, battery runtime is stated without a load level, and the room drawing omits cable-entry and service clearances. Three bidders return different scopes, yet each claims compliance.

The buyer issues a revised load schedule, source and bypass table, room layout, runtime basis, signal list, and document register. Each supplier marks deviations and returns a rating calculation, proposed single line, general arrangement, battery calculation, and draft FAT. The offers can then be compared by verified scope, and production begins only after the agreed documents are approved.

FAQ

What is the first document for a custom UPS RFQ?

Start with a load schedule linked to a site single-line diagram. The load schedule establishes kW, kVA, power factor, transient behavior, duty, criticality, and expansion. The single line establishes sources, bypass, earthing, phase arrangement, protection boundary, and distribution connections.

Should buyers provide an internal UPS circuit design?

Usually, buyers define required performance, external interfaces, site constraints, and applicable standards. The manufacturer develops and controls the internal design for the offered product. If the contract requires buyer approval of additional design detail, state the document level and approval responsibility explicitly.

Why are both kW and kVA required?

kW represents real power and kVA represents apparent power. UPS loading depends on both, together with load power factor and transient behavior. Specifying only one can result in an offer that appears correctly rated but does not match the actual load profile.

Is a 10–600kVA product range one standard configuration?

No. It identifies a family covering multiple ratings. Cabinet arrangement, battery system, protection, cables, footprint, bypass, and project engineering can vary by selected capacity and option. Request model-specific drawings, data, and test scope for the quoted rating.

When should production be released?

Release production after the compliance matrix, final scope, approved single line, general arrangement, battery basis, interfaces, document register, and test requirements are frozen. Any open item that can change electrical design, cabinet construction, or acceptance should be resolved or formally controlled.

What is the difference between FAT and SAT?

FAT verifies the approved product configuration at the factory using agreed methods. SAT verifies installation-dependent performance at site, including source, bypass, cables, earthing, ventilation, protection, communication, load interaction, and operating procedures. The two tests should have separate acceptance criteria.

How should buyers compare custom UPS quotations?

Use a common compliance matrix and normalize rating, batteries, bypass, protection, communications, drawings, tests, commissioning, training, spares, warranty, and exclusions. A lower cabinet price does not indicate a lower complete project cost when scope differs.

What should buyers send for engineering review?

Send USYTU the load schedule, single-line diagram, input and bypass data, output requirements, runtime basis, battery-room details, installation layout, environment, communication list, protection information, required standards, document list, FAT criteria, delivery destination, and project schedule. Use the USYTU contact page to request a marked compliance and deviation response.

Buyers can also review the broader USYTU power and data-center product range when the project includes batteries, micro-module infrastructure, or related backup-power equipment beyond the UPS cabinet.