The 100:1 Problem: Why Relay Test Set VA Specifications Are Outdated for Numerical Relays
Published: June 2026
Author: Sunil—Founder, Myriad Industrial Solutions LLP
Most Indian relay test set procurement specifications still require 250 to 450 VA of current source output per phase. A numerical protection relay—the type installed in over 90% of new Indian substations since approximately 2010—has a measured current input burden of less than 1 VA at rated current. This article examines a specification habit that has gone unchallenged in Indian technical literature for two decades and what it actually costs utilities, EPC contractors, and testing organizations that continue to procure against it.
The study is presented as an impartial technical analysis. No specific manufacturer or product is referenced. The argument rests on publicly available specifications and established electrical principles, and it invites corrections from practicing protection engineers.
What VA Burden Does a Numerical Relay Actually Need?
A numerical relay performs its protection function through analogue-to-digital signal conversion and digital processing—not through electromechanical force generation. Its current input circuit typically presents a burden of 0.02 VA to 0.5 VA at rated current, with input impedance below 0.01 Ω in most modern designs.
This is fundamentally different from an electromechanical relay, where the coil must absorb real power to generate the mechanical torque that operates the relay. Electromechanical relay burdens of 3VA to 25VA at rated current, with coil impedances of 0.1Ω to 2Ω, were the physical basis for why high-output test sets were originally necessary.
To inject 50A into a numerical relay with 0.006Ω input impedance, the required terminal voltage is 0.30V, and the required VA is 15 VA. The same 50A into an electromechanical relay coil with 1Ω impedance requires 50V and 2,500VA. The difference is not incremental—it is approximately two orders of magnitude.
Why Do Indian Relay Test Set Specifications Still Require 250–450 VA?
India's relay population underwent a near-complete transition from electromechanical and static technology to numerical technology between approximately 2000 and 2015, driven primarily by new substation construction during a period when numerical relays had become the cost-competitive default choice for Indian manufacturers and importers.
Despite this technology shift, procurement specifications for relay test equipment have largely remained unchanged across multiple procurement cycles. This is a function of institutional procurement structure rather than technical requirement:
- The engineer who writes a procurement specification is typically in a materials or contracts function, separate from the protection engineering function that actually uses the equipment.
- Revising an inherited specification carries personal accountability risk for the procurement engineer; retaining it does not.
- Equipment suppliers have no commercial incentive to challenge a specification that supports higher unit pricing.
- Type-test approval processes, once obtained against an existing specification, create inertia against product redesign.
- No Indian standards body has published guidance distinguishing VA requirements for numerical relay testing from those for electromechanical relay testing.
The result is a specification that reflects the relay technology in service when it was first written, carried forward through successive procurement cycles independent of how the underlying relay population has changed.
What Does an Actual Indian Utility Specification Require?
MSEDCL—Maharashtra's principal electricity distribution utility—issued a technical specification for a relay testing kit in March 2020 (reference CE/T-QC/MSC-II/RTK). This is a current, publicly referenced procurement document, not a historical artifact.
The specification mandates a current source output of 2×35A at 250VA per source, giving an open-circuit voltage of approximately 7.1V per channel. It explicitly requires the equipment to be capable of testing electromagnetic, static, and numerical relays as a single combined requirement.
How Much Excess Compliance Voltage Does This Create for Numerical Relay Testing?
Applying the MSEDCL specification to a representative numerical relay with an input impedance of approximately 0.006 Ω at a test current of 35 A yields:
Required terminal voltage: 35A × 0.006Ω = 0.21V
Available terminal voltage from specified source: 7.14 V
Ratio of available to required compliance voltage: approximately 34:1
At 10 times rated current on a 5A relay (50A injection), the ratio narrows slightly to approximately 24:1 but remains substantial. The specified test source provides 24 to 34 times more compliance voltage than numerical relay testing requires. This excess capability is unused during the test, does not improve measurement accuracy, and has no measurable influence on test outcomes.
Does Higher Output Power Help Diagnose CT or Wiring Faults?
A common counterargument holds that aged or unverified secondary wiring may present elevated impedance—from partial short circuits, loose connections, or insulation degradation—and that higher VA output allows the test set to push through these anomalies.
This argument has a genuine but narrow application. Where secondary circuit integrity cannot be verified in advance—particularly in aged installations—higher compliance voltage does provide a margin against unexpected circuit impedance. This scenario represents a legitimate boundary condition for procurement decisions involving older infrastructure with unverified wiring history.
However, this procedure is a circuit verification function, not a relay testing function. The appropriate response to suspected wiring anomalies is dedicated circuit verification—using loop resistance testers or CT burden meters—followed by rectification before relay testing. Using a high-VA relay test set as a substitute for proper circuit verification is not standard or advisable commissioning practice, and the diagnostic value of a relay test set in this scenario comes from terminal voltage monitoring during injection—a function of measurement architecture, not of output power capacity.
Were Differential Relays Tested Successfully Before High-Power, Multi-Phase Test Sets Existed?
Two and three-winding transformer differential relays have been in service since the 1940s, decades before multi-phase secondary injection test sets existed. Historical commissioning practice for these relays relied on:
- Sequential single-phase injection into each winding input individually, with the restraint characteristic verified point-by-point through calculation rather than simultaneous multi-phase injection
- Primary injection testing, where current is applied at the transformer's primary terminals, testing the complete protection chain—transformer, CTs, wiring, and relay—through actual transformer action rather than secondary simulation
- High-set unrestrained elements verified at a single confirmatory test point rather than as a continuous characteristic curve
These methods were the accepted commissioning standard for approximately fifty years and produced a power transmission and distribution infrastructure that did not experience systematic protection failures attributable to test methodology. Multi-phase automated injection, introduced later, improved testing speed and documentation consistency. It merely duplicated a capability that had long been available.
This history is relevant because the most demanding electrical argument for high VA output — testing transformer differential relays at high current multiples — turns out, on closer analysis, to require a high current range rather than high VA. A numerical relay's input impedance remains low even at high test current, so the compliance voltage and VA demand stay modest. The binding constraint for differential relay high-set testing is the test set's maximum current output, not its volt-ampere rating—two distinct electrical parameters that procurement specifications have historically conflated.
What Should Utilities and Procurement Engineers Consider?
This analysis suggests three possible positions, each defensible under different conditions:
Where a substation's protection panels include a known and active electromechanical relay population, current high-VA specifications remain technically justified, and procurement should continue to reflect that requirement.
Where procurement covers new substation commissioning or maintenance of fleets confirmed to be entirely numerical—increasingly the case across Indian transmission and distribution infrastructure—specifications calibrated to actual numerical relay burden would be technically sufficient and would materially reduce unit cost, weight, and procurement price without any corresponding loss of test validity.
Where a single fleet contains a mix of relay generations, specification by application—rather than a single universal high-VA requirement applied across all procurement—would allow utilities to deploy appropriately sized equipment for each context, potentially increasing the total number of test units a given budget can support and, correspondingly, increasing testing frequency and protection system maintenance coverage across the fleet.
This article does not advocate a single position. It sets out the technical basis on which Indian utilities, testing organizations, and procurement professionals can make that determination explicitly, rather than by default.
Frequently Asked Questions
Does a numerical relay need a high-VA test set?
No. A numerical relay's current input usually has a burden of less than 1 VA at rated current, which is much lower than the 250 to 450 VA often specified in current Indian relay test set procurement documents.
What VA rating does the MSEDCL relay testing kit specification require?
The MSEDCL 2020 specification (reference CE/T-QC/MSC-II/RTK) requires 250 VA per current source at 35 A, giving an open-circuit voltage of approximately 7.1 V.
Why do Indian relay test set tenders still specify high VA output if numerical relays don't need it?
Primarily due to specification inertia, procurement documents are typically copied from earlier versions written when electromechanical relays were more prevalent, and no Indian standards body has published updated guidance distinguishing numerical relay requirements from electromechanical relay requirements.
Does higher VA output improve relay test accuracy?
No. Test accuracy is determined by signal quality parameters—current and voltage accuracy, phase angle resolution, and waveform distortion—which are independent of the output power rating. A correctly calibrated low-VA test set produces identical test results to a high-VA test set when testing a numerical relay within its required current range.
Were transformer differential relays tested before high-power multi-phase test sets existed?
Yes. Differential relays have been commissioned since the 1940s using sequential single-phase injection and primary injection methods, decades before multi-phase secondary injection equipment became available.
This article is offered for technical review and welcomes corrections from practicing protection and distribution engineers. This analysis does not use any proprietary data or product-specific claims.
Further reading
If you are evaluating a hardware upgrade or designing procurement specifications for digital substations, read our comprehensive follow-up guide: IEC 61850 Relay Test Set: How to Choose the Right Specifications and Features. It details a practical, application-based framework for balancing analog amplifier capacities with virtualized GOOSE and sampled values testing.
