Protection
A thermal replica relay, ANSI device 49R, models the heating of a motor, cable or transformer from the current it carries. Testing injects overload currents, times the trip along the curve from cold and hot states, and checks the alarm, the reset and the output.
A device 49R relay does not measure temperature. It calculates it. From the current flowing in the protected equipment, the relay runs a mathematical model of the heating, the thermal replica, that tracks how much of the equipment's thermal capacity has been used. When the model reaches its limit, the relay trips, and an earlier alarm level may warn operators. The element is standard in motor protection relays and appears on transformers, cables and capacitor banks. Settings reflect the equipment: full-load current, service factor, the overload curve class, separate treatment of starting and running, and the heating and cooling time constants. The field test is secondary injection of current from a test set, with a timer on the output contact and attention to the thermal state the relay holds before each test.
Motors and cables tolerate overload for a while, and the time allowed falls as the overload grows. A fuse or instantaneous element cannot follow that tradeoff, and a simple timer ignores the heat already stored from earlier running. A thermal replica remembers: a motor that was just running hot has less margin than a cold one, and the relay accounts for that. A relay that trips too soon causes needless stoppages, especially in processes that hit repeated starts. One that trips too late allows insulation to age or fail, and the damage may surface months later. Since the model is internal and its constants come from settings, only injection can prove that the curve, the memory and the cooling behave as the study assumed.
Heat generated in a conductor grows with the square of the current, and the equipment loses heat to its surroundings at a rate set by its thermal time constant. A thermal replica uses the same idea in calculation: each step adds heat in proportion to the square of the current and removes heat in proportion to the difference from the surroundings, so a steady overload drives the model toward a limit along a curve that depends on the size of the overload. The result is an inverse-time characteristic, where a heavy overload trips quickly and a mild one takes much longer. The model has a starting value that depends on prior history, so the same overload trips faster from a hot state than from a cold one. Many relays also model unbalance, which adds heating, and an alarm level on the percentage of thermal capacity used. After a trip the model cools, and a restart may be blocked until it has.
With the relay isolated and its trip output blocked, the technician records the identity and the thermal settings: full-load current, service factor, curve class, time constants and alarm level. The thermal memory is cleared, or the relay is allowed to settle to a known cold state, and a three-phase current test set is connected to the current inputs with the output contact timed. A chosen overload is applied from the cold state, and the time to trip is measured and compared with the curve, at more than one multiple so the shape is verified. The test is then repeated from a hot state, produced by running the model up through a prior load and then reapplying the overload, to confirm that thermal memory shortens the time. The alarm level is observed as the model fills. After the trip, the cooling and the restart inhibit are checked. Because thermal memory carries over between runs, a reset is made before each test, and the order of tests is planned.
The record gives the relay identity and thermal settings, the overload multiples applied, the thermal state at the start of each test and the trip time measured for each. It notes the alarm response, the cooling and reset behavior and whether the memory was cleared before each run. It names the test set and its calibration status and the as-found and as-left settings, and records any adjustment.
Results are compared with the curve produced by the motor or equipment study and with the manufacturer's published curve, since the right settings depend on the equipment thermal limits. The shapes of the criteria are a trip time that follows the published curve at each level tested inside the stated accuracy, a hot-state time that is shorter than the cold-state time in the way the model predicts, an alarm that precedes the trip, a restart inhibit that releases as the model cools, and a clean output contact. Maintenance results are also trended against earlier tests. The scope of relay testing is described in ANSI/NETA ATS-2025 for acceptance and ANSI/NETA MTS-2023 for maintenance. For tolerances and clause detail, consult your own licensed copy of those standards and the manufacturer's data.
The equipment classes and standard sections that call for this test, and the UltraDb forms that carry it.
It is a relay that calculates equipment heating from the measured current instead of reading a temperature. The calculation includes stored heat and cooling, so it protects a motor, cable or transformer against sustained overload.
Heat from earlier load is already stored in the model when the machine is hot, so less additional heating is needed to reach the limit. A cold start has more margin, so the same overload takes longer to trip.
Each injection adds heat to the model, which would shorten the next result. Clearing the memory, or allowing a known cooldown, gives every test a defined starting state so that results can be compared.
No. It estimates temperature from current and cannot see ambient conditions, blocked cooling or a hot spot in the winding. A temperature relay, device 49T, measures actual temperature and complements it.
UltraDb references ASTM, ANSI, IEEE, NETA, NFPA, NEMA, FAA, TIA, and IEC standards descriptively, to identify the published methods its forms are built from. UltraDb and Verlecta are not affiliated with, endorsed by, certified by, or licensed by NETA, ASTM, ANSI, IEEE, NFPA, NEMA, FAA, TIA, IEC, or Megger. Standard names and section numbers are the property of their respective organizations. Always test to the edition your contract specifies. Where a standard is sold through the ANSI webstore, Verlecta may earn a commission on a purchase made through our link, at no extra cost to you. A commission never decides which standards we cite or how a form is built.