Disputed fast meter: lab error test and creep test on a domestic static meter
A domestic consumer in Nagpur disputes a bill. His single-phase static meter is marked 230 V, 10-60 A, 2400 imp/kWh, accuracy class 1.0. He says his consumption doubled from about 400 units a month to nearly 800 units after the meter was replaced, although he bought no new appliance. The meter is sent to the divisional test laboratory. The technician sets up a series circuit: supply to a class 0.2 reference standard meter, then to the meter under test, then to a 3 kW resistive load bank, all at 230 V and unity power factor. In the test run the reference meter records 2.500 kWh and the meter under test records 2.575 kWh. In a separate no-load creep test at 230 V with the load bank switched off, the pulse indicator of the meter under test blinks 3 times in 10 minutes. The meter's seals are intact and there is no visible damage or burn mark.
1. Calculate the percentage error of the meter under test and decide whether it passes as a class 1.0 meter. State any limits of this single test.
Read the worked answer
Calculation
Percentage error = (energy recorded by meter under test - energy recorded by reference) / energy recorded by reference x 100
= (2.575 - 2.500) / 2.500 x 100 = 0.075 / 2.500 x 100 = +3.0 percent
The meter is recording 3 percent fast. A class 1.0 meter is permitted roughly plus or minus 1.0 percent error at unity power factor in this load range, and the class 0.2 reference contributes only about 0.2 percent uncertainty, far too small to explain a 3 percent gap. The meter fails and must be replaced or recalibrated.
Limits of this single test
- Only one load point was tested: 3 kW at 230 V is about 13 A, a mid-range current. A proper verification tests several points (roughly 5 percent, 10 percent, rated and maximum current).
- Only unity power factor was tested. A phase-shift load at 0.5 pf lag is needed to check the meter's performance with reactive load, which matters for motor and compressor consumers.
- The reference meter's calibration certificate must be valid, and bench supply voltage and frequency should be stable and recorded.
- A single fast reading should be repeated to exclude connection or timing error before the meter is formally rejected.
- The test proves the meter is fast, but a 3 percent error cannot by itself explain a doubled bill, so the case is not closed here.
2. Quantify what the creep test means in units and explain whether creep can explain the jump from about 400 to about 800 units a month.
Read the worked answer
Converting pulses to energy
Meter constant = 2400 imp/kWh, so one pulse = 1/2400 = 0.000417 kWh.
Three pulses in 10 minutes = 3 x 0.000417 = 0.00125 kWh per 10 minutes.
Per hour: 0.00125 x 6 = 0.0075 kWh.
Per 30-day month: 0.0075 x 24 x 30 = 5.4 kWh per month.
So the creep observed would add roughly 5 units a month, which is around 1 percent of a 400-unit bill and effectively invisible.
Does creep explain the jump? No.
- Creep of this size accounts for about 5 units, not 400.
- The 3 percent fast error found in the load test accounts for only 3 percent of 400 units, about 12 units a month, and only on consumption that actually passes through the meter.
- Together, creep plus the fast error still leave about 380 units unexplained.
Reasoning conclusion: the meter is genuinely defective on both counts (it creeps at no load and it over-registers under load, so it must be replaced), but the consumer's doubling claim must be investigated elsewhere. The technician should not sign off the case as 'meter error explains bill'; the fault is real but too small to account for the bill, and a further site investigation is needed.
3. List the further checks you would make at the consumer's premises and the billing action you would recommend, keeping the meter's confirmed error and the unexplained consumption separate.
Read the worked answer
Keep two questions separate: (a) is the meter within limits? (already answered: no, +3 percent and creeping), and (b) what caused the extra consumption?
Further site checks
- Reading and billing check: confirm the meter serial number on the bill matches the meter, confirm the reading was actually read and not estimated, confirm the multiplying factor is 1 (direct-connected meter), and check for arrears, duty changes or a tariff-category change loaded into the bill.
- Register check: confirm the bill is based on kWh and not mistakenly on a kVAh or MD field.
- Load audit: list appliances and their ratings; a 2 kW geyser used one hour a day is 60 units a month on its own, and a defective immersion heater thermostat or a continuously running pump is a common cause of doubled consumption.
- Leakage and shared-load check: switch off the consumer's main switch and see whether the meter still pulses. If it does, load is either tapped ahead of the main switch, shared with another occupant, or the meter is creep-affected. Insulation resistance test of the wiring (below 1 Mohm indicates leakage to earth, which the meter correctly records).
- Known-load verification on site: switch on a 1.5 kW heater for one hour and confirm the meter advances about 1.5 units (3600 pulses). This is a simple field cross-check of the meter constant.
- Compare history: compare the same month of the previous year and the three months before replacement. If the old meter was under-registering, the 'jump' may be the first correct bill rather than a fault.
Recommended billing action
- Replace the defective meter and give the consumer a copy of the test report, since the error exceeds the permitted limit.
- Revise the bill for the period the faulty meter was in service using the average consumption of the preceding months (typical Maharashtra practice under the Supply Code; confirm the number of months applicable in the current MERC regulations).
- Do not apply a penalty or theft provision: an accuracy error with seals intact is a meter defect, not malpractice.
- Separately advise the consumer in writing of the outcome of the load audit, because the confirmed meter error explains only a small part of the increase.