Engineering Guide
Open-loop vs closed-loop Hall current sensors: accuracy, speed, power and cost
How each Hall sensor architecture works, a side-by-side comparison of accuracy, response time, drift, power and size, and a simple rule for choosing between them.
Open-loop and closed-loop are the two basic ways to build a Hall-effect current sensor. Both measure DC, AC and pulsed current with galvanic isolation, and both appear in EV drives, battery systems, chargers and industrial power. They differ in how they turn the magnetic field into a signal, and that difference decides accuracy, speed, power consumption and cost.
How an open-loop Hall sensor works
The conductor passes through a magnetic core with a small air gap. A Hall element sits in the gap. The current in the conductor creates a magnetic flux in the core, the Hall element produces a voltage proportional to that flux, and an amplifier turns it into the output signal.
It is simple and direct, but the output depends on everything in that chain: the linearity of the core, the sensitivity of the Hall element and how both change with temperature. Large currents can also drive the core towards saturation, which limits linearity at the top of the range.
Strengths: compact, low power consumption, low cost, wide current range. Limits: accuracy and drift are set by the core and Hall element; response is good but not the fastest.
A variant is the programmable Hall sensor, which uses an integrated Hall IC calibrated electronically at the factory. It keeps the open-loop principle but improves consistency and runs from a low supply voltage.
How a closed-loop Hall sensor works
A closed-loop sensor adds a secondary (compensation) winding on the core. The Hall element no longer measures the field directly. It acts as a null detector. An amplifier drives a current through the secondary winding that exactly cancels the flux created by the primary conductor, so the core is held at zero flux. That is why closed-loop sensors are also called zero-flux or magnetic balance sensors.
The compensation current is proportional to the primary current, divided by the number of secondary turns. It becomes the output, either directly as a current or across a measuring resistor as a voltage.
Because the core always works around zero flux, its non-linearity and saturation hardly matter, and the Hall element’s drift only affects the small error signal. The result is better accuracy, better linearity, lower drift and very fast response.
Strengths: high accuracy and linearity, low temperature drift, very fast response, wide bandwidth. Limits: the compensation current comes from the supply, so power consumption rises with measured current; larger and more expensive at high current ratings.
Side-by-side comparison
| Criterion | Open-loop Hall | Closed-loop Hall |
|---|---|---|
| Principle | Hall element measures core flux directly | Compensation winding holds core at zero flux |
| Accuracy (EVsense families) | 1% | 0.5% |
| Response time (EVsense families) | < 7 µs | < 1 µs |
| Linearity and drift | Set by core and Hall element | Very good; core works at zero flux |
| Supply current | Low, roughly constant | Rises with measured current |
| Supply (EVsense families) | ±12 V to ±16 V (programmable: 3.3–6 V) | ±12 V to ±24 V |
| Output | Voltage | Current or voltage |
| Size and cost at high current | Smaller, lower | Larger, higher |
| Measuring range (EVsense families) | 15–5000 A | 10–5000 A |
Values in the table come from the EVsense open-loop and closed-loop family specifications. Check the datasheet of the specific model for the exact combination of ratings.
Which one should you choose?
A simple rule works for most projects:
- Choose closed-loop when the current signal is part of a fast control loop or when accuracy and drift directly affect product performance. Examples: traction inverter and servo drive phase current, variable-frequency drives, precise overcurrent protection, and test equipment.
- Choose open-loop when the signal is used for monitoring or protection, when power and space are tight, or when cost matters at high current ratings. Examples: welding machines, power supplies, frequency converters, DC-DC converters and general current monitoring.
- Choose programmable open-loop for compact embedded designs on a low-voltage supply, such as PV inverters, combiner boxes and on-board power electronics.
- Choose high-precision sensors (0.0001–0.02% in the EVsense high-precision family) only for calibration, laboratory, medical or rail applications that really need ppm-level performance.
Watch the power budget of closed-loop sensors
The most common surprise with closed-loop sensors is supply current. At full scale, the secondary winding carries the full compensation current, on top of the sensor’s own consumption. On a large-current sensor this can be tens of milliamps or more per sensor, and it flows from your ±12 V or ±15 V rail. Add it up for all sensors in the system before you size the auxiliary power supply. Also choose the measuring resistor so the output voltage stays within the sensor’s specified range at peak current.
Practical tips for both types
- Centre the conductor in the aperture and keep return conductors away from the sensor, so external fields do not add error.
- Check offset at zero current after warm-up, and compensate in software if your accuracy budget is tight.
- Respect the isolation rating: EVsense open-loop and closed-loop families are rated 2.5–6 kV withstand voltage depending on the model.
- Filter sensibly: fast sensors also pass switching noise; place the ADC filter to suit your control bandwidth.
Summary
- Open-loop: Hall element measures the core flux directly. Compact, low power and economical, with about 1% accuracy.
- Closed-loop: compensation winding keeps the core at zero flux. 0.5% accuracy, response under 1 µs and low drift, at higher power and cost.
- Use closed-loop for fast control loops and precise measurement, open-loop for monitoring, protection and cost-sensitive designs.
- For a full EV application map, read how to choose an EV current sensor, or send us your requirements for a model recommendation.
Frequently asked questions
What is the main difference between open-loop and closed-loop Hall current sensors?
An open-loop sensor amplifies the Hall element's signal directly, so the output depends on the linearity and drift of the core and the Hall element. A closed-loop sensor drives a compensation current through a secondary winding to keep the core flux at zero; the compensation current is the output. That gives the closed-loop sensor better accuracy, linearity and response time, but it draws more supply current.
Is a closed-loop Hall sensor always better?
No. It is more accurate and faster, but it consumes more power (the compensation current flows from the supply), is usually larger and costs more, especially at high currents. For monitoring, protection and many converter applications, an open-loop sensor with about 1% accuracy is the more economical choice.
How accurate are EVsense open-loop and closed-loop sensors?
EVsense open-loop Hall sensors are specified at 1% accuracy with response time below 7 us over 15A to 5000A. Closed-loop Hall sensors are specified at 0.5% with response time below 1 us over 10A to 5000A. For ppm-level work there is a separate high-precision family at 0.0001-0.02%.
What is a programmable Hall current sensor?
It is an open-loop design built around an integrated Hall IC whose gain and offset are calibrated electronically during production. That gives compact size and a low-voltage supply (3.3-6V for EVsense programmable sensors), which suits embedded power electronics, PV inverters and combiner boxes.
Need help choosing a sensor?
Send the nominal and peak current, AC or DC, supply voltage, output interface and aperture size. Our engineers reply with a model recommendation, datasheets and sample options.