1 Application Brief

Introduction to Linear Motor Transport Systems

Linear motor transport systems are an emerging industrial application to replace and enhance conventional conveyer-belt-based factory automation systems. Unlike a conveyer belt, where rotating electric motors drive the belt at a certain speed, linear motor transport systems use multiple static, non-moving coil-based linear motor modules to drive magnetic movers, which carry the product to be assembled or processed. There are several system advantages over a conventional conveyer-belt based system.

Due to the versatility of the mechanism, it is now possible to achieve smaller and new geometries of a coil-based linear motor segment modules which include straight, curved, crossing or even two-dimensional movement. Since every linear motor within a segment can be controlled individually, it is possible to drive multiple movers on the same segment at different speeds and positions. For example, one mover stops to process or analyze the product carried by that mover, while the other products on the next movers are traveling on to the next process state at maximum speed. This independent control increases throughput significantly compared to traditional conveyer belt driven systems.

Figure 1-1 shows a simplified linear motor transport system, with the static linear and curved segments and the movers. Each segment typically includes the Linear Motor Power Stage with the corresponding coils to realize multiple 3-phase linear motors, a Linear Motor Position Sensor to sense the position and speed of each mover per segment and a Linear Motor Segment Controller for real-time position and motion control of the movers. A simplified picture of a mover is shown in Figure 1-2.

Requirements

Linear motor transport systems enable multiple magnetic movers traveling in one or even two dimensions with speeds up to 10 m/s and a linear position accuracy and repeatability as low as 0.01 mm. The magnetic field range present at the magnetic sensor depends on the mover’s sense magnet and the distance between the mover magnet to the static multi-position sensor printed-circuit board (PCB). Typically, the magnetic field ranges from 50 mT to 300 mT. Due to space requirements, small packages with highly integrated 3D hall sensor system-on-chip (SoC) are an advantage. An ambient operating temperature of the sensor beyond 85°C, such as 125°C, enables higher power densities, while still capturing accurate sensor data under these extreme conditions.

Since the position of multiple movers within one segment need to be detected at the same time, simultaneous sampling and low latency position measurement are important. 3D Hall-effect sensors with low latency digital interface offer an advantage over analog output SoC due to higher robustness against noise. Further advantages of an SoC with digital interface is the option to integrate diagnostics and monitoring of the SoC, for example die temperature, hall element or supply voltage diagnostics to increase system reliability.

TMAG5170 3D Hall-effect Sensor

TMAG5170 is a high precision linear Hall-effect sensor with sensitivity in three axes. The device is highly customizable with the ability to detect each component of the B-field vector using mutually orthogonal Hall-effect elements. This device samples each channel serially and then converts the result using an integrated 12-bit ADC at a maximum sample rate of 20 ksps.

Figure 1-7 TMAG5170 Block Diagram

Among the many features of TMAG5170 are integrated self-diagnostics, programmable alert thresholds, and customizable triggering for deterministic sampling.

The self-diagnostics of TMAG5170 are capable of monitoring Vcc status, power on reset, output pin status, device memory, temperature, and various other internal checks during live operation. This benefit allows the microcontroller to easily verify each sensor in the system remains operational and helps to flag system level issues that can lead to reliability or safety risks.

There are three triggering modes to initiate magnetic field conversions from TMAG5170. The alert pin of the device can function as an input pin that can be driven by any appropriate IO pin from the host controller. This hardware triggering is convenient and simple. Additionally, the conversion can be triggered through individual SPI command or by the CS pin of the device. With a known trigger timing, the conversion rate of the device can be used to correlate any given measurement to the actual position of the linear mover.

Figure 1-8 shows an example timing with an 8 kHz position sample rate using the ALERT pin to trigger a new conversion through the host MCU. For this example, the TMAG5170 is configured for pseudo-simultaneous sampling mode of the two axes, Z and X. The effective sample to data-transmit latency is around 60 us.

The SPI transfer of the 32-bit data with the TMAG5170 takes 3.2 us at 10 MHz SPI clock and corresponding setup and hold time, so the overall latency is around 64us when multiple TMAG5170 SDO data lines are connected to a host MCU in parallel. When multiple TMAG5170 use the same multiplexed SDO data line, the overall latency depends on the number of TMAG5170 sensors read out sequentially. The overall latency equals the 60 us plus N-times 4 us to 5 us, when N is the number of SPI transfers including an overhead for the chip select signals setup and hold times.

One benefit of using either the SPI command trigger or the CS trigger, is that it leaves the alert pin available to provide additional feedback to the controller. When using the programmable threshold limits, each TMAG5170 in the system can provide feedback about the relative proximity of any individual linear mover. By identifying only the sensors which are receiving useful input, the system can implement more efficient SPI reads over arrays of devices.

For more information, please consider the following alternate devices and supporting documentation.