要約

In this design solution, we review some of the challenging operating conditions and common failure mechanisms for high-side switch circuits used in actuators operating in a factory environment. We present a controller IC that integrates a diverse range of safety features to monitor circuit operation and take appropriate action to prevent damage should these conditions occur.

Introduction

“By failing to prepare, you are preparing to fail” is an adage well worth bearing in mind when designing equipment for the modern automated factory, where failure is not an option. With dozens of actuators simultaneously opening and closing relay and solenoids valves (Figure 1), these harsh, noisy, and demanding environments can stretch the performance of even the most robust components to their limit and even then, carefully designed protection circuitry is required to ensure that they are not pushed beyond their safe operating boundaries. Nonetheless, even the best laid plans can go awry and an unanticipated overload or undetected short-circuit can quickly destroy a switching component, leading to costly production downtime.

In this Design Solution we review the operation of the high side switch circuit and consider some common operating and fault conditions, which if not detected and responded to in a timely manner, have the potential to permanently damage or even destroy a switch and/or its controller. We then present a new controller IC that integrates a myriad of safety and protection features making it ideally placed to provide the highest levels of protection to high-side switch circuits if these (or other) fault conditions were to occur.

What is a High-Side Switch?

A transistor that switches on/off a ground-connected load is commonly referred to as a “high-side switch”, since it is used to conduct current from a ‘high’ voltage rail (typically 24V in an industrial application) to the load, which is connected to a lower voltage (typically 0V). An n-channel FET operating in the saturation region is required for this type of arrangement (Figure 2). A low voltage on/off signal is sent from a microcontroller to the switch controller IC which then provides the higher gate voltage required to turn on and off the switch, as required. These simple circuits sometimes also referred to as “Digital Outputs”, are often used in industrial actuators to open/close valves, energise solenoids and for motor braking.

In some versions of this circuit e.g. MAX14915, the switch is integrated into the controller IC, while in other implementations, it is an external discrete component. While integration is the obvious appeal of the former, the advantage of the latter approach is that it provides the freedom to choose a switch with a drive current appropriate to a specific load.

We next consider some common operating and fault conditions that can occur in the factory environment, their potential implications for a high-side switch circuit and some approaches to protecting the circuit from lasting damage should these conditions occur.

Inductive Voltage Surge

When a valve is closed (or opened, depending on the polarity of the system), the switch is turned off and current flowing into the inductive load (e.g. a solenoid) is abruptly stopped, starting the process of demagnetization. The nature of an inductor is to oppose this effect and in so doing creates a large kick-back voltage, in an attempt to keep current flowing. This kick-back voltage, which appears at the S terminal of the controller must be limited/clamped at a voltage in the middle of the range of the “Absolute Maximum Ratings” of the controller IC to prevent damage being caused by the large negative voltage that develops between the S and G terminals of the device. A common way to do this is to use a Transient Voltage Suppression (TVS) diode that clamps the voltage level on the source terminal at a predetermined safe voltage level. When choosing a TVS diode, its rated peak current should be greater than the peak current flowing through it in during normal application. During voltage clamping, the peak power dissipated in the TVS should be within its rated specification (for the highest operating temperature). To prevent damage, it is also critical that the source terminal of the driver IC is also able to withstand the maximum clamping voltage of the TVS.

Stuck Valve

Sometimes in an industrial process, a valve may become temporarily stuck in a position that prevents it from opening or closing fully. This can cause an ‘overcurrent’ condition to occur where the switch driver IC gradually increases current to the load to try and free up the valve. If the valve is suddenly freed, the problem is resolved and the current flowing in the switch will quickly return to normal. However, if the valve remains stuck in one position, then current may continue to increase beyond that rated value for the switch. If this condition persists for more than a short period of time, excess heat can permanently damage or even destroy the switch. Therefore, it is important to limit on the size and duration of overcurrent that can occur.

Short-Circuit

A short-circuit of the LOAD and RETURN terminals is the most extreme type of overcurrent condition. In this case, the current is limited only by the on resistance, R_DSON) of the switch and if not quickly detected and/or guarded against will cause the switch to burnout, causing irreparable damage. Protecting against a short-circuit is done in the same way as for an overcurrent condition.

Overheating

In the event of an unexpected rise in temperature e.g. a cooling fan failure, the ambient operating conditions of industrial equipment can exceed maximum ratings. In an actuator, this can push the transistor switch outside of its rated operating temperature range. At best this may cause degraded performance, at worst it can cause total burnout. It is important to continuously monitor equipment ambient operating temperature so that it can be shut down quickly if necessary.

Incorrect Wiring

Humans are not infallible and even the best trained technicians can make mistakes, especially when under pressure to commission a new production line and get it up and running quickly. When confronted with a ‘rats-nest’ of wires, one of the easiest and therefore most common mistakes a technician can make is to connect a positive terminal to negative or vice-versa. High-side driver circuits must be robust enough to safely handle this condition.

Power Supply Overvoltage

While 24V DC is the nominal voltage level used by most industrial equipment, it is not uncommon for this voltage to vary considerably (due to cross-contamination from surges in adjacent high-current switching equipment). While a degree of variation in supply voltage is to be expected and is usually planned for, a fault in the power supply used to generate the 24V rail could cause it rise substantially beyond the safe operating threshold of the switch and/or controller. Guarding against this event requires having some means to detect that the supply voltage has risen beyond a pre-set threshold, beyond which the switching circuit can be shut down if this occurs.

Taking Control

The high-side switch circuit shown in Figure 3 features a high-side switch controller IC (MAX14922) that provides an array of safety features to protect the complete circuit, making it reliable and robust, even when confronted with the most challenging operating and fault conditions, such as those considered previously.

This controller provides fast inductive load turn-off at the S (Source) input, which can be achieved by using a ground connected high-voltage TVS diode to provide voltage clamping against voltage (and ESD) surges from -70V (max) up to V_DD+6V.

It also enables current limiting when a sense resistor R_S is connected between the V_DD and the SNS input terminals. The maximum overcurrent is then I_OC = V_OC/R_S. If an overcurrent condition occurs, the controller takes a number of precautionary actions. Firstly, an open-drain diagnostic output (active-low OVCURR), which is high during normal circuit operation, transitions low. This can be used as a flag to the microcontroller to take appropriate action, if desired. The switching controller then begins to adjust V_GS of the switch to actively regulate the current for a fixed “blanking-time” (set by the capacitor value on the t_BLANK input). If the overcurrent condition persists for a period longer than the “blanking-time”, the controller then turns the switch off for protection purposes. After an off-delay equal to about 50 “blanking time” intervals (i.e., 10ms), the switch is automatically turned on again. Auto-retry on/off cycling continues until the cause for overcurrent is removed by a technician. The active-low OVCURR output remains low until the overcurrent condition is removed.

For example, in Figure 4, a 1nF capacitor has set the “blanking-time” interval to 200µs.

After an off-delay equal to about 50 “blanking time” intervals (or 10ms), the switch is automatically turned on again (Figure 5), and this “auto-retry” cycle repeats indefinitely until the fault has been resolved.

In the event of a short-circuit, the controller turns the switch off for approximately 5µs and then turns it back on at a controlled rate so that the short-circuit load current is then determined by the sense resistor value (Figure 6). Similarly, to the overcurrent condition, the regulation phase and auto-retry intervals are determined the by the value of the C_BLANK capacitor.

Another benefit of this IC is that is includes integrated temperature monitoring and a protective shutdown feature. An integrated temperature sensor signals a thermal warning at 110°C (typical). When this occurs, the active-low THW logic output goes low indicating an overtemperature event although the device continues to function normally. If the temperature cools down by 10°C, the active-low THW logic output returns high. However, if the temperature continues to rise above 150°C, the controller enters shutdown mode, regardless of the state of the IN input (from the microcontroller). In shutdown mode, the G output is turned off, forcing the switch to turn off completely. When the temperature reduces by 10°C, the device returns to normal operation with the active-low THW output going low once the temperature has fallen below 110°C.

Supply overvoltage detection is another useful safety feature of this controller. If the V_DD rises beyond the overvoltage threshold (approximately +39V), the avtive-low OV output goes active low. This does not affect the switch controller which continues to operate normally (and will continue to do for V_DD up to 70V), but the active-low OV output acts as a flag to the microcontroller to indicate that the supply voltage is higher than the system is designed for. For some applications, this signal could be used to gate the signal from the system microcontroller to prevent IN going high if an overvoltage condition occurs. For even further robustness, the part provides integrated protection against supply voltage miswiring.

Apart from the safety features of this IC, it also includes an internal charge-pump to provide higher drive current to the gate of the switch. This ensures that it is fully saturated with minimum on-resistance (R_DSON). This reduces I²R power dissipation (and associated heating), which is undesirable if the circuit is housed in a small enclosure and allows switching rates up to 50KHz. Conveniently, this switching controller IC also includes an on-board 5V LDO regulator, capable of delivering up to 50mA of current to external circuitry, where required.

Summary

At first glance the industrial high-side switch appears to be a relatively trivial circuit. However, when confronted with the task of operating in a harsh factory environment, there are many ways in which it can fail, with the potential to cause costly production downtime. In this Design Solution we have considered some of the many challenging operating and failure conditions that a high-side switch can encounter. We then presented a new high-side switch controller IC that includes a multitude of diagnostic and safety features to confer these circuits with an unsurpassed degree of robustness, while in operation on the factory floor. Available in a 3mm x 3mm 16-TQFN package, it is ideal for use with a wide selection of n-channel FET devices, in relay/solenoid valves and motor braking applications that require between 1A and 10A of current.

A similar version of this design solution originally appeared in Electronic Product Design & Test (UK) on November 01, 2020.