As Electronic devices proliferate and regulations governing user safety evolve, designers are looking for options to enhance device protection while minimizing cost and board space. The problem is that circuit protection is a lot like insurance: it may seem like an unnecessary expense until it’s needed. This protection is needed against a variety of internal and external aberrations and faults, including internal and external short-circuit, overcurrent, and voltage-surge situations. These occurrences can temporarily or permanently disable a system; damage the system, its internal components, or the load; and even result in harm to a user.
No single protection solution works for all faults and situations. For example, when it comes to implementing overvoltage protection (OVP), crowbars such as gas discharge tubes (GDTs) are generally better for long-term faults, while clamps such as a metal oxide varistor (MOV) are better-suited for transient events. However, GDTs suffer from “holdover current”, and MOVs can fail permanently and can reach dangerously high temperatures due to thermal runaway. Using both components connected together in series in a hybrid approach compensates for any potential issues, but this approach complicates board layout and adds cost. Advances in design are needed to remove this compromise.
This article describes the importance of OVP protection and various approaches to achieving it. It then introduces IsoMOV technology, which combines the benefits of GDTs and MOVs in a single device with longer life and no holdover current. It will then introduce example devices from Bourns Inc., describe their salient characteristics, and show how to select and use them for effective, efficient, low-cost protection.
Protection has multiple perspectives
There is no “one size fits all” solution for circuit and system protection. There are two reasons for this: first, there are many types of faults and occurrences against which protection is needed; second, the magnitude and duration of the fault condition determines the type and ruggedness of protection needed.
Among the many general fault situations are:
- Overcurrent, where the load draws too much current due to an external fault, short circuit, or internal component failure (including insulation failure)
- Overvoltage, where a part of the system is stressed by excessive voltage due to misconnection
- Thermal, where a component overheats due to poor design, inadequate thermal management, or excessive ambient heat
- Component failure, where an internal component fails which leads to an overcurrent/overvoltage situation which damages other components or the load
Faults also often have consequences beyond just affecting or even damaging a system, as they can result in a shock hazard to users.
Crowbars and clamps for surge protection
Among the most challenging fault condition on both AC and DC circuits is the overvoltage surge, called a temporary overvoltage (TOV) event. This short pulse or spike is often due to nearby lightning strikes or electrical switching which injects harmful transients into electrical equipment and its sensitive electronics.
Two broad classes of surge protection devices (SPDs) are used to deal with overvoltage and TOV events: the crowbar and the clamp. (Note that these terms are sometimes used interchangeably in “casual” discussion, but they are not the same.)
In brief, a crowbar becomes a short circuit across the line being protected, thus diverting the surge and its current to ground, preventing it from reaching the circuitry (Figure 1). The crowbar is triggered to go into this low-impedance mode when the overvoltage situation occurs.
Interesting side note: The term “crowbar” supposedly comes from the action of industrial workers in the early days of electricity, who would throw an actual metal crowbar across the power and ground bus bars when an overvoltage situation occurred.
Figure 1: When the crowbar protection function triggers, it becomes a low-impedance path between the line it is protecting and ground, thus diverting the overvoltage surge to ground. (Image source: Bourns, Inc.)
The crowbar stays in a low-impedance mode until the current decreases below the “holding current”, at which time it returns to the high-impedance, normal-operation state. It must be able to handle the current flowing through it for the amount of time the supply is in an overvoltage state.
In contrast, a clamp prevents the voltage from exceeding a preset level (Figure 2). When the transient voltage reaches the limiting level for which the clamping device is rated, it will clamp the voltage until the fault extinguishes, at which time the line will return to its normal operating mode. It is important that the rated clamping voltage be higher than the normal operating voltage.
Figure 2: In contrast to the crowbar, the clamp limits the overvoltage surge to a predefined value. (Image source: Bourns, Inc.)
A clamp conducts just enough current to maintain the voltage across it at a safe, desired value while the transient is above the clamp’s conduction voltage. This current, although small, can lead to some safety-related issues that must be addressed, and which may require additional protection, an issue which is discussed further below. It must be rated for the power it will have to dissipate for a specific time, which is usually a relatively short transient event.
Implementing the OVP functions
Since crowbars and clamps are critical protection devices, it is essential that they be simple, reliable, and have well-understood and consistent performance attributes. In this way, they are like the thermally activated fuse, the classic overcurrent-protection component often used as an additional layer of protection.
The crowbar device: The most common crowbar device is the GDT, a carefully constructed and dimensioned spark gap in a hermetic housing filled with an inert gas. In normal operation, before a TOV event, it appears like a near-infinite resistance (Figure 3). However, when the overvoltage surge occurs and exceeds the GDT design voltage, the gas ionizes and the tube will “flash over” like a spark gap, and switch from high impedance to a very low impedance. This change will temporarily short out the line until the fault extinguishes.
Figure 3: The GDT is a sophisticated spark-gap device which conducts only when the voltage across its terminals exceeds its design value; until then, it looks like a nearly perfect open circuit. (Image source: Bourns, Inc.)
The GDT is commonly used in DC circuits, telecommunication circuits, and signal circuits, all of which are generally at a fairly low current of one amp or less. Note that contrary to the dramatic GDTs seen in movies, the GDT for low-level surges is a small, encased, pc board mountable component, and the flashover spark is not visible. Smaller GDTs are offered with ratings from 75 to 600 volts; larger ones are offered with ratings to thousands of volts. One problem with GDTs is their follow-on current (also called holdover current), which is current that continues to flow even after the fault has been extinguished.
The clamping device: Two of the most widely used options for clamping are the power transient voltage suppressor (PTVS) diode and the metal oxide varistor (MOV), both commonly used for high-current protection in AC and DC circuits, motors, communication lines and sensing circuits (Figure 4). MOVs are available with voltage ratings in the tens to over a thousand volts.
Figure 4: The metal oxide varistor (and power transient voltage suppressor) provide a clamping voltage covering a wide design range. (Image source: Bourns, Inc.)
MOVs typically conduct a small amount of leakage current, even with applied voltages that are well below their nominal threshold voltage. If an MOV is subject to voltage surges beyond its rating, permanent damage can occur that causes the leakage current to increase. Even though this current is usually only a few milliamperes, it can present a shock hazard in some circumstances.
Further, self-heating will occur inside the MOV if this leakage current becomes sufficiently high. When an MOV is connected continuously across the AC mains, this self-heating can create positive feedback where higher leakage current leads to increased self-heating, which in turn leads to even higher leakage current. Subsequent surges can further accelerate this cycle.
At some point, the MOV will go into a thermal-runaway mode that generates considerable heat and destroys the MOV. In some situations, the heat produced by the MOV can become a potential ignition source (PIS) and cause nearby materials to catch fire. This effect must be considered and dealt with for basic safety and safety-related standards.
A better OVP solution
In order to provide an OVP solution which has virtually no leakage current and thus longer operational life, designers often use a dual-component arrangement. This hybrid approach combines two discrete devices: a series-connected GDT and MOV (Figure 5), with a combined voltage-versus-time curve (Figure 6).
Figure 5: The hybrid approach of connecting a GDT and MOV in series provides a more-effective OVP solution. (Image source: Bourns, Inc.)
Figure 6: The response versus time of the hybrid GDT + MOV arrangement shows how it combines the basic response attributes of each device. (Image source: Bourns, Inc.)
This is an effective way to have each device compensate for the possible shortcomings of the other. However, there are costs associated with this approach:
- It requires more circuit board real estate
- The bill of materials (BOM) has another component added to it
Another challenge is that the circuit board layout in the region of the MOV and GDT is complicated by regulatory-driven requirements defining minimum creepage and clearance distances, where:
- Clearance is the shortest distance in air between two conductive parts
- Creepage is the shortest distance along the surface of a solid insulating material between two conductive parts
The problem is that clearance and creepage distances increase with voltage. As a result, the placement of the MOV and GDT components adds another mandate and constraint to factor into board layout.
To help designers address these cost, space, and regulatory issues, Bourns, Inc. has developed the IsoMOV series of hybrid protection components. The family provides an alternative solution that combines both an MOV and a GDT in a single package, offering the equivalent functionality of a discrete MOV and GDT in series (Figure 7).
Figure 7: The schematic symbol for the IsoMOV (right) shows it as a merger of the GDT (center, left) and MOV (top and bottom, left) individual standard symbols. (Image source: Bourns, Inc.)
A look at the construction of the IsoMOV shows that is it not just an obvious, simplistic co-packaging of a MOV and a GDT in a single shared enclosure (Figure 8).
Figure 8: The physical construction of the IsoMOV is a completely different realization of the hybrid function, rather than just a co-packaged arrangement of the two individual existing devices. (Image source: Bourns, Inc.)
After assembly of the core, the leads are attached, and the unit is epoxy coated. The result is a familiar radial disc MOV package that is only slightly thicker, with a smaller diameter than the similarly rated conventional devices (Figure 9). Further, due to the patent-pending design of the metal oxide technology, the IsoMOV component also has a higher current rating for the same size. Both the footprint penalty and creepage/clearance issues are eliminated.
Figure 9: The radial lead disk package of the IsoMOV looks like a standard MOV, except it is smaller in diameter and has a higher current rating than an equivalent MOV alone. (Image source: Bourns, Inc.)
The IsoMOV is more than just “the best of both worlds”, as there are other advantages to the design. MOV failures are generally characterized by a so-called “surge hole” at the edge of the metalized area, which is typically caused by an elevated temperature inside the MOV during a surge. Bourns’ unique EdgMOV technology is designed to substantially reduce or eliminate this failure mode.
A look at one IsoMOV model provides more detailed insight. The ISOM3-275-B-L2 features a maximum continuous operating voltage (MCOV) rating of 275 volts root mean square (rms)/350 volts DC; the current rating is 3 kiloamps (kA)/15 operations), 6 kA/1 operation (maximum). Also of special interest is its low capacitance of 30 picofarads (pF) at 20 kilohertz (kHz) which makes it a good fit for high-speed data lines, and it has a low leakage of under 10 microamps (µA).
The role of standards
Design engineers must implement various forms of surge (and other) protection for many reasons ranging from prudent design practice to being mandated by various regulatory standards. Some of these standards are universal and apply to any equipment meeting a general operating scenario, such as AC line operation; others are specific to a certain class of applications, such as medical devices. Among the standards-setting organizations are UL, IEEE, and the IEC; many of their standards are “harmonized” and thus are identical, or nearly so.
All of these standards are complex with numerous mandates; they also include exceptions calling out steps or features which can be eliminated in some circumstances, as well as additional requirements which must be added in others. For example, both IEC 60950-1, “Information technology equipment – Safety” and UL/IEC 62368-1, and “Standard for Audio/video, information and communication technology equipment – Part 1: Safety requirements” (which replaced IEC 60950-1 in 2020), require that the rated voltage of the MOV be at least 125% of the rated voltage of the equipment. As a consequence, the rated voltage of a MOV must be at least 300 volts rms for a 240-volt rms mains circuit.
Consider the common case of the AC line plug, which comes in two and three-prong versions. In theory, the three-wire version provides a safety ground, but in practice, that ground is often not connected or available. The lack of a true Earth safety-ground connection can lead to a potentially dangerous condition when only the hot and neutral wires are available. In that case, it is necessary to add protection components to the design to prevent possible electric shock if users touch conductive parts that are supposed to be grounded but aren’t. However, in this case, the small amount of MOV leakage current can become an electric shock hazard.
The most common solution for preventing MOV leakage current from becoming this hazardous is to place at least one GDT in series with the MOV (Figure 10). By using an IsoMOV device, the functions of both MOV and GDT are in one space-saving package. Thus, the IsoMOV is also a problem-solving component which simplifies meeting the safety requirement called out by UL/IEC 62368-1.
Figure 10: To eliminate user shock hazard due to unavoidable leakage current in an ungrounded application, two devices—an MOV and a GDT—can be placed in series between the hot and neutral AC lines. (Image source: Bourns, Inc.)
Figure 11: The alternative to using an individual MOV and GDT is to use a single IsoMOV device, resulting in the same or better performance, providing a much smaller overall solution. (Image source: Bourns, Inc.)
Engineers are often tasked with deciding which solution is “best.” In most cases, there are tradeoffs with no single, simple answer. In general, when it comes to implementing overvoltage protection, crowbars are better for long-term faults, while clamps are better-suited for transient events. However, using both devices increases footprint and complicates board layout.
Now, however, there’s no need to compromise. Bourns’ IsoMOVs provide much longer operational life than a MOV alone yet without the follow-on current issues of the GDT. The devices provide surge and overvoltage protection that meets all relevant standards in a small footprint. In addition, their low leakage current minimizes follow-on problems, while their very low capacitance makes them suitable for protection of low-voltage, high-speed circuits.