November 4, 2016
To assure reliable operation, transient voltage suppression should be considered at early stages of the design process. This can be a complex task as electronic components are increasingly sensitive to stray electrical transients. The designer must define the types of transient threats and determine what applications are needed while meeting the product agency norms and standards.
Varistors are increasingly used as the front-line solution for transient surge protection. Littelfuse provides expertise to the designer and offers the broadest range of circuit protection technologies to choose from.
Littelfuse varistors are available in a variety of forms to serve a wide range of applications. Options include ultra small surface mount multi-layer suppressor (MLV) devices for small electronics applications, and traditional mid-range metal oxide varistors (MOVs) and axial metal oxide varistors for protection of small machinery, power sources and components. Littelfuse also offers larger terminal mount MOVs for industrial applications.
A more recent innovation to the the Littelfuse product line, MLVs address a specific part of the transient voltage spectrum – the circuit board level environment where, although lower in energy, transients from ESD, inductive load switching, and even lightning surge remnants would otherwise reach sensitive integrated circuits. Each of these events can relate to a product's ElectroMagnetic Compatibility (EMC), or its immunity to transients that could cause damage or malfunction.
Littelfuse offers five distinct versions of MLVs including the MHS Series ESD Suppressor for high data rates, the ML Series which supports the broadest application range, the MLE Series intended for ESD while providing filter functions, the MLN Series Quad Array in a 1206 & 0805 chip and the AUML Series characterized for the specific transients found in automotive electronic systems.
Surface mountable MOV (Metal Oxide Varistor) devices facilitate customs in SMT assembly process and resolve the PCB space limitation issue. They are reflow and wave solderable and include the CH, SM7, SM20, MLE, MHS, ML, and MLN series.
Traditional radial through-hole MOV (Metal Oxide Varistor) devices are available in diameters of 5mm, 7mm, 10mm, 14mm, 20mm and 25mm. They are fit for providing voltage surge protection for a wide variety of applications and include the C-III, iTMOV, LA, TMOV, RA, UltraMOV, UltraMOV25S, and ZA series.
Bare disc varistors are industrial high-energy elements. They are designed for special applications requiring unique electrical contact or packaging methods asked for by customers. The CA Series of transient surge suppressors are industrial high-energy disc varistors (MOVs) intended for special applications requiring unique electrical contact or packaging methods provided by the customer.
Thermal protective Metal Oxide Varistors (TMOVs) are designed to meet abnormal overvoltage requirements of UL 1449. They can be wave soldered without any need for special or expensive assembly processes and include the iTMOV, TMOV,TMOV25S, and TMOV34S series.
Industrial high energy varistors provide a much higher surge and energy rating than regular MOVs (Metal Oxide Varistors) and also possess various terminals to fit different assembly requests or conditions. They include the BA, BB, CA, DA, HA,HB34, HC, HF34, HG34, TMOV34S, UltraMOV25S, C-III, FBMOV, and TMOV25S series.
Integrated varistors consist of a 40kA varistor building block (MOV) with an integral thermally activated element. These devices are recognized as an independent Type 1 SPD by UL.
The Littelfuse FBMOV Series thermally protected and Non Fragmenting varistor represents a new development in circuit protection. It consists of a 40kA varistor building block (MOV) with an integral thermally activated element designed to open in the event of overheating due to abnormal over-voltage, limited current conditions.
Littelfuse plants for PolySwitch devices are ISO/TS 16949:2009 and ISO 9001:2008 certified.
Introduction to Overvoltage Suppression
Voltage transients are defined as short duration surges of electrical energy and are the result of the sudden release of energy that was previously stored, or induced by other means, such as heavy inductive loads or lightning strikes. In electrical or electronic circuits, this energy can be released in a predictable manner via controlled switching actions, or randomly induced into a circuit from external sources.
Repeatable transients are frequently caused by the operation of motors, generators, or the switching of reactive circuit components. Random transients, on the other hand, are often caused by Lightning (Figure 1) and Electrostatic Discharge (ESD) (Figure 2). Lightning and ESD generally occur unpredictably, and may require elaborate monitoring to be accurately measured, especially if induced at the circuit board level. Numerous electronics standards groups have analyzed transient voltage occurrences using accepted monitoring or testing methods. The key characteristics of several transients are shown below in Table 1.
Figure 1. Lightning Transient Waveform
Table 1. Examples of Transient Sources and Magnitude
Characteristics of Transient Voltage Spikes
Transient voltage spikes generally exhibit a "double exponential" wave form, shown in Figure 1 for lightning and figure 2 for ESD. The exponential rise time of lightning is in the range 1.2µs to 10µs (essentially 10% to 90%) and the duration is in the range of 50µs to 1000µs (50% of peak values). ESD on the other hand, is a much shorter duration event. The rise time has been characterized at less than 1 ns. The overall duration is approximately 100ns.
Figure 2. ESD Test Waveform
Why are Transients of Increasing Concern?
Component miniaturization has resulted in increased sensitivity to electrical stresses. Microprocessors for example, have structures and conductive paths which are unable to handle high currents from ESD transients. Such components operate at very low voltages, so voltage disturbances must be controlled to prevent device interruption and latent or catastrophic failures. Sensitive devices such as microprocessors are being adopted at an exponential rate. Microprocessors are beginning to perform transparent operations never before imagined. Everything from home appliances, such as dishwashers, to industrial controls and even toys, have increased the use of microprocessors to improve functionality and efficiency.
Vehicles now employ many electronics systems to control the engine, climate, braking and, in some cases, steering systems. Some of the innovations are designed to improve efficiency, but many are safety related, such as ABS and traction control systems. Many of the features in appliances and automobiles use modules which present transient threats (such as electric motors). Not only is the general environment hostile, but the equipment or appliance can also be sources of threats. For this reason, careful circuit design and the correct use of overvoltage protection technology will greatly improve the reliability and safety of the end application. Table 2 shows the vulnerability of various component technologies.
|Device Type||Vulnerability (volts)|
TABLE 2. RANGE OF DEVICE VULNERABILITY.
Transient Voltage Scenarios
ESD (Electrostatic Discharge)
Electrostatic discharge is characterized by very fast rise times and very high peak voltages and currents. This energy is the result of an imbalance of positive and negative charges between objects.
Below are some examples of the voltages which can be generated, depending on the relative humidity (RH):
- Walking across a carpet:
35kV @ RH = 20%; 1.5kV @ RH = 65%
- Walking across a vinyl floor:
12kV @ RH = 20%; 250V @ RH = 65%
- Worker at a bench:
6kV @ RH = 20%; 100V @ RH = 65%
- Vinyl envelopes:
7kV @ RH = 20%; 600V @ RH = 65%
- Poly bag picked up from desk:
20kV @ RH = 20%; 1.2kV @ RH = 65%
Referring to Table 2 on the previous page, it can be seen that ESD that is generated by everyday activities can far surpass the vulnerability threshold of standard semiconductor technologies. Figure 2 shows the ESD waveform as defined in the IEC 61000-4-2 test specification.
Inductive Load Switching
The switching of inductive loads generates high energy transients which increase in magnitude with increasingly heavy loads. When the inductive load is switched off, the collapsing magnetic field is converted into electrical energy which takes the form of a double exponential transient. Depending on the source, these transients can be as large as hundreds of volts and hundreds of Amps, with duration times of 400ms.
Typical sources of inductive transients are:
These examples are extremely common in electrical and electronic systems. Because the sizes of the loads vary according to the application, the wave shape, duration, peak current and peak voltage are all variables which exist in real world transients. Once these variables can be approximated, a suitable suppressor technology can be selected.
Figure 3. Automotive Load Dump
Lightning Induced Transients
Even though a direct strike is clearly destructive, transients induced by lightning are not the result of a direct strike. When a lightning strike occurs, the event creates a magnetic field which can induce transients of large magnitude in nearby electrical cables.
Figure 4, shows how a cloud-to-cloud strike will effect not only ove RHead cables, but also buried cables. Even a strike 1 mile distant (1.6km) can generate 70V in electrical cables.
Figure 4. Cloud-to-Cloud Lightning Strike
Figure 5, on the following page, shows the effect of a cloud-to-ground strike: the transient–generating effect is far greater.
Figure 5. Cloud-to-Ground Lightning Strike
Figure 6, shows a typical current waveform for induced lightning disturbances.
Figure 6. Peak Pulse Current Test Waveform
Technological Solutions for Transient Threats
Because of the various types of transients and applications, it is important to correctly match the suppression solution to the different applications. Littelfuse offers the broadest range of circuit protection technologies to ensure that you get the proper solution for your application. Please consult our online library of Application Notes and Design Notes for further information on common design issues encountered at http://www.littelfuse.com.
Metal Oxide Varistors and Multi-Layered Varistors
Varistors are voltage dependent, nonlinear devices which have electrical characteristics similar to back-to- back Zener diodes. They are composed primarily of ZNO with small additions of other metal oxides such as Bismuth, Cobalt, Magnese and others. The Metal Oxide Varistor or "MOV" is sintered during the manufacturing operation into a ceramic semiconductor and results in a crystalline microstructure that allows MOVs to dissipate very high levels of transient energy across the entire bulk of the device. Therefore, MOVs are typically used for the suppression of lightning and other high energy transients found in industrial or AC line applications. Additionally, MOVs are used in DC circuits such as low voltage power supplies and automobile applications. Their manufacturing process permits many different form factors with the radial leaded disc being the most common.
Multilayer Varistors or MLVs are constructed of ZNO material similar to standard MOVs, however, they are fabricated with interweaved layers of metal electrodes and supplied in leadless ceramic packages. As with standard MOVs, Multilayers transition from a high impedance to a conduction state when subjected to voltages that exceed their nominal voltage rating. MLVs are constructed in various chip form sizes and are capable of significant surge energy for their physical size. Thus, data line and power supply suppression are achieved with one technology.
The following parameters apply to Varistors and/or Multilayer Varistors and should be understood by the circuit designer to properly select a device for a given application.
Introduction to Varistor Technology
The varistor body structure consists of a matrix of conductive ZNO grains separated by grain boundaries providing P-N junction semiconductor characteristics. These boundaries are responsible for blocking conduction at low voltages and are the source of the nonlinear electrical conduction at higher voltages.
FIGURE 1. TYPICAL VARISTOR V-I CHARACTERISTIC
The symmetrical, sharp breakdown characteristics shown in Figure 1, enable the varistor to provide excellent transient suppression performance. When exposed to high voltage transients the varistor impedance changes many orders of magnitude from a near open circuit to a highly conductive level, thus clamping the transient voltage to a safe level. The potentially destructive energy of the incoming transient pulse is absorbed by the varistor, thereby protecting vulnerable circuit components.
Since electrical conduction occurs, in effect, between ZNO grains distributed throughout the bulk of the device, the Littelfuse Varistor is inherently more rugged than its single P-N junction counterparts, such as Zener diodes. In the varistor, energy is absorbed uniformly throughout the body of the device with the resultant heating spread evenly through its volume. Electrical properties are controlled mainly by the physical dimensions of the varistor body which is sintered in various form factors such as discs, chips and tubes. The energy rating is determined by volume, voltage rating by thickness or current flow path length, and current capability by area measured normal to the direction of current flow.
MOVs are designed to protect sensitive circuits against external transients (lightning) and internal transients (inductive load switching, relay switching and capacitor discharges). And other high level transients found in industrial, AC line application or lower level transients found in automotive DC line applications with peak current rating ranging from 20A to 500A and peak energy rating from 0.05J - 2.5J.
An attractive property of the MOV is that the electrical characteristics are related to the bulk of the device. Each ZnO grain of the ceramic acts as if it has a semiconductor junction at the grain boundary. A cross-section of the material is shown in Figure 2, which illustrates the ceramic microstructure. Varistors are fabricated by forming and sintering Zinc Oxide-based powders into ceramic parts. These parts are then electroded with either thick film Silver or arc/flame sprayed metal.
The ZnO grain boundaries can be clearly observed. Since the nonlinear electrical behavior occurs at the boundary of each semiconducting ZnO grain, the varistor can be considered a "multi-junction" device composed of many series and parallel connections of grain boundaries. Device behavior may be analyzed with respect to the details of the ceramic microstructure. Mean grain size and grain size distribution play a major role in electrical behavior.
FIGURE 2. OPTICAL PHOTOMICROGRAPH OF A POLISHED AND ETCHED SECTION OF A VARISTOR
The bulk of the varistor between contacts is comprised of ZnO grains of an average size "d" as shown in the schematic model of Figure 3. Resistivity of the ZnO is <0.3 Ω-cm.
FIGURE 3. SCHEMATIC DEPICTION OF THE MICROSTRUCTURE OF A
METAL-OXIDE VARISTOR, GRAINS OF CONDUCTING ZnO (AVERAGE
SIZE d) ARE SEPARATED BY INTERGRANULAR BOUNDARIES.
Designing a varistor for a given nominal varistor voltage, (VN), is basically a matter of selecting the device thickness such that the appropriate number of grains, (n), are in series between electrodes. In practice, the varistor material is characterized by a voltage gradient measured across its thickness by a specific volts/mm value. By controlling composition and manufacturing conditions the gradient remains fixed. Because there are practical limits to the range of thicknesses achievable, more than one voltage gradient value is desired. By altering the composition of the metal oxide additives it is possible to change the grain size "d" and achieve the desired result.
A fundamental property of the ZnO varistor is that the voltage drop across a single interface "junction" between grains is nearly constant. Observations over a range of compositional variations and processing conditions show a fixed voltage drop of about 2V-3V per grain boundary junction. Also, the voltage drop does not vary for grains of different sizes. It follows, then, that the varistor voltage will be determined by the thickness of the material and the size of the ZnO grains. The relationship can be stated very simply as follows:
The varistor voltage, (VN), is defined as the voltage across a varistor at the point on its V-I characteristic where the transition (v) is complete from the low-level linear region to the highly nonlinear region. For standard measurement purposes, it is arbitrarily defined as the voltage at a current of 1mA. Some typical values of dimensions for Littelfuse Varistors are given in Table 1.
|VARISTOR VOLTAGE||AVERAGE GRAIN SIZE||n||GRADIENT||DEVICE THICKNESS|
|VOLTS||MICRONS||V/mm AT 1mA||mm|
NOTE: Low voltage formulation.
Theory of Operation
Because of the polycrystalline nature of metal-oxide semiconductor varistors, the physical operation of the device is more complex than that of conventional semiconductors. Intensive measurement has determined many of the device's electrical characteristics, and much effort continues to better define the varistor's operation. However from the user's viewpoint, this is not nearly as important as understanding the basic electrical properties as they relate to device construction.
The key to explaining metal-oxide varistor operation lies in understanding the electronic phenomena occurring near the grain boundaries, or junctions between the ZNO grains. While some of the early theory supposed that electronic tunneling occurred through an insulating second phase layer at the grain boundaries, varistor operation is probably better described by a series-parallel arrangement of semiconducting diodes. In this model, the grain boundaries contain defect states which trap free electrons from the n-type semiconducting ZNO grains, thus forming a space charge depletion layer in the ZnO grains in the region adjacent to the grain boundaries. (See reference notes on the last page of this section).
Evidence for depletion layers in the varistor is shown in Figure 4, where the inverse of the capacitance per boundary squared is plotted against the applied voltage per boundary. This is the same type of behavior observed carrier concentration, N, was determined to be about 2 x 1017 per cm3. In addition, the width of the depletion layer was calculated to be about 1000 Angstrom units. Single junction studies also support the diode model.
It is these depletion layers that block the free flow of carriers and are responsible for the low voltage insulating behavior in the leakage region as depicted in Figure 5. The leakage current is due to the free flow of carriers across the field lowered barrier, and is thermally activated, at least above about 25°C. For semiconductor abrupt P-N junction diodes. The relationship is:
(Vb) = barrier voltage,
(V) = applied voltage,
(q) = electron charge,
(es) = semiconductor permittivity and
(N) = carrier concentration.
From this relationship the ZnO carrier concentration, N, was determined to be about 2 x 1017 per cm3.
In addition, the width of the depletion layer was calculated to be about 1000 Angstrom units. Single junction studies also support the diode model.
FIGURE 4. CAPACITANCE-VOLTAGE BEHAVIOR OF VARISTOR RESEMBLES
A SEMICONDUCTOR ABRUPT-JUNCTION REVERSED
BIASED DIODE Nd ˜ 2 x 1017/cm3
Figure 5, shows an energy band diagram for a ZnO-grain boundary-ZnO junction. The left-hand grain is forward biased, VL, and the right side is reverse biased toVR. The depletion layer widths are XL and XR, and the respective barrier heights are fL and fR. The zero biased barrier height is fO. As the voltage bias is increased, fL is decreased and fR is increased, leading to a lowering of the barrier and an increase in conduction.
The barrier height fL of a low voltage varistor was measured as a function of applied voltage, and is presented in Figure 6. The rapid decrease in the barrier at high voltage represents the onset of nonlinear conduction.
FIGURE 5. ENERGY BAND DIAGRAM OF A ZnO-GRAINBOUNDARY-ZnO JUNCTION
FIGURE 6. THERMAL BARRIER vs APPLIED VOLTAGE
Transport mechanisms in the nonlinear region are very complicated and are still the subject of active research. Most theories draw their inspiration from semiconductor transport theory and is not covered in detail in this document.
The process of fabricating a Littelfuse Varistor is illustrated in the flow chart of Figure 7. The starting material may differ in the composition of the additive oxides, in order to cover the voltage range of product.
FIGURE 7. SCHEMATIC FLOW DIAGRAM OF LITTELFUSE VARISTOR FABRICATION
Device characteristics are determined at the pressing operation. The powder is pressed into a form of predetermined thickness in order to obtain a desired value of nominal voltage. To obtain the desired ratings of peak current and energy capability, the electrode area and mass of the device are varied. The range of diameters obtainable in disc product offerings is listed here:
Of course, other shapes, such as rectangles, are also possible by simply changing the press dies. Other ceramic fabrication techniques can be used to make different shapes. For example, rods or tubes are made by extruding and cutting to length. After forming, the green (i.e., unfired) parts are placed in a kiln and sintered at peak temperatures in excess of 1200°C. The B ismuth oxide is molten above 825°C, assisting in the initial densification of the polycrystalline ceramic. At higher temperatures, grain growth occurs, forming a structure with controlled grain size.
Electroding is accomplished, for radial and chip devices, by means of thick film silver fired onto the ceramic surface. Wire leads or strap terminals are then soldered in place. A conductive epoxy is used for connecting leads to the axial 3mm discs. For the larger industrial devices (40mm and 60mm diameter discs) the contact material is arc sprayed Aluminum, with an overspray of Copper if necessary to give a solderable surface.
Many encapsulation techniques are used in the assembly of the various Littelfuse Varistor packages. Most radials and some industrial devices (HA Series) are epoxy coated in a fluidized bed, whereas epoxy is "spun" onto the axial device.
Radials are also available with phenolic coatings applied using a wet process. The PA Series package consists of plastic molded around a 20mm disc subassembly. The RA, DA and DB Series devices are all similar in that they all are composed of discs or chips, with tabs or leads, encased in a molded plastic shell filled with epoxy. Different package styles allow variation in energy ratings, as well as in mechanical mounting.
TABLE 2. BY-TYPE CERAMIC DIMENSIONS
|Leadless Surface Mount||CH, AUML† , ML† , MLE† , MLN† Series||5mm x 8mm Chip, 0603, 0805, 1206, 1210, 1812, 2220|
|Axial Leaded||MA Series||3mm Diameter Disc|
|Radial Leaded||ZA, LA, C-III, TMOV®,
i TMOV® ,UltraMOV™, TMOV25S® Series
|5mm, 7mm, 10mm, 14mm, 20mm Diameter Discs|
|Boxed, Low Profile||RA Series||5mm x 8mm, 10mm x 16mm, 14 x 22 Chips|
|Industrial Packages||BA, BB Series
DA, DB Series
HA, HB Series
HC, HF Series
|32mm, 40mm Diameter Disc, 34mm Square Disc, 40mm Diameter Disc, 60mm Diameter Disc|
|Industrial Discs||CA Series||60mm Diameter Discs|
Figure 9A, 9B and 9C (below) show construction details of some Littelfuse varistor packages. Dimensions of the ceramic, by package type, are above in Table 2.
FIGURE 9A. CROSS-SECTION OF MA SERIES
FIGURE 9B. CROSS-SECTION OF RADIAL LEAD PACKAGE
FIGURE 9C. PICTORIAL VIEW OF HIGH ENERGY DA, DB AND BA/BB SERIES
Electrical Characterization Varistor V-I Characteristics
Turning now to the high current upturn region in Figure 10, we see that the V-I behavior approaches an ohmic characteristic. The limiting resistance value depends upon the electrical conductivity of the body of the semiconducting ZnO grains, which have carrier concentrations in the range of 1017 to 1018 per cm3. This would put the ZnO resistivity below 0.3Ωcm.
FIGURE 10. TYPICAL VARISTOR V-I CURVE PLOTTED ON LOG-LOG SCALE
Varistor electrical characteristics are conveniently displayed using log-log format in order to show the wide range of the V-I curve. The log format also is clearer than a linear representation which tends to exaggerate the nonlinearity in proportion to the current scale chosen. A typical V-I characteristic curve is shown in Figure 10. This plot shows a wider range of current than is normally provided on varistor data sheets in order to illustrate three distinct regions of electrical operation.
Equivalent Circuit Model
An electrical model for the varistor can be represented by the simplified equivalent circuit of Figure 11.
FIGURE 11. VARISTOR EQUIVALENT CIRCUIT MODEL
Leakage Region of Operation
At low current levels, the V-I Curve approaches a linear (ohmic) relationship and shows a significant temperature dependence. The varistor is in a high resistance mode (approaching 109Ω) and appears as an open circuit. The nonlinear resistance component (RX) can be ignored because (ROFF) in parallel will predominate. Also, (RON) will be insignificant compared to (ROFF).
FIGURE 12. EQUIVALENT CIRCUIT AT LOW CURRENTS
For a given varistor device, capacitance remains approximately constant over a wide range of voltage and frequency in the leakage region. The value of capacitance drops only slightly as voltage is applied to the varistor. As the voltage approaches the nominal varistor voltage, the capacitance decreases. Capacitance remains nearly constant with frequency change up to 100 kHz. Similarly, the change with temperature is small, the 25°C value of capacitance being well with +/-10% from -40°C to +125°C.
The temperature effect of the V-I characteristic curve in the leakage region is shown in Figure 13. A distinct temperature dependence is noted.
FIGURE 13. TEMPERATURE DEPENDENCE OF THE CHARACTERISTIC CURVE IN THE LEAKAGE REGION
The relation between the leakage current (I) and temperature (T) is
The temperature variation, in effect, corresponds to a change in (ROFF). However, (ROFF) remains at a high resistance value even at elevated temperatures. For example, it is still in the range of 10MΩ to 100MΩ at 125°C.
Although (ROFF) is a high resistance it varies with frequency. The relationship is approximately linear with inverse frequency.
If however, the parallel combination of (ROFF) and (°C) is predominantly capacitive at any frequency of interest. This is because the capacitive reactance also varies approximately linearly with 1/f.
At higher currents, at and above the mA range, temperature variation becomes minimal. The plot of the temperature coefficient (dV/dT) is given in Figure 14. It should be noted that the temperature coefficient is negative (-) and decreases as current rises. In the clamping voltage range of the varistor (I > 1A), the temperature dependency approaches zero.
FIGURE 14. RELATION OF TEMPERATURE COEFFICIENT DV/DT TO VARISTOR CURRENT
Nominal Varistor Region of Operation
The varistor characteristic follows the equation:
I = kVa, where (k) is a constant and the exponent (a) defines the degree of nonlinearity. Alpha is a figure of merit and can be determined from the slope of the V-I curve or calculated from the formula:
In this region the varistor is conducting and RX will predominate over C, RON andROFF. RX becomes many orders of magnitude less than ROFF but remains larger than RON.
FIGURE 15. EQUIVALENT CIRCUIT AT VARISTOR CONDUCTION
During conduction the varistor voltage remains relatively constant for a change in current of several orders of magnitude. In effect, the device resistance, RX, is changing in response to current. This can be observed by examining the static or dynamic resistance as a function of current. The static resistance is defined by:
Plots of typical resistance values vs current (I) are given in Figure 16A and 16B.
FIGURE 16A. RX STATIC VARISTOR RESISTANCE FIGURE
FIGURE 16B. ZX DYNAMIC VARISTOR RESISTANCE
Upturn Region of Operation
At high currents, approaching the maximum rating, the varistor approximates a short-circuit. The curve departs from the nonlinear relation and approaches the value of the material bulk resistance, about 1Ω-10Ω. The upturn takes place as RXapproaches the value of RON. Resistor RON represents the bulk resistance of the ZNO grains. This resistance is linear (which appears as a steeper slope on the log plot) and occurs at currents 50A to 50,000A, depending on the varistor size.
FIGURE 17. EQUIVALENT CIRCUIT AT VARISTOR UPTURN
Speed of Response and Rate Effects
The varistor action depends on a conduction mechanism similar to that of other semiconductor devices. For this reason, conduction occurs very rapidly, with no apparent time lag – even into the nanosecond (ns) range. Figure 18, shows a composite photograph of two voltage traces with and without a varistor inserted in a very low inductance impulse generator. The second trace (which is not synchronized with the first, but merely superimposed on the oscilloscope screen) shows that the voltage clamping effect of the varistor occurs in less than 1.0 ns.
FIGURE 18. RESPONSE OF A ZnO VARISTOR TO A FAST RISE TIME (500ps) PULSE
In the conventional lead–mounted devices, the inductance of the leads would completely mask the fast action of the varistor; therefore, the test circuit for Figure 18, required insertion of a small piece of varistor material in a coaxial line to demonstrate the intrinsic varistor response.
Tests made on lead– mounted devices, even with careful attention to minimizing lead length, show that the voltages induced in the loop formed by the leads contribute a substantial part of the voltage appearing across the terminals of a varistor at high current and fast current rise. Fortunately, the currents which can be delivered by a transient source are invariably slower in rise time than the observed voltage transients. The applications most frequently encountered for varistors involve current rise times longer than 0.5μs.
Voltage rate-of-rise is not the best term to use when discussing the response of a varistor to a fast impulse (unlike spark gaps where a finite time is involved in switching from nonconducting to conducting state). The response time of the varistor to the transient current that a circuit can deliver is the appropriate characteristic to consider.
The V-I characteristic of Figure 19A, shows how the response of the varistor is affected by the current waveform. From such data, an "overshoot" effect can be defined as being the relative increase in the maximum voltage appearing across the varistor during a fast current rise, using the conventional 8/20μs current wave as the reference. Figure 19B, shows typical clamping voltage variation with rise time for various current levels.
FIGURE 19. RESPONSE OF LEAD-MOUNTED VARISTORS TO CURRENT WAVEFORM
FIGURE 19A. V-I CHARACTERISTICS FOR VARIOUS CURRENT RISE TIMES
FIGURE 19B. OVERSHOOT DEFINED WITH REFERENCE TO THE BASIC 8/20?s CURRENT PULSE
How to Connect a Littelfuse Varistor
Transient suppressors can be exposed to high currents for short durations in the nanoseconds to millisecond time frame.
Littelfuse Varistors are connected in parallel to the load, and any voltage drop in the leads to the varistor will reduce its effectiveness. Best results are obtained by using short leads that are close together to reduce induced voltages and a low ohmic resistance to reduce I • R drops.
This is the most complete protection one can select, but in many cases only Varistor 1 or Varistor 1 and 2 are selected.
FIGURE 25A. 3 PHASE 220V/380V, UNGROUNDED
FIGURE 25B. 3 PHASE 220V OR 380V, UNGROUNDED
FIGURE 25C. 3 PHASE 220V, ONE PHASE GROUNDED
FIGURE 25D. 3 PHASE 220V
FIGURE 25E. 3 PHASE 120V/208V, 4-WIRE
FIGURE 25F. 3 PHASE 240V/415V
For higher voltages use same connections, but select varistors for the appropriate voltage rating.
DC applications require connection between plus and minus or plus and ground and minus and ground.
For example, if a transient towards ground exists on all 3 phases (common mode transients) only transient suppressors connected phase to ground would absorb energy. Transient suppressors connected phase to phase would not be effective.
FIGURE 26. COMMON MODE TRANSIENT AND CORRECT SOLUTION
On the other hand if a differential mode of transient (phase to phase) exists then transient suppressors connected phase to phase would be the correct solution.
FIGURE 27. DIFFERENTIAL MODE TRANSIENT AND CORRECT SOLUTION
This is just a selection of some of the more important variations in connecting transient suppressors.
The logical approach is to connect the transient suppressor between the points of the potential difference created by the transient.The suppressor will then equalize or reduce these potentials to lower and harmless levels.
Varistor Terms and Definitions
Definitions (IEEE Standard C62.33, 1982)
A characteristic is an inherent and measurable property of a device. Such a property may be electrical, mechanical, or thermal, and can be expressed as a value for stated conditions.
A rating is a value which establishes either a limiting capability or a limiting condition (either maximum or minimum) for operation of a device. It is determined for specified values of environment and operation. The ratings indicate a level of stress which may be applied to the device without causing degradation or failure. Varistor symbols are defined on the linear V-I graph illustrated in Figure 20.
FIGURE 20. I-V GRAPH ILLUSTRATING SYMBOLS AND DEFINITIONS
Voltage Clamping Device
A clamping device, such as an MOV, refers to a characteristic in which the effective resistance changes from a high to low state as a function of applied voltage. In its conductive state, a voltage divider action is established between the clamping device and the source impedance of the circuit. Clamping devices are generally "dissipative" devices, converting much of the transient electrical energy to heat.
Choosing the most appropriate suppressor depends upon a balance between the application, its operation, voltage transient threats expected and sensitivity levels of the components requiring protection. Form factor/package style also must be considered.
At high current and energy levels, varistor characteristics are measured, of necessity, with an impulse waveform. Shown in Figure 21, is the ANSI Standard C62.1 waveshape, an exponentially decaying waveform representative of lightning surges and the discharge of stored energy in reactive circuits.
The 8/20μs current wave (8μs rise and 20μs to 50% decay of peak value) is used as a standard, based on industry practices, for the characteristics and ratings described. One exception is the energy rating (WTM), where a longer waveform of 10/1000μs is used. This condition is more representative of the high energy surges usually experienced from inductive discharge of motors and transformers. Varistors are rated for a maximum pulse energy surge that results in a varistor voltage (VN) shift of less than +/-10% from initial value.
FIGURE 21. DEFINITION OF PULSE CURRENT WAVEFORM
Power Dissipation Ratings
When transients occur in rapid succession the average power dissipation is the energy WTM (watt-seconds) per pulse times the number of pulses per second. The power so developed must be within the specifications shown in the Device Ratings and Characteristics Table for the specific device. Certain parameters must be derated at high temperatures.
FIGURE 22. DEVICE RATINGS AND CHARACTERISTICS
TABLE 3. VARISTOR CHARACTERISTICS(IEEE STANDARD C62.33-1982 SUBSECTION 2.3 AND 2.4)