Considerations for using tantalum capacitors in portable battery-powered medical devices

Background

There are many types of portable battery-powered medical devices, and there are many choices for charger control circuits that can reliably power these devices. Passive components such as tantalum capacitors (such as surface mount tantalum capacitors and chip capacitors) can improve the overall performance of the charger control and energy storage system in portable devices. Portable battery-powered medical devices can be powered by either disposable batteries or backup rechargeable batteries charged by a battery charger. The demand for portability and ease of use of medical devices has led to many improvements in charging control circuits. Charger and battery systems have evolved from circuits composed of many components to integrated microprocessor-based systems that use fewer passive components and less board space.

Given the high reliability requirements of medical devices, this article provides examples of design trade-offs for commercial and medical tantalum capacitors and describes some new developments that help improve performance. This article also highlights general selection criteria for capacitor technology and advances in packaging technology that can be used in portable medical devices. The most commonly used types of bulk capacitors in portable medical devices are multilayer ceramic capacitors (MLCCs), aluminum electrolytic capacitors, and solid tantalum capacitors. Table 1 describes some general characteristics and possible drawbacks of each capacitor technology.

Table 1. Types of bulk capacitors used in portable medical devices

Battery Charger Basics

For portable devices using rechargeable secondary batteries, several types of chargers can be used: buck chargers, offline chargers, or linear regulator/chargers. The most common type is the buck charger. This charger converts the battery source voltage to a lower voltage and regulates it. The converter can be powered by an external AC/DC adapter or internal adapter circuit. Linear regulators are compact and ideal for low-capacity battery charger applications. Single-chip integrated solutions can power portable devices and charge the battery separately at the same time.

Figure 1 is an example of a small DC/DC switching regulator. It can provide synchronized pulse switching for a battery charger. This pulse battery charging system dissipates little heat and is packaged in a TSSOP package with a height of only 1.2 mm. This device has a rich feature set, including the ability to isolate the battery (Vbat) from the external power supply when shut down.

There are many types of capacitors used in chargers. Input decoupling capacitors are used to bypass noise. Generally, 0.1μF MLCC capacitors are placed near the Vcc pin to filter out high-frequency noise.

Figure 1. A Li-Ion or NiCd/NiMH microprocessor battery charger using the Siliconix Si9731.

The selection of the output capacitor type should be based on the appropriate ESR to meet the stable load line range, and the following items should be evaluated:

1. Ability to reduce power consumption

2. Ability to reduce ripple voltage

3. Able to meet the requirements of system load lines.

The converter is responsible for providing the load current and voltage. As the load changes, the current increases and the voltage drops. The regulator can maintain a constant voltage, but it cannot respond quickly to changes in load current, so a bulk capacitor is used to cope with such changes and prevent the voltage from dropping. If the current output of the converter had to pass through an inductor, it would not be able to respond instantaneously, and a parallel capacitor bank would be needed across the load to pull up the voltage. MLCCs and tantalum capacitors are sometimes mixed to reduce the overall ESR of the bulk capacitor. Due to the lower impedance of the MLCC, it will charge first, followed by the bulk tantalum capacitor.

Power supply and output capacitor requirements

Batteries used in portable medical devices are either disposable or secondary. Disposable batteries are generally used only once. During the operation of the circuit, the active chemical substances are consumed. Once the discharge is complete, the circuit will stop working and a new battery must be replaced. Secondary batteries can be recharged after discharge because the chemical reaction that produces the electricity can be reversed, allowing the battery system to be recharged. The choice of power source and battery type depends on the application. Common types of disposable batteries used in medical devices are alkaline batteries and lithium batteries.

Secondary batteries include lithium batteries, nickel-cadmium batteries (NiCad), nickel-metal hydride (NiMH) batteries and lead-acid batteries. Among them, lithium batteries are the most commonly used because they have the highest volume energy density and mass energy density and a very low discharge rate, which means they have good charge retention when idle.

Table 2 Power consumption and capacity range of tantalum capacitors

Portable device circuits require output capacitors, which are usually powered by primary or secondary batteries to reduce voltage overshoot or undershoot during load transients. To effectively filter noise, the equivalent series resistance (ESR) of the capacitor is a key parameter to consider. The output capacitor is used to handle the ripple current and voltage of the circuit. The overheating of the capacitor bank needs to be controlled so that the maximum allowable power dissipation is not exceeded during circuit operation. It needs to be ensured that the ripple current through the output capacitor does not exceed the allowable value.

Table 2 summarizes the maximum allowable power ratings for various packages (by case size) at +25°C and f=100kHz. For applications with temperature rises above +25°C, further derating is recommended. Please refer to the capacitor manufacturer's power derating recommendations for applicable tantalum packages.

The maximum allowable AC ripple current (Irms) can be calculated using the formula P=Irms2 x ESR, where P represents the maximum allowable power corresponding to the case size of the tantalum capacitor, and ESR can be calculated based on the operating frequency of the capacitor.

For tantalum capacitors, it is also necessary to comply with appropriate voltage derating specifications and not exceed the manufacturer's recommended ratings. The operating voltage of the output capacitor should be determined by the voltage circuit state. It can be calculated according to the formula Vrated=Vpeak+Vdc, which is the ripple voltage plus the DC voltage noise. The calculation method for the allowable ripple voltage is E=IxZ, where Z represents the capacitor resistance. In general, lower ESR can help reduce output ripple noise.

Adding bulk capacitors to the circuit also provides power-up in no-load conditions (when the battery is not operating and the circuit is powered by line current). When powered by line current, the derating specifications should be followed when selecting the rating of bulk tantalum capacitors.

Selecting Output Capacitors for Battery-Powered Low-Dropout Regulators (LDOs)

Linear voltage regulators or low dropout regulators (LDOs) in portable devices are powered by batteries. The size of the capacitor is very important because LDOs are typically packaged in small SOT packages. LDOs are often used to ensure high accuracy when the load changes. A 90mV dropout at a 50mA load current is typical. For example, if the manufacturer of the LDO specifies the use of capacitors for noise reduction, then the capacitor type should be considered:

●Performance requirements for medical equipment

● Specified ESR safe operating range

● Size and cost of capacitors

● Rated voltage

Table 3 ESR requirements for various types of capacitors

There are several choices in capacitor technology to meet the ESR requirements shown in Table 3. By examining the stability of the circuit load line, the appropriate capacitor technology can be selected for proper operation of the line.

Load-line stability analysis of a low dropout (LDO) regulator can yield the minimum and maximum ESR values ​​for various load conditions.

For example, if a 10μF tantalum capacitor is used for load-line transient stability, the safe operating range of the ESR measured at 10kHz is a maximum of 10Ω and a minimum of 10mΩ (see Figure 2).

Figure 2 ESR requirements for stable operation of LDO regulators

In this case, if the LDO is to operate efficiently, a minimum size capacitor with low ESR is required. There are a variety of low ESR capacitor technologies that meet the requirements for this application. The ESR of tantalum capacitors is generally specified by the manufacturer at 100kHz. This application requires an ESR at 10kHz to achieve proper load line stability.

The selection of the appropriate capacitor can be determined by the impedance-frequency relationship at 10kHz. As shown in Table 2, there are several solid tantalum capacitors suitable for this application. The corresponding ESR of MLCC, tantalum capacitors, and aluminum electrolytic capacitors can be found in Table 2. Although tantalum polymer capacitors have lower ESR than standard solid tantalum capacitors with manganese cathodes, due to recent improvements in tantalum capacitor structure using manganese dioxide (MnO2) cathodes, some standard solid tantalum capacitor products have an ESR of less than 50mΩ, which is fully acceptable for LDO applications.

Figure 3 Impedance-frequency curve of 0603 tantalum capacitor

Figure 4 shows a capacitor from the Vishay TM8-298D series in an M or 0603 case size. The 0603 tantalum capacitor has an ESR of 1.19 Ω at 10kHz, as shown in the tantalum capacitor impedance-frequency curve in Figure 3. This ESR is well within the safe operating range, which allows for excellent circuit load line stability. In this case, if an MLCC capacitor with an ultra-low ESR of less than 10 mΩ is used, a small resistor would need to be added in series with the capacitor in the circuit to provide a safe operating range for the ESR. Due to limited space and the number of components, a single 0603 tantalum capacitor can meet both the ESR and space requirements.

Figure 4: Size reduction of tantalum capacitors

In some cases, a circuit requires both bulk capacitors to reduce voltage drop and ultra-low ESR to handle ripple. The best balance between higher efficiency and lower power consumption favors the use of lower ESR capacitors.

Other capacitor technologies with higher ESR can also be used. MLCC0 0805 is a 400 layer 0805 size X5R dielectric capacitor with a specification of 10μF ~ 10V. There are also 10μF ~ 10V capacitors with 0603 X5R dielectric layers. Their ESR is 20mΩ at 10kHz. Compared with tantalum capacitors, MLCC capacitors have very low ESR. However, for capacitors used in LDOs in this application, lower ESR is not an advantage.

Board space and cost are also factors to consider in capacitor selection for this application.

Figure 5 MAP tantalum capacitor package

More advanced tantalum capacitor packages remove the leadframe, improving volumetric packaging efficiency and electrical performance. Figure 6 compares the multi-array packaging (MAP) assembly technology with traditional packaging technology. Eliminating the leadframe assembly in standard tantalum capacitor packages can save more space to accommodate more tantalum cores. In traditional leadframe packaging, the main part of the tantalum capacitor package is the plastic packaging material or packaging material. As shown in Figure 5, the positive lead connected to the leadframe also takes up packaging space. In general, the effective utilization rate of the available volume of traditional leadframe packaging is only 30%.

As shown in Figure 6, the MAP process can improve the placement accuracy of the tantalum core in the package, thereby reducing the overall package size and achieving tighter dimensional error control. The package implemented using the MAP process can also reduce the clearance and provide a better "reference line" for high-density lines in the vertical direction. For example, the maximum height of the standard injection molded lead frame tantalum capacitor D type is 4.1mm, while the height of the D type produced using the MAP process is 1.65mm.

Figure 6 The latest MAP tantalum capacitor package has the highest volume efficiency

With the help of MAP process, the case size of tantalum capacitors has been reduced from A to 0805 (current technology) to 0603 or 0402. The improvement of tantalum powder can reduce the case size of 0805 with 10?F~10V capacity to the case size of 0603, as shown in Figure 4.

Capacitor DC leakage / insulation resistance comparison

When using batteries as a power source, capacitor DC leakage current (DCL) should be considered as a loss because the capacitor affects the battery's condition and life. In addition to batteries, large-capacity capacitors are also used as supplementary power sources in portable devices to cope with changes in circuit loads.

Many portable device applications require low DCL to achieve long, efficient battery life. To cope with load changes, a large output capacitor in parallel with the battery can maintain power storage capacity. In some applications, the device operation time is intermittent short cycles, and the battery is idle most of the time. Therefore, the capacitor needs to have a very low DCL to meet the application requirements of portable devices and maximize the battery life.

DC leakage is a very small value and is present in all capacitors. Tantalum capacitors have leakage currents in the microamps range, while MLCCs have leakage currents in the picoamps range. DC leakage is measured by applying a DC voltage to an equivalent series resistor-capacitor circuit and measuring the current at room temperature. A 1000Ω resistor should be connected in series with the capacitor to limit the charging current.

Figure 7 DCL curve of tantalum capacitor

The terminology and unit of measurement used to describe DCL varies with capacitor technology. DCL is the unit of measurement used for tantalum capacitors, while insulation resistance (IR) is the unit of measurement used for MLCCs. MLCCs have an IR limit based on the dielectric type. For large capacitor MLCCs with X5R dielectrics, the IR limit is >10,000MΩ or (R x C) ≥ 500ΩF, whichever is lower. MLCCs are screened for IR minimums using an automated IR tester that complies with Military Specification 55681.

DCL can be calculated using Ohm's law using the capacitor's IR and rated voltage. For example, the MLCC IR limit is 100MΩ-?F, which is equivalent to a tantalum capacitor standard DCL limit of 0.01, i.e. (capacitance x voltage) = 0.01?A/?FV.

Tantalum capacitors are screened according to the specified minimum DCL value, or not exceeding the specified maximum value. The DCL test of tantalum capacitors is based on military product specification 55365F. The DCL differences between tantalum capacitors of various specifications are quite obvious, so the limit value of each specification of tantalum capacitor is separately specified.

In portable applications, DCL at a longer soak time is an important indicator of capacitors. For tantalum capacitors with specific specifications and tantalum core designs, the DCL distribution in a production batch can be quantified. If the application requires extremely low DCL, it is easy to automatically screen out tantalum capacitors with a specific DCL at a rated voltage that meet the conditions for use in portable devices from a batch.

Figure 8 shows a 47uF-10V tantalum capacitor. Although its maximum DCL is 4.7? A, after screening according to a specific holding time, it can provide ultra-low DCL for the application. Taking the component in Figure 8 as an example, the batch can be screened according to the standard of 10 seconds DCL 600nA, thereby reducing the overall DCL from 4.7uA to 600nA limit.

Figure 8 A 47uF-10V tantalum capacitor

The DCL limit should be determined based on the operating and non-operating time of the battery-powered device. For example, if a portable device operates for only a few seconds and then remains idle for a long time, the bulk capacitor should have a low DCL to ensure a long battery life. In addition, the overall quiescent current and operating current of the circuit should be evaluated to determine whether a low DCL capacitor is required.

Battery Runtime and DCL

For rechargeable secondary batteries, DCL is also important to extend the time between charges, but a certain degree of leakage current from the output capacitor can be allowed in the overall operating current. Evaluating the current requirements of the circuit under various usage conditions and understanding the DCL of the capacitor can significantly extend the battery life.

By measuring DCL or IR, we can understand the performance of the capacitor dielectric and the quality of the dielectric layer. When powered, the DCL current will flow through or across the capacitor dielectric isolation layer. For capacitors made of oxide film such as tantalum capacitors, the main component of the DCL current is a mixture of multiple currents, including surface leakage current flowing through the dielectric, dielectric absorption (DA) current due to polarization of the dielectric material, and native leakage current flowing through the dielectric material. Similarly, the leakage current of MLCCs using ceramic dielectrics based on barium titanate is mainly leakage current flowing through the dielectric, as well as DA loss and native leakage current.

MLCCs have good low DCL characteristics, but in some cases, tantalum capacitors can provide the same low DCL in a smaller size. Table 5 compares the calculation method for correctly evaluating and selecting the appropriate capacitor based on the DCL requirements. As shown in Table 5, tantalum capacitors are generally specified according to the maximum DCL value. Tantalum capacitors with standard manganese dioxide (MnO2) construction are graded by the manufacturer according to (.01xCV). Some capacitor manufacturers also provide specific hold times along with the DCL information and pre-screen capacitors based on specific DCL limits that are much lower than the maximum DCL value of the same grade.

Selecting Appropriate Low DCL Capacitors

For example, a portable battery-powered medical device with a short duty cycle requires the circuit to start the motor for a few seconds every day and then shut it down. Such an application can use a low DCL bulk capacitor.

Specific use:

● DC/DC converter for motor drive

● Input voltage: 1.5V

● Fixed output voltage: 3.3V

● Output current: 200mA@2V

● Bulk output capacitor: 47?F

● DCL = 200nA when the holding time is 60 seconds

If the 47uF bulk capacitor is a tantalum capacitor, appropriate voltage derating should be performed. The derating should be based on the derating specifications of the tantalum capacitor manufacturer. For specific examples, see Table 4. In this example, a rated voltage of 10V is selected.

Table 4 Derating specifications for tantalum capacitors

The rated voltage of an MLCC can be the same as or slightly higher than the operating voltage, so a rated voltage of 6V is sufficient. For MLCCs, if the IR (see Table 5) and the operating voltage (4V) are known, the DCL can be calculated. MLCCs suitable for low DCL applications have two dielectrics, X5R and X7R. Based on the rated operating voltage, the DCL can be calculated using the IR value of the component according to Ohm's law.

Table 5 Low DCL capacitor selection

To determine the DCL limit for tantalum capacitors, MAP 47? F-10V capacitors with case sizes D and F from multiple production batches were batch tested, and the DCL and corresponding hold time at different hold times (60 seconds) for each capacitor were recorded, as shown in Figure 7. Statistical analysis was then used to determine the lower DCL for each batch. In addition, a unique molding process was used to strengthen the negative electrode to improve and reduce the DCL performance of the capacitor. Any DCL curve that deviated from the standard batch was paid attention to, and finally the lower limit of DCL was found.

Figure 4 shows the various package options and the volume requirements for each. Vishay’s 572D series tantalum capacitors meet the DCL requirements while offering the highest volumetric efficiency at only 8.39 mm3. If space requirements are not so critical, MLCCs can also be used for this application. X5R dielectric MLCCs have a DCL as low as 187nA, and as with tantalum capacitors, only one bulk capacitor will meet the requirements. MLCC X7R dielectric capacitors have a better capacitance temperature coefficient than X5R, but two MLCC capacitors are required in parallel to form a bulk capacitor.

The ability of a capacitor to maintain capacitance after voltage is applied is an important consideration in some circuits. For X5R dielectric MLCCs, the voltage coefficient of capacitance (VCC) should be considered when selecting the component's rated voltage. If the DC application voltage, including the ripple voltage, is close to the MLCC's rated voltage, the VCC effect will cause the component to lose some capacitance. Capacitor loss may affect circuit operation. Also, consider the effect of temperature on the IR of the MLCC and the temperature coefficient of capacitance (TCC) when selecting components. Manufacturers provide curves for the degradation of IR with increasing temperature for specific dielectrics. Temperature effects should be evaluated during design.

Improving DCL of Tantalum Capacitors

The dielectric layer of the tantalum capacitor is a thin film of tantalum pentoxide, covering the surface of each tantalum core. It is anodized and is composed of a 5nm~10nm thick N-type tantalum oxide layer and a pure semiconductor layer of tantalum pentoxide. The layer thickness is proportional to the anodizing voltage and determines the rated voltage of the component. For solid tantalum capacitors used in 6V battery applications, the final tantalum dielectric layer thickness is 0.04 microns or 40 nanometers.

Ultra-large capacity MLCCs are manufactured by pouring a thin layer of ceramic dielectric with a thickness of 2.0 microns, which is much thicker than tantalum capacitors. MLCCs use a stacking process to ultimately create multilayer capacitors. Like tantalum capacitors, the thickness of the dielectric layer of MLCCs determines the rated voltage, and the number of dielectric layers determines the capacity. The difference in dielectric constants leads to huge differences in IR.

The DCL of tantalum capacitors will increase due to mechanical damage to the positive electrode surface or cracks in the oxide layer surface. As shown in Figure 8, the outer surface of the positive electrode is a vulnerable part that is affected by thermal, mechanical and electrical effects. The surface DCL will be affected by humidity and cause instability during long-term operation.

Improving the production process of tantalum cores and better controlling the thickness of the oxide layer can help eliminate the surface DCL problem shown in Figure 8. A thicker dielectric film is generated on the outer surface of the tantalum core to protect it from mechanical damage, thereby greatly improving the DCL performance and reducing DCL. In addition to improving the positive electrode structure of tantalum capacitors, the manganese dioxide negative electrode structure of tantalum capacitors has better DCL performance compared to the polymer negative electrode structure because the material has better conductivity.

Figure 9 shows a new MAP 0603 package with excellent DCL performance made with this new technology. Combined with the improvements to the tantalum core, the latest MAP series tantalum packages can improve assembly, packaging and termination processes, avoid mechanical damage, and increase the volumetric efficiency of the capacitor.

Fig. 9

Improving DCL Reliability of Medical-Grade Tantalum Capacitors

Because some medical equipment requires high reliability, especially for mission-critical applications, capacitor manufacturers provide robust and conservative designs to meet performance requirements. Through careful tantalum core and tantalum powder design, the performance of medical tantalum capacitors is higher than that of standard commercial tantalum capacitors and high-reliability products produced using traditional technology.

Manufacturers evaluate the appropriate tantalum powder for each design. As the capacitor CV increases, the failure rate increases, so the appropriate particle size of tantalum powder should be selected for the specific design. For medical-grade designs, the goal is to provide more reliable DCL performance within the available case size range. For commercial-grade designs, the goal is to minimize costs and maximize design benefits by providing higher-k CV tantalum powder in the smallest available case size. Therefore, the DCL of commercial tantalum capacitors will generally be higher than that of medical tantalum capacitors.

The following example illustrates the comparison between the improved medical TM8 series DCL and the traditional high-reliability 194D series.

Figure 10 compares the 194D series design with the TM8 series design in the F case size. The 194D is an older design used in many high-reliability applications. The tantalum core design uses a high-k CV powder of 23kCV. The TM8 is a newer medical grade design that uses 10Kvc powder, greatly improved DCL performance, and uses the latest MAP assembly process without increasing board space.

Fig.10

High Energy Storage Tantalum Capacitors in Medical Devices

Small portable or implantable cardioverter defibrillators (ICDs) are used for patients who are at risk of sudden cardiac death from ventricular tachyarrhythmias. Portable defibrillators have similar functions to ICDs and are designed to deliver electrical therapy to the heart to restore normal heart rhythm. The electrical therapy circuit uses a high-energy charged capacitor to shock the heart tissue.

Some designs use high energy aluminum electrolytic capacitors, but require a backup battery and a procedure to implement a recuperation period to maintain good charge efficiency over the life of the device. High energy wet tantalum rechargeable capacitors do not require recuperation and have higher energy density than aluminum electrolytic capacitors.

The energy storage capacity of a capacitor depends on the value of the relative permittivity of the dielectric and the maximum permissible voltage in the material. When an electric field is present, any conduction of the dielectric of the capacitor results in capacitance losses. And the losses increase as the electric field changes, such as with alternating current. The molecules of the dielectric are polarized to a certain extent, and when the electric field is present, the displacement of these molecules is initially opposite. Part of the energy is consumed in the displacement of the molecules and is consumed in the process. When the electric field changes or disappears, this loss manifests as heat.

Foil-type aluminum electrolytic capacitors are immersed in a conductive electrolyte. The dielectric consists of an oxide film on the surface of the aluminum foil, which is generally 50 to 100 nanometers thick and determines the capacity per unit electrode area. Tantalum capacitors also have an oxide film layer, but the thickness is much smaller, generally only 5 to 10 nanometers. When choosing the type of capacitor to use in an energy storage device, it is necessary to consider operating life, board space and cost requirements. Because cardiac defibrillation requires very high energy, only aluminum electrolytic capacitors and wet tantalum capacitors are suitable.

in conclusion

This article discusses various applications of portable medical devices and the circuits used in them. There are a variety of capacitors available for these portable applications. When selecting a capacitor for this application, the preferred electrical parameters are the capacitor's DCL and ESR. Due to the extremely high reliability and battery life requirements of some medical applications, some capacitors are not suitable.