When you have finished, press the "Technical" button on the left to return to the Main Page, or press an appropriate button to go to a different section.
If there are no buttons on the left, you probably got directly to this page via a search engine.
There are probably more different types of capacitors, or condensers as they used to be called, than any other type of electrical component, so I will limit the scope of this article to a brief description of the various types and their uses. The different types of capacitor have different circuit symbols, as shown below.
As with resistors, capacitors come in a large range of physical forms and may be either fixed or variable. Many are available with axial or radial leads, or are primarily intended for printed circuit mounting or are fitted with threaded bushes or solder tags. There are types with no leads at all that are either intended for surface mounting using metalled ends intended to be directly soldered onto copper tracks on a PCB, or in the case of bare discs, with surface metallising intended to be soldered directly to other components or copper track. There are encapsulated variants for use in adverse environmental conditions, high power types, sub-miniature surface mounting types, low self-inductance types, high voltage types, the list goes on and on.
But first, what is a capacitor? For the purposes of this article, we can regard a capacitor, other than a varicap diode or varactor diode, as two conductive plates separated by an insulator, the dielectric, or a stack of such devices connected in parallel. The capacity value is inversely proportional to the distance between the plates and is proportional to the number of plate pairs, the area of the plates and the dielectric constant, or permittivity, of the insulator. A vacuum, or for all practical purposes, air, has a dielectric constant of 1, with all other insulators having values greater than unity.
Capacity values are now stated in Farads (F), using the normal decimal sub- multipliers of pico (p), nano (n), and micro (µ), although "jars" were originally used (1.0µ = 900jars). As the farad is such a large unit, it is not normally necessary to use multipliers. Thus, for example, 0.1 picafarad is normally written as 0p1, 1.0 picafarad as 1p0, 1.5 picafarads as 1p5, 1000 picafarads as 1000p or 1n, 0.1 microfarads as 100n, 1.0 microfarad as 1µ and 1000 microfarads as 1000µ. Alpha-numeric marking is used for larger value types , whereas standard colour coding or alpha-numeric markings are used to indicate component values on low value types. The latter method is self evident and in the former method, a series of coloured bands is used to indicate value and tolerance, with the first band located near one end of the component and often wider than the other bands.
The Standard EIA Colour Code Table per EIA-RS-279 is as follows:-
Colour 1st band 2nd band 3rd band 4th band Temperature
(Multiplier) (tolerance) Coefficient
Black 0 0 ×1 ±1%
Brown 1 1 ×10 ±2% 100 ppm
Red 2 2 ×100 50 ppm
Orange 3 3 ×1000 15 ppm
Yellow 4 4 ×100000 25 ppm
Green 5 5 ×1000000 ±0.5%
Blue 6 6 ×10000000 ±0.25%
Violet 7 7 ×100000000 ±0.1%
Grey 8 8 ×1000000000 ±0.05%
White 9 9 ×10000000000
Gold ×0.1 ±5%
Silver ×0.01 ±10%
None ±20%Although it is theoretically possible to manufacture capacitors having any nominal value of capacity, the smaller value types are normally produced with "standard" picafarad values. These values are normally in the E12, also known as the 10% series (10, 12, 15, 18, 22, 27, 33, 39, 47, 56, 68 and 82) or the E24, also known as the 5% series (10, 11, 12, 13, 15, 16, 18, 20, 22, 24, 27, 30, 33, 36, 39, 43, 47, 51, 56, 62, 68, 75, 82, 91) series. Larger value types are normally manufactured with "decade" microfarad values, e.g. 1.0, 10, 100, 1000. Electrolytics are often manufactured with intermediate microfarad values, such as 2, 8, 16, 32, 50, 64, 100, 150, 200 and 500.
Temperature coefficients are stated in "parts per million per degree centigrade" and can be either positive or negative, although negative values are more common. For example, the capacity of a 100pf capacitor, with a stated temperature coefficient of N750 would reduce by 7.5pf if its temperature were to be raised by 100 degrees centigrade.
Before continuing, let us consider voltage, current and temperature ratings. Most people understand the voltage rating of a capacitor but many do not appreciate that capacitors also have series resistance, leakage resistance and maximum current ratings. These are not normally a problem when capacitors are used as coupling or de-coupling components in low power circuits. However, in applications such as reservoir and smoothing capacitors in power supplies, coupling and tuning capacitors in high power RF transmitters and high power pulse applications, such as are required for flash discharge tubes, the current rating and series resistance are very important, as currents of many amperes may be encountered.
The series resistance of a capacitor is that intrinsic resistance existing in the terminal wires, the internal connections to the plates and in the plates themselves. Any current flowing in this resistance will generate heat, which could result in excessive temperature rise. Leakage resistance is that intrinsic resistance existing in parallel with the capacitor plates. Any voltage across the capacitor will cause current to flow in this leakage resistance with consequent generation of heat. With modern capacitors, except perhaps electrolytic types, series and leakage resistances are not usually a problem.
The voltage specified on any given capacitor is the "working voltage" but a test, peak or surge voltage may also be specified. Working voltage is the maximum continuous voltage that the capacitor will withstand for extended periods of time, although it is good engineering practice to assume an actual working voltage of about 75% of the stated value. The test, peak or surge voltage is that non- repetitive voltage that the capacitor will withstand for extremely limited periods of time.
The ability of all capacitors to maintain their values, pass current or withstand voltage is adversely affected by excessively low or high temperatures. Hence capacitors have temperature, as well as voltage and current, ratings. Electrolytic capacitors are particularly sensitive to temperature.
Large electrolytic capacitors used in smoothing circuits may have to pass several tens of amps of ripple current at the same time withstanding an applied DC voltage of scores, or even hundreds, of volts. A transmitter delivering 1000W into a 50ohm feeder via a DC blocking capacitor will require that capacitor to pass a current of 4.47A.
The last "parasitic" element to be considered is self-inductance. This appears in series with the capacitive element of the component and becomes increasingly important as the operating frequency increases, becoming of paramount importance in UHF and microwave applications, where a capacitor can actually become a resonant circuit.
Let us now consider the different types of capacitor that are available and their applications.
FIXED CAPACITORS
These types are categorised according to the dielectric used and most of these can be sub-divided into many different types.
Vacuum Dielectric
Fixed vacuum capacitors are basically non-adjustable versions of the vacuum variable (see below). Versions are available with capacities of up to 1000pF at working voltages of up to 50kV. They are capable of carrying RF currents in excess of 30A. Needless to say, these capacitors are extremely expensive. The photos below are reproduced with the kind permission of Dave Knight G3YNH.
Air Dielectric
Fixed air dielectric capacitors are rare and are normally limited to very high power transmitter applications. They usually consist of flat metal plates separated by glass or ceramic insulators. Working voltages up to hundreds of kV can be obtained, with capacity values of up to several hundred pF. The series resistance is extremely low and the leakage resistance is extremely high. Consequently, the Q is also very high. This type of capacitor would be a purpose built item for a specific application.
Paper Dielectric
Paper capacitors are now virtually obsolete but were very common prior to the advent of the more modern plastic film types in the early 1960s. They employed metal foil, to which the leads were connected, on either side of a thin paper strip. When a sufficient length of strip had been manufactured, depending on the required value of the finished capacitor, it was wound into a tube which was then inserted into an outer case, often of cardboard but sometimes consisting of a sealed metal tube. The entire assembly was then wax impregnated before the ends of the outer case were sealed.
Some more specialised types were housed in sealed metal cans, often fitted with terminals. These capacitors were sometimes manufactured as multiple units, where two or more individual capacitors were mounted in a single can.
They were used for general purpose coupling, de-coupling and smoothing purposes, with values ranging from about 1000pF to several microfarads. Working voltages between 100V and several kV were available. The value tolerance was usually ±20%. Some types suffered from an effect called "migratory placticisers", where the paper insulation becomes brittle, which causes it to crack and break down after prolonged use. It is never advisable to apply voltages greater than half the working voltage to this type of capacitor and, if used in a mains filter, the DC working voltage should be at least three times the applied RMS voltage, unless the component is specifically designed for AC applications.
This type of capacitor exhibits fairly low-Q and poor temperature stability. Even so, block paper capacitors housed in sealed metal cans and having values of 4µF or more, were often used instead of the much smaller but, at the time, much less reliable, electrolytic alternatives.
Motor start, motor run and power factor correction capacitors are usually metalised paper types, although special AC rated, reversible electrolytic types exist for motor applications. Metalised paper capacitors for these applications are usually housed in sealed metal cans. These capacitors are designed to carry fairly large AC currents and therefore have substantial terminals, rather than wire connections.
Plastic Film Dielectric
Plastic film capacitors employ a metal deposition, to which the leads are connected, on either side of a thin plastic film. When a sufficient length of strip has been manufactured, depending on the required value of the finished capacitor, it is wound into a tube, and either encapsulated in a plastic compound or covered in more film before sealing the ends. This type of capacitor has almost entirely replaced the paper dielectric type in coupling, de-coupling, timing and similar low to medium frequency applications. Specialised types are available for motor run and start, power factor correction, mains filtering, sample and hold, high stability RF tuning, high current pulse and many other applications. The plastic film can be mylar, polystyrene, polyester, polycarbonate or polypropylene, depending on application. PTFE is sometimes used as a dielectric but metal cannot easily be deposited on this material and its use is limited to very specialised applications. Values range from a few pF to several µF and working voltages range from around 50V to a few kV, depending on type and application. Value tolerances are normally better than ±10% for the higher values and better than ±2% for the smaller values. This type of capacitor exhibits medium to high-Q and has very good temperature stability, particularly in the lower values.
Ceramic Dielectric
Ceramic capacitors are available in three main types, namely tubular, disc and surface mounting. They two former types employ metallising, to which the leads are attached, on either side of the ceramic and both are available in values ranging from less than 1pF to 1µF, although higher values are sometimes encountered. The surface mounting versions are similar to the disc type except that the leads are replaced by metalled end caps. Normally, the temperature stability and Q decreases as the value increases. The higher values are normally used as de-coupling capacitors but the lower values can be used as the capacitive element in LC tuned circuits. Components with values less than 100pF are available with closely controlled temperature coefficients of between P100 and N750, including NP0 and are often used to compensate for temperature drift in other components. Value tolerances are normally better than ±20% for the higher values and better than ±2% for the smaller values. The higher value types have large, ill-defined temperature coefficients. Working voltages depend on type and value and are typically 63V for the higher values and up to 250V for the smaller values. Special types are available, such as discs for use in the high voltage circuits of video monitors and TV receivers. These types are available with working voltages up to 5kV.
Special types of ceramic capacitors have been designed for use in high power RF transmitters. These can have working voltages up to more than 30kV and can carry more than 50 amperes of RF without damage. These types exhibit high temperature coefficients and are therefore not suitable for use in stable frequency determining applications.
Bare disc capacitors without leads or encapsulation are also available. In this type, the plates comprise metallising on either side of the disc and they are intended to be soldered directly to PCB tracks or other components. The main use of this type is in UHF and microwave applications, where low series inductance is very important. Values range from a few pF to about 1000pF, with value tolerances better than ±10% and working voltages of around 100V.
Another variety of ceramic capacitor is the monolithic or "Monobloc" type. These are multi-plate ceramic components where the plates and connection wires, dielectric and overall casing are combined into a single, homogeneous block, hence the name monolithic, or "single stone". Similar comments apply to these types as apply to ordinary ceramic discs, except high voltage and high current types are not available. This type of capacitor is also available in surface mounting versions.
The feed-through capacitor is a special type of ceramic capacitor, whereby a lead may be passed through a screened plate or chassis. They consist of a tubular ceramic capacitor, where one lead is connected to the metallising inside the tube and is passed through the tube to provide a connection at each end. The other plate is formed by metallising on the outside of the tube, which is either soldered to a connection ring or to a threaded bush. Those with the connecting ring are soldered directly to the chassis, whereas those with a threaded bush are fixed with a nut. The purpose of this type of capacitor is to provide a convenient method of passing a lead through a metal plate and efficiently de-coupling it to earth at the same time. They are often used as the input and output connections to sealed, metal cased, filter units. The range of values available is usually limited to 100pF, 500pF and 1000pF, although other values are occasionally encountered. The working voltage is usually 100V but high voltage versions for mains filtering are also manufactured.
Some versions are available which incorporate ferrite beads. These are usually manufactured as pi-section filters but other configurations are sometimes found. All types are available as multiple units.
