So far, we have chosen a component and assembly method, and hence the component specification as regards compatibility with the assembly processes. We now have to look at other aspects of the component which one has to consider when ordering parts, specifically values, tolerances and ratings, and at how you might be able to judge from visual appearance whether or not you have the right part.
Whilst it is possible to buy custom passive components with any desired value, standard parts are manufactured in ranges of preferred values. The reason behind this is that the method of manufacture of many components results in a spread of characteristics and it is expensive to make/select to accurate values, whilst most circuits will work adequately with loose tolerance parts. It makes good economic sense to make and stock only a limited range of values.
In the 1870s, a French army officer, Colonel Charles Renard, tried to rationalise the procurement of ropes for use with military balloons! He derived a preferred number system (the ‘Renard system’) using a geometric progression based on the number 10. The numbers created from the 5th (10th, 20th, 40th . . . ) root of 10, rounded off to one/two decimal places, created the R5 (R10, R20, R40 . . . ) series. Successive terms of the series differ by approximately 60% (25%, 12%, 6% . . . ).
Still in common use (for example, for fuse ratings) are the R10 series (Table 1), which for the first three decades is:
Resistor manufacturers adopted a modified system, based on the 6th (12th, 24th, . . . ) roots of 10, creating the E6 (E12, E24, . . . ) ‘Electrical’ series, corresponding to ±20% (±10%, ±5%, . . . ). The sequences of values, given in Table 2, ensure that the entire production falls within one or other of the values. The tighter the tolerance, the more preferred values there are in the series: components near top limit for one value overlap with those near bottom limit for the next highest value.
E6 Series (six values per decade)
|Used for 20% components. Values in bold are the E3 series|
E12 Series (12 values per decade)
|Used for 10% components|
E24 Series (24 values per decade)
|This series contains the following in addition to the above E12 values:|
Only components with tolerances of 5% or better are available in this series.
Precision components frequently use an E96 series, which has three values between each E24 series value, to give a total of 96 values per decade.
These preferred value series have been harmonised internationally, and are now used for parameters other than the original resistance, capacitance and inductance. For example, it is not uncommon to find capacitors with a 6.3V rating.
Tolerances traditionally use 1%, 2%, 5% values and their decade multiples and sub-multiples. There is, however, a tendency for capacitor manufacturers to use letter codes to show tolerances (Table 3).
Note that, for low capacitance values, tolerances are given in ±units rather than ±percent, and different code letters are used.
Voltage ratings are usually given as part of the code marking on electrolytic, AC-rated and high voltage capacitors. Otherwise, ratings must always be verified from the manufacturer’s data sheet. Note that any voltage rating quoted is probably the maximum operating voltage at 25ºC, and good designers will select parts with a higher rating than the in-circuit voltage, especially where the application involves extended temperature operation or requires high reliability. For electrolytic capacitors, the voltage rating will be for application in the intended direction – reverse ratings are very much lower.
Power and current ratings may be estimated from the physical dimensions of the part, but should always be checked from the manufacturer’s data sheet. As reliability reduces with increased body temperature, it is unusual for components to be operated at their maximum ratings.
All components drift with temperature, and are stated to have a ‘Temperature Coefficient of Resistance’ (or Capacitance, etc.), usually abbreviated as TCR (TCC, etc.). These parameters are normally quoted in parts per million per degree Celsius – a resistor with a ±250ppm TCR will vary by up to 0.025% for each ºC of temperature change. Depending on the materials and construction, temperature coefficients may be linear or non-linear.
Similarly, component characteristics such as capacitance may vary with applied voltage. Where this is quoted, it will again be as per cent or parts per million per volt.
The amount of information about a component (manufacturer, type, value, tolerance, quality screening level, etc.) printed on the device varies enormously. As components have become smaller, the area available to display this information has diminished, until it has become vanishingly small, as in the case of the smaller sizes of chip capacitors, where no data is displayed.
At the same time, component density has increased, and the former common practice of silk screening component identification legends (R20, C52, U201 etc.) on the PCB has declined through lack of space and the move to automated rather than hand assembly.
This problem of identifying the value of small components was recognised long ago, when carbon composition and, later, metal film axial resistors were given a colour code (Table 4) to identify resistance value and tolerance. The information is presented as a series of four (optionally five) coloured rings (Figure 1).
|Colour||Value digit||Value multiplier||Number of zeros|
|black||0||1 ( =100 )||0|
Similar codes have been used for capacitors, but tolerance band colours may be different, units will be pF or µF (depending on construction), and multipliers violet/grey/white rather than silver/gold.
Curiously, wire-wound and high stability metal film resistors usually bear alphanumeric data rather than colour coding. Both for resistors and capacitors, decimal points are generally not used, but indicated instead by the position of the multiplier. For example, 1M5 would be used for 1.5 MW, 19n2 for 19.2 nF, µ22 for 0.22 µF and R047 for 47 mW.
A resistor with a white body generally indicates a non-flammable resistor, and a blue body a fusible resistor. When a fusible resistor overheats, it cuts the current in the same way as a fuse, and both types are designed not to catch fire when they overheat. Never replace either type with a normal resistor, because this would create a fire hazard should a circuit fault occur. Otherwise the colour of the resistor body has no universal meaning, although some manufacturers use the body colour to differentiate between components with different temperature coefficients.
Chip resistors are normally marked with three digits (representing the first two value digits plus a multiplier digit). For example, 334 indicates a value of 33.104 W, or 330 kW. Other details must be obtained from the manufacturer’s data. Where non-standard values are needed for close tolerance resistors, a fourth digit may be added. For example, 1942 denotes a value of 19.4 kW.
Attempts have been made to extend the axial resistor colour code to small value capacitors (chiefly leaded radial types), using coloured spots on small components instead of rings, but this has not become standard practice. Great care should be taken, because there are differences between the codes for capacitors and resistors!
Normally electrolytic or plastic/paper capacitors are marked with capacitance value and working voltage, in plain text and in a single colour, such as 10 µF 10% 100 V. A '–' or '=' symbol following the voltage indicates a DC rating, and ~ that the capacitor is rated for AC operation. In addition, the polarity of tantalum and aluminium electrolytic capacitors is indicated, usually by a + sign at the anode end.
However, chip ceramic capacitors normally have no markings at all – it is necessary to refer to the container in which the chips are packed for the appropriate data. This does not help if it is necessary to verify on a completed assembly whether or not the correct value has been fitted! There has been pressure from the automotive industry to use laser-marked parts, but the process is not easy, given the small body sizes, and there is also a danger that the cut track may penetrate into the active area.
Manufacturer part codes are generally descriptive, a common form being:
XXXX X XX X X
The first four digits are the size, for example 0805, but may be in inches or metric, depending on maker. The next letter(s) indicate the dielectric type (for example A for NPO material). Then follow the first two digits of the value in pF, and a multiplier (102 represents 10.102 pF, or 1 nF) and tolerance letter (as Table 3). Extra letters may give information on packaging, voltage rating and termination material.
There are no general codes for marking other components, although manufacturers may provide basic details, e.g. potentiometer resistance and/or type number. Safety critical components may also bear the approval marks of national laboratories: VDE (German) and UL (USA) approval markings are frequently seen.
Ideally these should be marked with:
Even more important is some form of pin identification to ensure correct orientation on placement or insertion. For ICs, this usually takes the form of a dot or indentation next to pin 1, but practices vary.
For transistors and diodes, the package standard defines allowable lead sequences — for example, emitter, base, collector for a bipolar transistor, and source, gate and drain for an FET. In some cases, however, the lead positions may be different: SOT-23 transistors are frequently available with base and emitter connections in either ‘normal’ or ‘reversed’ configuration (Figure 2).
The amount of information is dependent on the available surface area of the component - cylindrical diodes may simply be marked with a ring at the cathode end, whereas most ICs carry sufficient data to enable identification of the device. However, for most discrete components, it will be necessary to consult manufacturer’s data books to identify a full device type from a specific code. SOT styles of package are an example of where coding needs to be interpreted , because the space available for easily visible coding is usually only sufficient for two or three alphanumeric characters, whereas transistors generally require both device type and hFE selection to be shown. Note, however, that where laser coding is used, it is possible to use additional characters to contain an encrypted version of the date of manufacture. These will often be in smaller type and/or at right angles to the main code.
IC markings can be split into four parts:
1 The prefix identifies the manufacturer - usually by a one to three letter code. However, a manufacturer may have several prefixes. 2 The device code identifies the specific IC type. 3 The suffix indicates package type and temperature range. Note that each manufacturer has its own set of suffices, subject to frequent modification. There are four common temperature ranges for devices, given below with typical applications:
|-65°C to +150°C||aerospace|
|-55°C to +125°C||military|
|-25°C to +85°C||industrial|
|0°C to +70°C||commercial|
4 The date code is normally a four digit code - the first two digits indicate the year, and the last two the week number. A date code of 9921 for example would indicate the date of manufacture as Week 21 of 1999. Some devices, however, are date coded in a form known only to the manufacturer.
Although some custom parts carry adhesive labels, the two most common coding methods are applying a marking ink to the device, and selectively abrading the top surface layer by particle blasting or (more usually nowadays) by laser machining.
Ink marking can be carried out by:
The quality of transfer printing can be poor, and the resolution is only capable of handling limited information. Pad printing has been an advance, and ink jet printing is a fast, flexible method capable of excellent results.
A variety of inks may be used, depending on the body material and the process chosen. The most common are epoxy or phenolic based and are cured by heat or UV radiation. Coding will frequently be carried out in two passes, the first applying generic information, and the second pass adding batch or unit detail.
A common problem associated with ink marking is the smearing or removal of the ink on subsequent handling, even when the ink is nominally ‘cured’. This can sometimes be alleviated by the choice of ink, or by improving the drying and curing process. More often, the problem is related either to contamination of the upper surface of the moulded part or to its texture, with matt or textured finishes giving improved results.
In laser marking, the information is ‘burnt in’, usually with a C02 or YAG laser which scans the surface to be abraded. This has the advantages that the coding is carried out immediately without any post-cure, and is indelible and that the process is very flexible, with the possibility of adding computer-generated unique identity markings. The technique can be used on all kinds of parts, including metal surfaces.
The major problem associated with laser marking is the relatively poor contrast between the exposed area and the background, the coding having only a slightly lighter and less reflective appearance than the black body of an integrated circuit. Although special moulding compounds have been developed to enhance the contrast, these do not give the same readability that can be achieved easily with white ink!
Look again at the sample board which you first examined in Introduction to components, examining particularly the coding on the components.
If you are able to examine any reels of components, record the part information given on the reel and compare it with any code which appears on the part. How does the amount of information not included in the part code vary with device type and size?
Never take components for granted:
Do not assume that you understand every aspect of their construction.
Do not assume that components that are called the same thing are actually made the same way, have the same functionality, or will withstand the same assembly processes.
Check when copying across package data from another manufacturer – the packages are likely to have different dimensions, even if they have the same JEDEC code.
And always read the data sheet!