What is a decoupling capacitor (or smoothing capacitor as referred to in the link below)?
How do I know if I need one and if so, what size and where it needs to go?
This question mentions many chips needing one between VCC and GND; how do I know if a specific chip is one?
Would an SN74195N 4-bit parallel access shift register used with an Arduino need one? (To use my current project as an example) Why or why not?
I feel like I'm starting to understand the basics of resistors and some places they're used, what values should be used in said places, etc, and I'd like to understand capacitors at the basic level as well.
To block the D.C (bias), to have pure A.C signal, coupling capacitor is used. Reactance offerd could depends on frequency Xc=1/(2(pai)FC). Coupling capacitor are used between the stages of amplification of signal, to block the D.C exist with the amplified signal.
Basically its protects the Micro-controller (or any other Chip). If u need one or not its written on the data sheet. Mostly the recommended setup is that was one needs.
http://pdf1.alldatasheet.com/datasheet-pdf/view/27391/TI/SN74195N.html
On Page 4 there is the basic setup shown in the diagram.
Do not hesitate to ask if something is unclear.
Proper functioning of digital circuits relays on communicating logical signals at specified voltage levels. One of critical points to maintain these levels is power supply for integrated circuits. Different IC families have different tolerance on their supply voltage, but other important aspect is uniformity of supply voltage over all cooperating IC. First aspect is assured with proper power supply selection (or design), e.g. better than +/- 5% from 5V for TTL family as SN74195 that you mention. Second aspect (uniformity) is fulfilled by proper circuit (and PCB) design.
Main cause of non-uniformity of supply voltage among IC, all connected in parallel to power supply (that in the first look, if resistance is neglected, would give same supply voltage for every IC), is inductivity of the wires or PCB traces, and it may get so worse so to make circuit functionality unpredictable at high switching speeds. During high switching speeds (short turn-on and turn-off times) high current pulses are required by IC from power supply. But inductivity will resist to this required change and voltage at this particular IC will drop. To prevent this drop an capacitor is connected in parallel with power pins of IC, that will behave as local power supply at these short moments of time. Leads that connect this capacitor to IC pins should be as short as possible. To make this job easier some manufacturers produced IC sockets (DIL) with already included decoupling capacitor.
Rigorous calculations for these capacitors is complicated, but in early days there was as a rule of thumb to place by one 10nF-100nF ceramics capacitor near every IC, and one electrolytic capacitor of 1uF-10uF (preferred Tantalum one) between main traces that supply groups of 10-20 IC. Maybe circuit will work without half of them, or maybe without most of them, even without all of them, but never forget about Murphy's law.
When price is not of concern, put all of them. During rapid prototyping with breadboards usually we do not put any of them, but common speeds in these cases are in low MHz range. If everything seems to be OK but circuit is having trouble from time to time put them.
Decoupling capacitors are easily noticed visually in commercial boards, especially around high complexity ICs with many pins tens of them can be very close around it, or directly under it (e.g. PC processors), or under it at other PCB side.
Capacitors can be used for "decoupling" in various contexts, so there is not a single answer to the question. However, a typical decoupling example is to connect a capacitor from a DC voltage supply point to ground, for example from the VCC pin of a digital chip to the ground or 0V level. By doing this ripples that may be on the DC voltage, for example digital pulses, AC sine waves or noise, are largely bypassed to ground and thus filtered from the VCC line.
Just to prevent any misunderstanding, a "decoupling capacitor" is not a label for a specific type of capacitors. It is common to use capacitor values on the order of 1nF up to 100nF for the above mentioned decoupling, and for that purpose common ceramic capacitors are often used, as they conduct readily high frequency pulses or noise. Electrolytic or tantalum capacitors, which have typically larger capacitance - on the order of microfarads or hundreds of microfarads, are not good at bypassing the higher frequencies (hundreds of kHz and higher), but good for bypassing lower frequencies (several kHz and lower). Because these two types of capacitors are good for different frequency ranges, you may see both used together - say, a 0.01uF ceramic capacitor and also a 10uF electrolytic capacitor both connected to the same point and ground. In that case, they work in parallel and each one may bypass its own set of frequency ranges.
With respect to the power connections to an integrated circuit or really any active circuit element, I would add that the local decoupling capacitor is used to attenuate any ripple remaining AND to remove any noise the power trace may have picked up along its path. This cap is meant to shunt to ground any of this high frequency noise, immediately before reaching the circuit element.
Like Kristinn Andersen says, larger capacitors used in the power supply circuit are often not good at removing high frequency ripple and noise from the power trace. So that, and also induced noise, has to be managed. I've used both. A small ceramic cap in parallel with the each filter capacitor in the power supply, plus that local decoupling cap at each active device's power lead.
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering