Typical 0.1 μF bypass capacitor value...

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
User avatar
Externet
Posts: 1798
Joined: Fri Feb 01, 2002 1:01 am
Location: Mideast USA
Contact:

Typical 0.1 μF bypass capacitor value...

Post by Externet » Fri May 04, 2018 5:24 am

Most data sheets recommend 0.1μF bypass capacitors next to power IC pins. They smooth any sagging of the supply for the current the ICs draw in normal operation, and absorb transients generated elsewhere in the circuits.
What if that capacitor value is raised; would it be 'better' or non-detrimental ? Is the 0.1 μF actually a minimum value ? What if 1 μF was used instead ?
- Abolish the deciBel ! -

User avatar
CeaSaR
Posts: 1626
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Typical 0.1 μF bypass capacitor value...

Post by CeaSaR » Fri May 04, 2018 5:56 am

As you know all to well, they are being used for frequency dependent situations.

The small (0.01 + -) next to the IC power pin is used to bypass high frequencies to ground, preventing them from entering the IC and causing all sorts of problems.

The larger larger ones (~10 and up) are used to reduce power supply ripple and add a small amount of power boost in times of great demand when the load can draw down the supply voltage.

That's why you tend to see the bypass near the IC and a smoothing cap near the power supply on the schematic, even though they may be physically close on the board.

CeaSaR
Hey, what do I know?

dyarker
Posts: 1651
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: Typical 0.1 μF bypass capacitor value...

Post by dyarker » Fri May 04, 2018 9:06 am

Depends.

Replacing a 0.1uF ceramic disk capacitor with a 1.0uf ceramic disk capacitor would be better. But, probably over-kill for not much improvement in performance, more expense, and more PCB space.

Replacing a 0.0uF ceramic disk capacitor with a 1.0uF electrolytic capacitor could (repeat could) actually be worse.

The different types of capacitors has different frequency reponse. Ceramic disk being closest to theoretical "pure" capacitances (mica and vacuum might be closer). Other types lose "effective" capacitance faster as frequency goes up than ceramic disk.

Cheers,
Dale Y

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Typical 0.1 μF bypass capacitor value...

Post by Robert Reed » Fri May 04, 2018 6:39 pm

The 0.1 cap is generally used whether you need it or not. Assuming you are using MLC ceramics, its good practice to install them with out much after thought. The bypass caps value depends on the rise time and bandwidth that the particular IC is going to deal with, but just as important is the trace/wire length that connects it to the chip terminals. The "0.1" bypass cap can vary in value by as much as +/- 30 % in the lower grade values. In the XR series its closer to +/- 10%. Then at the top of the line is the COG series at close to +/- 0%. In my opinion these are the best of any type of capacitor whether using it for RF or audio however the max values here stop short of 1 uF and are more costly. All capacitors have a self resonant frequency (SRF) and the larger the cap the lower the SRF. after that point they start to be come inductive. Some would say that they are useless beyond that point, but I tend to argue that regardless of reactance they still provide a low impedance path to ground for a moderate increase in frequency. In general the 0.1 cap will have an SRF at a few MHz.
As you continue go higher in frequency the values have to continually be smaller for best bypassing and 1000 or less is not uncommon in the VHF band. There have been times when prototyping circuits in the 500 MHz range I have used 100 pf and less where nothing else would do the job. If you are working only in the audio range or maybe up to 1 MHz, just go with the 0.1 regardless of the circuit design. That's assuming that any pulse waveforms in that circuit don't have super fast rise times like high speed comparators, etc. The internet is loaded with cap manufacturers info on SRF's of their products and is well worth the time spent for a primer on this subject.

User avatar
haklesup
Posts: 2843
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: Typical 0.1 μF bypass capacitor value...

Post by haklesup » Tue May 08, 2018 2:34 pm

for most digital circuits the smaller ceramic caps go as close to each IC power pin as possible and shunt higher frequency transients generated inside the IC from getting onto the power lines. Additional larger Electrolytic caps will go at the edge of the PCB where the power supply enters to shunt larger energy transients coming from supplies and still larger caps at the output of the supplies to filter ripple (if applicable). Many high performance chips will have preferred layouts for the filter caps and signal escaper routs right in the datasheet. The higher the frequency the more important layout is and making traces short, avoiding vias and decoupling supply to supply become actually necessary or else you risk data loss. Analog parts don't necessarily follow this rule of thumb and there is often guidance in the datasheet. if you are doing any 5V to 3.3V level shifts, this becomes even more important as the 3.3V parts can read in bad bits from the noise from the 5V parts. That's trickier to fix, best solution is don't route these traces near each other.

Post Reply

Who is online

Users browsing this forum: No registered users and 8 guests