Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Purpose Of Parallel Tiny Parallel Caps

Status
Not open for further replies.

meowth08

Member
Hi.

I'm wondering why many of the circuit designs have tiny capacitors connected to the power supply.
Sometimes, there are a number of (1uF) + a number of (.1uF) then another number (.01uF) connected to a single supply.

What is the rule/standard in choosing which values in capacitance should we tie to the supply? Computations? Is there any?
Why connect many small values in parallel instead of few larger ones?
 

Attachments

  • supply capacitors.jpg
    supply capacitors.jpg
    18.9 KB · Views: 317
They are decoupling capacitors, the supply electrolytics, although much larger in μf, often posses inductance value which can be significant, the smaller capacitors ensure that the +Ve & -ve supply is decoupled or essentially the same potential as far as the circuit signal is concerned.
Although shown in a group, they are each placed immediately at signal IC's etc.
Max.
 
In addition to the answer from MaxHR...

What is the rule/standard in choosing which values in capacitance should we tie to the supply?
The generally accepted "rule of thumb" is to connect a 0.1uF or 0.01uF capacitor as close to the supply pins of each IC as possible.

Computations? Is there any?
None that I know of.
I have seen technical articles which went into great depth analysing the effects of different values of capacitance in parallel and the effects on their combined self resonant frequency, but at a practical level..
Just fit the capacitors and be happy that it works well.
If the job is cost sensitive, once it is working try removing some of the capacitors and see what you can get away with.:eek:

JimB
 
The red line if the impedance of a 10uF ceramic capacitor. I added a green line for a larger cap and a yellow line for a smaller cap.
If you parallel a 100uf and 10uf and 1uf you will get a line that starts down on green, hops to red for the 2mhz area and jumps to yellow for 8mhz and above. So when you parallel caps you get the low frequency benefits of the large cap and the high frequency benefits of the small cap.
upload_2015-6-24_19-4-19.png
 
meowth08,

Notice the names on the caps in your image?

On a lot of schematics or prints, they do not show the decoupling capacitors on the chips. The prints are too busy(crowded) there. So, they show them at the supply rail.

The label of the cap shows you the area of the board or chip it is located at.

Does that help?
 
Those small caps are placed close to the ICs to minimize the effects of trace inductance
Even short traces can have a significant impedance at the high frequency currents generated between the power pins and ground pins by high speed ICs.
If those currents aren't decoupled near the pins, they can cause disruption of the circuit operation due to the voltage spikes generated by these currents across the trace inductance.
 
Small value caps or often placed directly in parallel with the large main smoothing capacitors for the same reason which often appear paradoxical to the newbie.
Max.
 
Hi,

In addition to the other great replies here i'd like to add a little also.

Decoupling caps are used for a variety of purposes, some very specific. For general supply filtering they may be there just because they can handle the high frequency currents better than the larger electrolytic capacitors. For TTL logic families each chip or group of chips had to have a small capacitor because the output stage was partly made up of a totem pole output which would generate large spikes on the power supply lines. Placing a cap close to the chip would nip that right at the source of the problem which would reduce the effect the spike would have had on other chips, which could interfere with the timing of the circuit.
Another specific use for example is with the 555 timer ic, where the 0.01uf cap is recommended to keep noise out of the internal circuit which could cause a fluctuation in the output pulse timing. In this case the cap is not on the power supply line (although one might be good there too) but on the control voltage input, which we dont want to change due to noise pickup.

So there are a number of uses for these small capacitors, some a little more general and some very specific.
 
Two interesting faults solved today. Both TDA2003 sound IC's in Chinese TV chassis... two sets.

Supply Voltage around 18VDC. According to datasheet....quiescent Current should be around 44Ma....not so.
They were drawing around 400 to 500 Ma and rising as it heats up.....no speakers connected. But got as hot as hell without working.

Turns out the 100nF Cap working in Parallel with the 470uF smoothing Cap from the SMPS supply was open. So it was oscillating and getting angry and getting as hot as hell.

Around an Hour to solve...worth the effort.

This fault is going to make easy money here. There are many sets like these flooding the Market here. Sooner or later the sound will sound wonky.

I will say the TDA2003 is one tough *******.....board is black/burnt under it...got as hot as hell...still sound though. But distorted.

Diagram here of this lovely chip: (And I know it's an old design:))
 

Attachments

  • tda2003.pdf
    222.4 KB · Views: 290
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top