The caps surrounding the LM7812 regulator?

[walid]

Hi

please look at the attached figure below:
u can see two caps around the LM7812
I go to datasheets and read about them,
After read that I felt I did not reach the result required. can you explain it more fore me
Thank you for your cooperation

[walid]

I am waiting for the answer, a little patience be useful.

[audioguru]

The regulator oscillates if its input is a high impedance. If the main filter capacitor is far away then the wires to it have inductace which is a high impedance. The input capacitor directly at the regulator prevents oscillation.

The output capacitor stops the output from ringing when the input voltage or the load current changes. It also smooths the regulator's transient response to those changes because the circuit takes time to respond.

[walid]

Hi guru
Thank u for reply
please what the differance between ringing and oscillating?
thanks

[audioguru]

Ringing is like a bell. It begins oscillating when the input voltage or output current changes then the amplitude gets less and less, then it is gone. Ringing is a sign of poor stability because it is almost oscillating.
Oscillation is a continuous signal.

[walid]

Hi guru

Tell me please, what could happen if we remove them, Are they very essential. I mean what the damage caused by the absence of these caps.

Second question :
If you have a 12V voltage regulator in a circuit. How must be the least value of the i/p DC voltage to get a 12-volt at the o/p?
thanks

[audioguru]

Tell me please, what could happen if we remove them, Are they very essential. I mean what the damage caused by the absence of these caps.

The datasheet says the input capacitor must be used and when. It also says the output capacitor stops certain problems. They are small and inexpensive so why not use them?

Second question :
If you have a 12V voltage regulator in a circuit. How must be the least value of the i/p DC voltage to get a 12-volt at the o/p?

The datasheet says "The input voltage must remain typically 2.0V above the output voltage" but some ICs need 3.0V. At 1A, the dropout voltage is typically 2.0V but then it is not regulating. It is typically a little less at lower currents.

[walid]

Hi guru

First, thank you very much

The datasheet says ............
The datasheet says ............

I swear I read it, but I do not understand, they put figures and curves and I want rules like what you learned me
Your answer almost picture more, While it is good to draw my attention to the Datasheet
thank you guru

[Kevin Weddle]

The chip designers don't have any idea where the regulator is going to be used. The operating parameters are also wide. It's nice to know that the regulator can be enhanced by adding additional capacitors, but I wouldn't throw them in without knowing.

[audioguru]

A voltage regulator IC has a very high internal gain. The external capacitors are use to increase its stability (so it doesn't oscillate) when wires from its input voltage source or to its load are inductive.

[Kevin Weddle]

I was also thinking that since the regulator current could be changing significantly, the phase could be changing, which would hinder the regulator's ability to regulate. A capacitor would then help the regulator to not be the cause of phase shift.

[Kevin Weddle]

A voltage regulator IC has a very high internal gain. The external capacitors are use to increase its stability (so it doesn't oscillate) when wires from its input voltage source or to its load are inductive.

Why must the gain be high within the regulator. If the load current brings the voltage down a volt, bringing it up a volt can't require too much gain, unless your reducing the feedback to the control element.

[audioguru]

The 7812 has a typical output voltage change of only 10mV when its input voltage changes from 14.5V to 30V.
Its output voltage changes typically only 12mV when its output current changes from 5mA to 1.5A.

A 7805 is even better because it has more negative feedback.

Its output resistance at DC is only 0.008 ohms.

There is an error amplifier circuit inside with a voltage gain of at least 10,000. If the gain isn't so high then the voltage regulation won't be as good.

[Kevin Weddle]

Sounds extraordinary, but the input to most regulators is a charged capacitor. All the regulator needs to do for the most part is not interfere with the capacitor voltage.

[audioguru]

You can buy a 7805 voltage regulator for $.60US each at Digikey. It has one power transistor, 16 small transistors, two zener diodes, two capacitors and countless resistors. It regulates well.

You can also buy a cheap TIP31 power transistor in the same case for nine cents more: $.69US.

[Kevin Weddle]

Do you suppose the circuit is designed with a load taken into account, or does an increased load just take away from the regulator?

[audioguru]

The load has nothing to do with a high gain regulator that has lots of negative feedback.
It works perfectly until the max load current is limited by the regulator.

Its output resistance at DC is only 0.008 ohms so the load resistance is much higher (1A at 5V= 5 ohms which is 625 times higher than the output resistance) and has a negligible effect on it.

[Kevin Weddle]

So the regulator really needs to regulate against the fluctuating input, and not worry about the 5 ohm load because it's nothing?

Passing 1A without affecting the regulator circuit at all seems farfetched. That is why I think that the regulator, which is a complicated DC circuit, directly regulates the load and worries little about the input.

The output resistance figure seems misleading. That regulator is doing everything it can possibly do to maintain 5volts at 1A. And in the same breath it has to maintain 5volts at 1mA

[audioguru]

So the regulator really needs to regulate against the fluctuating input, and not worry about the 5 ohm load because it's nothing?

Passing 1A without affecting the regulator circuit at all seems farfetched. That is why I think that the regulator, which is a complicated DC circuit, directly regulates the load and worries little about the input.

No. Read the datasheet.
A 7805 voltage regulator keeps the output voltage very close to 5V if the input voltage changes from 7.5V to 30V and/or if the load current changes from zero to 1.5A.

The output resistance figure seems misleading. That regulator is doing everything it can possibly do to maintain 5volts at 1A. And in the same breath it has to maintain 5volts at 1mA

The output resistance is spec'd on the datasheet at 0.008 ohms for a 7812 an is less for a 7805.
The regulator has high gain and a high amount of negative feedback like an audio amplifier. Many audio amplifiers also have a very low output impedance so that they can damp loudspeaker resonances very well. An audio amplifier with a damping factor of 500 at 8 ohms has an output impedance of only 8/500= 0.016 ohms.

[Kevin Weddle]

No. Read the datasheet.
A 7805 voltage regulator keeps the output voltage very close to 5V if the input voltage changes from 7.5V to 30V and/or if the load current changes from zero to 1.5A.

That's what I said. It regulates 5v with a 7.5V to 30V input. So this is what it must focus on because ....

The output resistance is spec'd on the datasheet at 0.008 ohms for a 7812 an is less for a 7805.

There is no load that can compete with that low of resistance. That means the load is nothing to the regulator up to 1.5A.

[audioguru]

There is no load that can compete with that low of resistance. That means the load is nothing to the regulator up to 1.5A.

The actual output resistance is about 1 ohm. The high gain and high amount of negative feedback make it function the same as if its output resistance is only 0.008 ohms.

With a load of 1.5A then its output voltage would drop 1.5V across its actual reistance of 1 ohm. The gain and negative feedback cause its output to rise almost 1.5V so the voltage drop is only 12mV.

[Kevin Weddle]

Okay. I think we weren't on the same page. About the design, the circuit values are chosen around an ideal circumstance. Do you think it was designed around an open load, or a mid value of about .75A. I wouldn't choose a 5A regulator over a 1.5A regulator for a 1A application.

[audioguru]

The load current doesn't affect the voltage regulator circuit. It works perfectly with zero current up to max current.
It also works perfectly with minimum input voltage up to its max input voltage.

[Herman the German]

The datasheet says the input capacitor must be used and when. It also says the output capacitor stops certain problems. They are small and inexpensive so why not use them?


The datasheet says "The input voltage must remain typically 2.0V above the output voltage" but some ICs need 3.0V. At 1A, the dropout voltage is typically 2.0V but then it is not regulating. It is typically a little less at lower currents.

Omit them if you like it hot! Like any oscillation (mechanic or electronic) it produces heat! Of course the poor regulator will some time die because of the tremendous heat inside. You can accelerate this process also omitting a heat sink.

Concerning the input voltage for a stable output three Volts is the best figure. Due to the "typically" built-in error two Volts will sometimes work, sometimes won't work. Electronically spoken this device will be a "perhaps-gate" using two Volts above the desired output voltage.