Voltage Regulator Question

[brutus]

Hi,

In the data sheet for a chip (qt301 from qprox.com) I am using it says the following:

"If the power supply is shared with another electronic system, make sure the supply is free of spikes, sags, and surges. The supply is best locally regulated using a conventional 78L05 type regulator, or almost any 3-terminal LDO device from 3V to 5V".

Ok... I can't use 78L05 because the supply is at 5V. So I will need to use a regulator that gives 3.3V or 4V ouput. As always in electronics, there are HUNDREDS of different parts to choose from, and many don't have basic circuits that show you how to make a basic "fixed" voltage regulator (i think it's called fixed). Also, there a very few regulators that are called "fixed", so I'm guessing the "fixed" regulators have the capacitors in the IC already???

I believe if they are not designated as "fixed" I need to use two capacitors...

If anyone can point to a part and a circuit that would be very nice...

B.

 

[windoze killa]

If you are trying to produce a 5V supply and you can't use a 7805 because you only have a 5V input then why not just use that 5V supply.

 

[brutus]

Where do I say that I want to use a 78L05???

I say I dont. I want to use something like 78L033 or 78L04, except in a TO-92 package.

Anyway... read my post if you're interested in replying...

 

[audioguru2]

A "fixed" voltage regulator has its output voltage fixed to one voltage. An "adjustable" voltage regulator can have an output voltage adjusted with a pot or have almost any voltage set with two resistors. You should use an input capacitor and an output capacitor with both types.

They say to use a LDO (Low DropOut) regulator. An ordinary 3.3V regulator needs an input voltage of at least 5.8V to regulate. A 3.3V LDO regulator needs an input voltage at least only about 3.8V. 

 

[brutus]

Thanks for the reply.

I did find a circuit. It uses a 0.1uF for the input film capacitor (poly...ester?), and a 2.2uF electrolytic capacitor for the output.

Final question:

Do I really need to use an electrolytic cap? They are too big. I would prefer to use something smaller, maybe another film or ceramic...

Having a stable voltage is hugely important, so I don't want to use something unsuitable.

Thanks again,
B.

 

[audioguru2]

Many low dropout regulators must have at least 100uF on their output.

 

[brutus]

Hmmm.... 100uF seems really high...

Attached is a JPEG of the "test circuit" where they use 2.2 uF instead of 100uF. The whole pdf document can be found at:

http://www.st.com/stonline/books/pdf/docs/2573.pdf

Am I missing something?

B

View attachment 39394

 

[audioguru2]

Am I missing something?

It is a nice regulator, but:
Can you buy a 2.2uF capacitor that is guaranteed to be at least 2.2uF?
Did you notice that every example circuit in the datasheet has a 10uF output capacitor?
Did you notice the lack of a graph showing the overshoot response with various output capacitors, as shown on other regulator datasheets?

 

[brutus]

>It is a nice regulator, but:
>Can you buy a 2.2uF capacitor that is guaranteed to be at least 2.2uF?
>Did you notice that every example circuit in the datasheet has a 10uF output >capacitor?

Ok... So I'll use a 10uF outpur cap... Solves the overshoot problem maybe?

I still want to use as small a cap as possible. Can I get away with a metallized film cap or maybe a X7R ceramic cap?

_OR_

is electrolytic the only way to go for the output cap?

???

 

[audioguru2]

Ok... So I'll use a 10uF output cap... Solves the overshoot problem maybe?

We don't know because the datasheet doesn't have a graph like other companies have showing transient behaviour.

I still want to use as small a cap as possible. Can I get away with a metallized film cap or maybe a X7R ceramic cap?

I have not seen a small 2.2uF film cap. I have not seen a ceramic cap with a value as high as 2.2uF.

_OR_

is electrolytic the only way to go for the output cap?

The example circuits in the datasheet show a polarized cap. The Equivalent Series Resistance of the cap is critical, a resistor in series with the cap might be necessary to prevent oscillation.
It is too bad that ST Micro don't talk about these things on their datasheet like National Semi does.

 

[MP1]

Ok... So I'll use a 10uF outpur cap... Solves the overshoot problem maybe?

I still want to use as small a cap as possible. Can I get away with a metallized film cap or maybe a X7R ceramic cap?

_OR_

is electrolytic the only way to go for the output cap?

I assume you are trying to use as little board space as possible? I design with "surface mount" tantalums instead of "through hole" Electrolytics in such a case. Solves many other problems as well.
I would not recommend metallic or ceramic caps for power supply bypass.

Hope that is helpful.

MP

 

[indulis]

John

You can get 100uF ceramic cap's nowadays if you can live with a low voltage.

2.2uF caps are a dime-a-dozen and nothing special... check TDK or Murata.

 

[brutus]

>I assume you are trying to use as little board space as possible? I design with "surface mount" tantalums instead of "through hole" Electrolytics in such a case. Solves many other problems as well.

You assume correctly, except it's not only board space but I also need a very low profile, like 3-4 mm. They have 2.2uF axial electrolytic caps that I might use, but your idea of using surface mount tantalum caps sounds like a good idea. I will look into it...

>I would not recommend metallic or ceramic caps for power supply bypass.

Just curious... why? Does it have to do with the speed that it takes to charge and discharge?

Thanks MP,
B.

 

[brutus]

Hi,

I read an article about LDO and mistakes people make that result in an unstable output.

http://www.national.com/nationaledge/jul02/article2.html

In that article (which is way over my head) they mention that tantalum and ceramic are not compatible (or something).

I will post the two paragraphs from my pdf again and ask a question after:

"If the power supply is shared with another electronic system, make sure the supply is free of spikes, sags, and surges. The supply is best locally regulated using a conventional 78L05 type regulator, or almost any 3-terminal LDO device from 3V to 5V.
For proper operation, a 0.1uF or greater bypass capacitor must be used between Vdd and Vss; the bypass cap should be placed very close to the device pins."

In the link article (www. national.com) it talks about interference (oscillation) caused by ceramic caps. If each of my chips will have it's own regulator, maybe a bypass cap between Vdd and Vss is not necessary? Maybe the 10uF cap I will use for the regulator output will be enough?

Thanks.

 

[audioguru2]

On the datasheet for the low-dropout regulator you selected, there is a listing of the allowable ESR of the output cap. A ceramic cap has an ESR much too low and will cause the regulator to oscillate unless a low value resistor is added in series with it. Even additional ceramic caps on the output cause trouble.

 

[brutus]

> Even additional ceramic caps on the output cause trouble.

Ok... Does this mean that any cap that is between the supply voltage (in this case the output of the regulator) and Ground can cause problems?

_BUT_

Capacitors that are in other parts of the circuit (are not directly connected to the supply voltage and GND) will not cause problems, and can be film or ceramic???

 

[audioguru2]

Does this mean that any cap that is between the supply voltage (in this case the output of the regulator) and Ground can cause problems?

The datasheet for the regulator shows the safe range of an output capacitor's ESR. Ceramic caps have an ESR much too low for them which would cause high frequency oscillation.

Capacitors that are in other parts of the circuit (are not directly connected to the supply voltage and GND) will not cause problems, and can be film or ceramic???

We are talking only about capacitors from the output of a low dropout regulator to ground.

 

[brutus]

Hi,

I just spoke to the guys that make the chip (that I need to 'voltage regulate') as well as a tech guy from digikey who knew about this issue.

They both said that this issue was true in the past (1980's 1990's) but is no longer an issue at all because the LDOs have internal components to deal with exactly that.

So as far as I can tell, YES you can use ceramic caps on the LDO output.

Just thought I would post my findings...

 

[audioguru2]

I just spoke to the guys that make the chip (that I need to 'voltage regulate') as well as a tech guy from digikey who knew about this issue.

They both said that this issue was true in the past (1980's 1990's) but is no longer an issue at all because the LDOs have internal components to deal with exactly that.

So as far as I can tell, YES you can use ceramic caps on the LDO output.

The March, 2005 datasheet of the LE00 low-dropout regulator is pretty new and says the ICs should have a bypass capacitor with an ESR from 0.1 to 10 ohms. Every schematic shown has a polarized output bypass capacitor.
Ceramic caps have an ESR much lower, and are not polarized.
Try a ceramic output bypass capacitor and tell us if the regulator oscillates.