linear voltage regulators design fault

Thread Starter

denison

Joined Oct 13, 2018
328
Hi, I came by accident on a big fault in linear voltage regulators in designing a circuit. I decided to use a fixed 24 volt regulator for my circuit. I was previously using a simple 24v Zener power supply but decided I needed better regulation.
I picked the L7824CV. The specifications for this show that Ro=28 million ohms. Looking up what this means on the internet is that it is the resistance between the output and ground of the chip. However on checking this resistance directly on the chip it is approx. 20k ohms. Because of this there is a connection of 20k between the 24v line and ground on the circuit. Therefore any resistor connected across the power line is in parallel to this resistance.
This has completed destroyed the functioning of my circuit as compared with the Zener power supply. If the connection was 28 million ohms this would not cause any problem in my circuit as this is close to infinity.
Either I don't understand the meaning of Ro or the chip is faulty. It is also easy to see that adjustable linear regulators would also be loading the circuit with this parasitic resistance. I have tried a 4.7uf across the power lines but this made no difference.
Does anyone know of a regulator that does not have this problem?
 

dl324

Joined Mar 30, 2015
16,839
Welcome to AAC!

A voltage regulator will provide much better regulation than a zener.

Where are you seeing this 28M ohm output impedance? How can a voltage regulator be loading your circuit? Post a schematic.
 

dendad

Joined Feb 20, 2016
4,451
Put a circuit up of you power supply. I think you have misunderstood the design by the sound of what you are saying.
 

ebp

Joined Feb 8, 2018
2,332
As Veracohr pointed out, you have misread the value by a factor of a billion and it more or less is a statement of imperfection of regulation.

All voltage regulators are going to have resistors between the output and ground. These are the "sampling" resistors that provide the feedback to the amplifier that controls the regulation. In a micropower regulator the resistance may be quite high, but the in the 7800 series it is moderate - and unspecified.

A typical 1 W zener is going to have an impedance of only a few tens of ohms at a current of a few milliamps.

Why is the output to ground resistance of a regulator destroying the operation of your circuit? On the surface, it makes no sense whatever. I've used three terminal regulators, both fixed and adjustable, in a myriad of circuits and never once had occasion to be concerned about the resistance between the output and ground.
 

crutschow

Joined Mar 14, 2008
34,280
I picked the L7824CV. The specifications for this show that Ro=28 million ohms. Looking up what this means on the internet is that it is the resistance between the output and ground of the chip. However on checking this resistance directly on the chip it is approx. 20k ohms.
Unless you know specifically how the chip is designed, a resistance measurement at the terminal of an unpowered device generally tells you little about the operation of the device when powered.

In this case you also misinterpreted the value of the specified resistance and the meaning of Ro.
Ro=28 milliohms and it's the equivalent active (operating) output resistance as determined by the internal regulation feedback loop of the regulator.
Basically it's a measure of the change in output voltage for a change in output load current.
For example, the output voltage will change by 28mV for a 1A change in output current.
 

Papabravo

Joined Feb 24, 2006
21,158
Whenever you think you have found a fault in a product that is used world wide in industrial quantities it might be time for you to fall back.
 

Audioguru

Joined Dec 20, 2007
11,248
A Zener diode is a SHUNT regulator. A voltage regulator IC is a SERIES regulator. A shunt regulator prevents the loaded output voltage from being externally driven too high. A series regulator prevents the loaded output voltage from going too low.
 

WBahn

Joined Mar 31, 2012
29,976
Hi, I came by accident on a big fault in linear voltage regulators in designing a circuit. I decided to use a fixed 24 volt regulator for my circuit. I was previously using a simple 24v Zener power supply but decided I needed better regulation.
I picked the L7824CV. The specifications for this show that Ro=28 million ohms. Looking up what this means on the internet is that it is the resistance between the output and ground of the chip. However on checking this resistance directly on the chip it is approx. 20k ohms. Because of this there is a connection of 20k between the 24v line and ground on the circuit. Therefore any resistor connected across the power line is in parallel to this resistance.
This has completed destroyed the functioning of my circuit as compared with the Zener power supply. If the connection was 28 million ohms this would not cause any problem in my circuit as this is close to infinity.
Either I don't understand the meaning of Ro or the chip is faulty. It is also easy to see that adjustable linear regulators would also be loading the circuit with this parasitic resistance. I have tried a 4.7uf across the power lines but this made no difference.
Does anyone know of a regulator that does not have this problem?
How has this completely destroyed the functioning of your circuit?

More to the point, WHAT circuit has it destroyed the functioning of?

Have you tried using the regulator as recommended by the datasheet?

There is very little we can do to help you understand what is going on with what you are doing unless you show us what you are doing.

What manufacturer are you using? The STM part has an Ro typical spec of 17 mΩ -- that's 0.017 Ω.
 

atferrari

Joined Jan 6, 2004
4,764
Hi, I came by accident on a big fault in linear voltage regulators in designing a circuit. I decided to use a fixed 24 volt regulator for my circuit. I was previously using a simple 24v Zener power supply but decided I needed better regulation.
I picked the L7824CV. The specifications for this show that Ro=28 million ohms. Looking up what this means on the internet is that it is the resistance between the output and ground of the chip. However on checking this resistance directly on the chip it is approx. 20k ohms. Because of this there is a connection of 20k between the 24v line and ground on the circuit. Therefore any resistor connected across the power line is in parallel to this resistance.
This has completed destroyed the functioning of my circuit as compared with the Zener power supply. If the connection was 28 million ohms this would not cause any problem in my circuit as this is close to infinity.
Either I don't understand the meaning of Ro or the chip is faulty. It is also easy to see that adjustable linear regulators would also be loading the circuit with this parasitic resistance. I have tried a 4.7uf across the power lines but this made no difference.
Does anyone know of a regulator that does not have this problem?
Show the actual schematic.
 

atferrari

Joined Jan 6, 2004
4,764
Does anyone know of a regulator that does not have this problem?
All what I have used to build my bench's PSU providing 5 different fixed voltages for each polarity.

Your remind me the hundreds of novices starting with micros, claiming they have a bug of sorts. Posts about fixed regulators compete in popularity with those about LCDs not working in the first try.
 
Top