Voltage Regulator Question

bertus

Joined Apr 5, 2008
22,266
Hello,

Every voltage regulator will do this.
An 78XX fixed voltage regulator will out put XX volts with any input between XX+3 to 30 Volts.
The efficiency will be worse when the input voltage gets higher.
The voltage difference * output current will be dissipated as heat.

A buck regulator will be more efficient, as it will change the pulse width when the input voltage changes.

Have a look at the data sheets given in this post created by MrChips:
http://forum.allaboutcircuits.com/showpost.php?p=448194&postcount=8

Bertus
 
Last edited:

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Thanks Bertus :)

I looked at some of the datasheets that Mr Chips has linked us to, and I don't understand how I am supposed to adjust the output voltage. Do I have to add the extra circuitry around the voltage regulator like in their sample applications they have and vary the potentiometer they have there that Iadj goes through?
 
Last edited:

Sampson

Joined Feb 18, 2013
4
I was using a 7805 voltage regulator in one of my projects but it created too much heat because of my variable input. (8V to 40V). I ended up using a buck style regulator and Yes, they do require quite a few extra parts to properly configure the device. However, one company has a web-site that allows you to enter your requirements and it will create a part list and schematic to get you up and running. (Worked great for me!)
It is called Simple Switcher and this is the site:
http://www.ti.com/ww/en/simple_swit.../index.html?DCMP=simple_switcher&HQS=switcher
 

spinnaker

Joined Oct 29, 2009
7,830
Thanks Bertus :)

I looked at some of the datasheets that Mr Chips has linked us to, and I don't understand how I am supposed to adjust the output voltage. Do I have to add the extra circuitry around the voltage regulator like in their sample applications they have and vary the potentiometer they have there that Iadj goes through?
You need to first define your needs. The 7805 that Bertus mentioned is a fairly common 5V regulator. It is simple and requires very few external components.

The LM317 is an adjustable regulator. You can choose to use a potentiometer or fixed resistor.

Keep in mind that these regulators will only limit voltage and not boost them. So say for example your input voltage goes to 4.5 volts, your 7805 regulator will only supply 4.5V minus the 7805 voltage drop.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Well, I asked if voltage regulators could step down to a constant DC voltage from a variable DC input (say, example, 20-80V). I just want to know how I can do this. Does it vary from voltage regulator to another and is there some specific circuitry I have to follow?
 

thatoneguy

Joined Feb 19, 2009
6,359
How much current will you be using at the lower voltage?

If the swing goes up to 80V, you'll need a switching regulator. Even then, 80V is on the edge of some off the shelf ICs.
 

mecha

Joined Nov 10, 2012
18
"Well, I asked if voltage regulators could step down to a constant DC voltage from a variable DC input (say, example, 20-80V). I just want to know how I can do this. Does it vary from voltage regulator to another and is there some specific circuitry I have to follow?"

The 78 series (7805 etc) step down from max about 30 volts, and you must have some headroom, ie 1.2 volts (?) above output voltage at all times. They'll handle an amp or two, but you can slave up a transistor for higher loads. 78 series require two external capacitors. Stepping down from 80v is a more dodgy - I'd limit your maximum input voltage.

The 317, being variable, needs another component or two to handle variability. But power capacity is much the same as 78
 

spinnaker

Joined Oct 29, 2009
7,830
Well, I asked if voltage regulators could step down to a constant DC voltage from a variable DC input (say, example, 20-80V). I just want to know how I can do this. Does it vary from voltage regulator to another and is there some specific circuitry I have to follow?
No one is going to be able to help you if you continue to refuse to define your requirements.

The circuit is going to depend on your requirements.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
I don't know what my current requirements are, or I would've specified them Spinnaker. I was trying to bring down anything that is above 50V DC (and it can go up as high up to 100V DC) down to 20V constant DC. I didn't know you needed that other info regarding current and whatnot. I thought voltage regulators simply brought the voltage down from some DC value to another, and I had no idea you needed external circuitry. I thought you buy a simple LMXXXX IC that is suited for your application and simply stick it in to your breadboard. Sigh, our education system sucks.
 

atferrari

Joined Jan 6, 2004
4,763
If 80V is a LOT for common linear regulators (with very few exceptions, totally out of range), 100V is MUCH more. Being imprecise does not help.

What if instead of commenting you try to define your idea in short simple sentences and they could go from there. Is asking for a circuit, too much?

If you do not know the current, say at least what you intend to feed with it.
 

Evil Lurker

Joined Aug 25, 2011
116
For stepping down voltage any sort of significant amperage with say more than 10v difference between input voltage and output voltage its pretty much absolutely necessary to use a switching dc-dc converter of the buck (or sometimes called step down) variety. There are many integrated IC packages out there now, the one I'm working with at the moment is the LM2596 which can handle a peak switching current of 3A. Its pretty much about as simple as it gets... the datasheet tells you which inductor, capacitors, and all the other components you need to use for various input/output voltages at varying currents to get maximum efficiency and the lowest amount of ripple on your output.

If you need something that can take say, 12v and boost it up to 20v, or take 32 volts and reduce it down to 20v, and have the same power supply be able to do the job, then you need a SEPIC, or single ended primary inductor coupled. These are not exactly easy to design as generally they need to be high frequency to keep the ripple voltage from destroying your coupling capacitor nor are they the most efficient. And, just like with any high frequency circuit layout of the components and a ground plane are absolutely critical.

For just about any voltage about 35v and you are going to need a transformer based design.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Guys, you need to take it easy. This isn't a conventional problem I'm tackling, stop it.

My problem is I don't know what the hell the voltage is going to be because it varies with distance in my wireless power transfer setup. As I bring the receiver closer to the transmitter, the voltage increases and may burn out my load. However, when I take it away from the transmitter, the voltage goes down (to 0, if you take it far enough). When it's really close, it can go as far up as 2x the input voltage due to the flux linkages and hence the induced voltage changes with distance.

Now, I wanted to design this system such that when the receiver is brought close to the source, the voltage at the load (which is attached to the receiver) to be constant. When I take it far away, I designed the system to ignore that ( I didn't want to put a boost converter, so I designed it to work, within say transmitter + 1 meter away so I would only step down and not up. I've designed the system to give 10V at 1 meter. when I bring this thing closer to my transmitter this voltage climbs to 25V or so and can burn my load. I need a voltage regulator that takes any input above 10V (10-30 would be good) and drops it down to 10V constant. Of course, with different inputs I'm testing this system with, I get different voltage outputs and that's why it's all unknown to me as of now. I just wanted to know if I could simply pop in a voltage regulator indicated on its datasheet to take 10-30 and drop it to 10 or if I needed external circuitry alongside the regulator.
 

thatoneguy

Joined Feb 19, 2009
6,359
With all of the equipment and brainpower at MIT, when they made their resonant circuit, what did they use to indicate power was being transferred?

That might give you a hint.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
I'm assuming they just used measurements at the input and output and figured out how much power went through and how much of it lost in the transmitter circuitry.
 

thatoneguy

Joined Feb 19, 2009
6,359
Their load was an incandescent light bulb, which effectively emits all power to heat and light.

The one in the TED conference was a modified version, like a high power charging pad. They aren't in common use due to the inefficiency of the system.

Transmitting power over distances isn't a new and wonderful breakthrough. It has been around since Tesla. Radio waves are sending out electricity which is picked up by receivers hundreds of miles away.

There is no benefit to pushing a large amount of power over a very short distance with the losses involved. So we will have an electric grid far into the future, and that's even with the efficiency losses of the grid, which are rather high if measured (between generation, step up, transmission, step down, transmission, step down, etc.).
 
Top