Link to what, the power supply article? Look at the front of this thread. It is where I am working on it. I write the articles within the threads/posts because it lets me work on the article from any work station.

If that is not what you are asking, then link to what?

Before my title changed to SuperModerator it was E-Book Developer. My goal is to have the book finished in my lifetime.

There is always room for more writers, it will get finished much sooner.

 

I started reading the first post and noticed a number of typos. Then I noticed the date of the post. So I assumed that there must be a revised version somewhere.

 

Nope. Correct away! It may be changed later as I add, delete, or modify the text. Not all the remote workstations have good online spellcheckers I can install, so stuff creeps in.

I've sat on this article hoping the muse would strike, so far nothing. I am good at putting words on paper, but I am not the sharpest pencil in the cup (not the dullest either). So I invite anyone to have a say or critique. So far no trolls have struck, some folks think if you ask for help you don't have a clue what you're talking about.

 

Here are some minor corrections for now, hilighted in RED or $$$

(I'm allowed only six smilies and images - so some of my tags are $$$)
(I have removed the images)

Status: Work In Progress. If you see something that is not right feel free to comment, but as a work in progress it may be I haven't corrected something that I will be getting to. If you have a suggestion please make it.

There are three major kinds of power supplies: unregulated (also called brute force), linear regulated, and switching. All of them have various advantages and disadvantages, which will be covered in a case by case basis. Much of the material has been covered in depth, this section will show how it fits together as a whole.

$$$ (grammar, either and or start new sentence).


Unregulated

An unregulated power supply is the most rudimentary type, consisting of a transformer, rectifier, and low-pass filter. These power supplies typically exhibit a lot of ripple voltage (i.e. rapidly-varying instability) and other AC "noise" superimposed on the DC power. If the input voltage varies, the output voltage will vary by a proportional amount. The advantage of an unregulated supply is that its cheap, simple, and efficient.

The ripple can be very small. It depends on the filter and the load.

Changed to "This type of power supply can exhibit". The point is this particular type of supply can have excessive ripple.

Power supplies as a class try to emulate batteries. They have some unique characteristics such as ripple, but some of the basic concepts, such as internal resistance, still apply.

Perhaps need to clarify the difference between a diode and a rectifier.

Not sure what you mean, will add clarification as to type of diodes. Note the referral to the chapter in the AAC book at end of paragraph.

Diodes are one constant in almost all power supplies, at least those that take AC from line voltage and convert it into DC. The drop around 0.6V at minimum currents, but as the current goes up so does the voltage drop. A 1N4001 power diode will drop around 1.1V at 1Amp, its maximum rating. This drop must be accounted for whenever designing a power supply. All of the following configurations have been covered in detail in Chapter 3, Diodes and Rectifiers. If you have a question about the following circuits you may need to review that section.

Changed to "The drop for a silicon diode is around 0.6V"

While it is possible to make a power supply without a transformer, it is extremely hazardous, the nature of which can not be understated. Only a transformer provides decent isolation from the mains. There are very few ways a power supply without a transformer can be made to operate safely around people.

proper Need more input, sentence looks OK to me.

A transformer will drop the voltages of a power supply to much safer levels, as well as isolating the user from the AC power. Power transformers are not treated the same as impedance matching transformers, though the theory of operation is much the same. Transformers are usually rated in VA of output, or as AC volts out and a current rating. The VA rating determines what the maximum current transferred through the transformer, You can calculate the fuse or breaker required by dividing the primary voltage by the VA rating. So if a transformer is a 120VAC on the primary side, and the VA rating is 120VA, then you would use a 1 amp fuse.

$$$ strike out much Changed to "operation is very similar" as there are slight differences in how it is taught.

Note that transformers in general will produce their rated voltage at their rated current. Low current models tend to change more than high current versions, since a higher percentage of loading will be used. With no load you can have substantially higher voltages out. Since it is partly a quality issue, a perfect transformer would not do this, it can be hard to find a specification on it, so the user has to be aware of the fact.

Impedance matching transformers have to handle a wide range of frequencies, while a power transformer is optimized for only one frequency. Many cases a 50Hz transformer can be used for a 60Hz, but there will usually be a small detrimental change in performance, in efficiency and heat produced. Transformers are one of the few man made devices that can exceed 95% conversion efficiencies.

Another bit of math common to all power supplies is line voltage is given in RMS values, while power supplies will translate this into peak values. The conversion for this is 1.414 (square root of 2). So the output of a power supply will be the output of the transformer times 1.414 minus the diode drop(s).


Basic Power Supply Configurations

Different configurations of power supplies will have dramatically different characteristics. The following design uses only one half of a cycle. It's Effective DC voltage (defined as measured by a meter without filtering) is around 31.8% of the AC output voltage. This design, while simple, wastes much of the power possible from the transformer as well as allowing excessive ripple due to fact that there is no power being delivered to the load 50% of the time, and tries to use the capacitor to make up for the deficit. This circuit is referred to as a Half Wave Rectifier power supply.


......................................................Figure 1

Note that the fuse is the first thing connecting the power supply from line current. If it blows the entire circuit will go dead. The switch is optional, but recommended. These are standard features for any power supply.

handled simply in several ways Used as is.

The excessive ripple can be handled pretty simply several ways. Increasing the capacitance works, but large value capacitors tend to be physically large. Large capacitors also appear as a dead short for the transformers/diodes while they charge, this is called surge current. In extreme cases it can damage or burn out the diodes.

$$$ charge resulting in high charge current. Disagree, for now will leave as it, but will allow other members to input. I have always heard it called surge current.

If the transformer has a center tap then it can be used for full wave rectification as shown. The effective DC voltage for the circuit below is around 63.6%, but since only one half of the transformer the voltage will be half of the transformers AC output. This circuit is referred to as a Full Wave Center Tap power supply.


......................................................Figure 2


The full power of the transformer can be squeezed out of it by using the following full wave bridge rectifier. Its effective DC power rating is around 90%, but it uses the full AC voltage of the transformer. This circuit is referred to as a Full Wave Bridge power supply.


......................................................Figure 3


There are many circuits out there that need a ± power supply, notably op amps. With a simple center tap on the transformer you can generate a decent power supply for that application. This circuit is referred to as a Dual Complimentary Rectifier power supply.


......................................................Figure 4


Ripple

Ripple is a serious problem with all these designs. 50 Hz is slightly worse than 60 Hz, since the lower frequency requires a larger capacitor to compensate, but ripple should be planned and calculated for any design you make. It is a problem in RC discharge curves. Even if you are planning on using a regulator to clean up the output (which will be covered later) you must get the ripple low enough for the regulator to do its job. The greater the load the larger the capacitor must be to compensate, since the discharge of the capacitor is that much faster.

While ripple is very undesirable, it is also unavoidable. The first thing you have to define is the peak to peak value of ripple you are willing to accept. In audio and RF it can cause hum, where the line frequency can be heard riding on the signal. In digital circuits it can cause false triggers of gates if it is bad enough, though in general digital circuits are less sensitive ripple since they are a go/no go system. It is important to figure out how much ripple can be lived with and still allow the circuit being powered to function correctly.

sensitive to ripple Used change as is.

A capicitor on a power supply with no load will do something odd. It will charge up to the peak value, and stay there. Ripple will be almost none existent, since there is no load resistance to discharge the capacitor between charge cycles. It is the combination of the line frequency, load resistance, and capacitance that ultimately define how much ripple is on the power supply.

The math for calculating ripple can be quite complex, so over time it has been broken down to several rules of thumb. In the beginning it was done with tables and charts, but math is probably an easier tool to use for the job, understanding that it is somewhat simplified.

<snip>

 

My class of Electronics #1 used the Sedra/Smith book to calculate inrush diode currents and ripple voltages depending on the load. I think I 'll come up with some formulas tomorrow.

 

I have already done this in a previous post. Send me your work and we'll compare notes.

 

Thank you. Most of the changes are implemented, I have added my commentary to your posts in green (moderator privileges). It managed to put too many characters in the post, so I sniped off a little text to make it fit.

 

Sorry, my typo. I meant to say:
charge resulting in high surge current.


$$$ charge resulting in high charge current. Disagree, for now will leave as it, but will allow other members to input. I have always heard it called surge current.

 

Added the wall warts section today. Think a picture of a standard wall wart would be appropriate?

 

I have already done this in a previous post. Send me your work and we'll compare notes.

Okay, as promised to Bill and MrChips:

Legend:
\(V_r\): Ripple voltage
\(r_d\): Diode internal resistance
\(V_{D0}\): Diode threshold voltage drop
\(V_s\): Source voltage
\(PIV\): Peak Invert Voltage on the diode
\(I_p\): Peak diode current
\(I_L\): Mean load current
\(V_o\): Output voltage
\(V_p\): Peak source voltage
\(R_L\): Load resistance

For half rectification without filter capacitor:
\(V_o=\frac{R_L}{R_L+r_d}(V_s-V_{D0})\simeq V_s-V_{D0}\\
\bar{V_o}=\frac1{\pi}V_s-\frac{V_{D0}}{2}\\
I_p=\frac{V_s-V_{D0}}2\)

With filter capacitor and for \(RC>>\frac1{\text{Line Frequency}}\):
\(V_o\simeq V_p-\frac{V_r}2\\
V_r\simeq \frac{V_p}{f R_L C}
\text{Diode conduction time } \Delta t=\frac{\sqrt{2 \frac{V_r}{V_p}}}{2 \pi f}\\
\text{Conduction current}
i_{DAv}=I_L \left(1+\pi \sqrt{\frac{2 V_p}{V_r}} \right)
i_{DMax}=I_L \left(1+2 \pi \sqrt{\frac{2 V_p}{V_r}} \right)\)

Full/Bridge rectification with split transformer without filter capacitor:
\(PIV=2V_s-V_{D0}\\
I_p=\frac{V_s-V_{D0}}R_L\\
\bar{V_o}=\frac2\pi V_s-V_{D0}\)

With filter capacitor:
\(f'=2f\\
V_r=\frac{V_p}{f' R C}\\
I_{DAv}=I_L \left(1+\pi \sqrt{\frac{V_p}{2 V_r}} \right)\\
I_{DMax}=I_L \left(1+2 \pi \sqrt{\frac{V_p}{2 V_r}} \right)\)

Proofreading Needed!

 

I am currently in need how to calculate the ripple for set capacitance. I have a half built capacitance box I'll be working on as an experimental tool to verify some of the math.

There are some ideas being thrown around in this discussion which might help.

 

To Geo:

Thank you. I don't understand all that I see, but I'll work on it. Ideally it can be simplified a little, where we put in the 50 or 60 Hertz, the load, the capacitance, the transformer VA, and get some sense of the expected ripple. If it can not be simplified then so be it. It gives me something to work with either way.

 

Don't forget this useful formula that calculates input capacitor value:

ΔE = 0.7 I / C / F

Where:
ΔE = peak to peak ripple voltage
I = DC load current
C = input filter capacitor in farads
F = ripple frequency (full-wave = 120 for 60hZ line)

The assumption is that the capacitor is powering the load 70% of the time--the remaining 30% of the time, the transformer /rectifier is both charging the capacitor and powering the load.

Note that % ripple is a relatively meaningless figure.

 

To Geo:

Thank you. I don't understand all that I see, but I'll work on it. Ideally it can be simplified a little, where we put in the 50 or 60 Hertz, the load, the capacitance, the transformer VA, and get some sense of the expected ripple. If it can not be simplified then so be it. It gives me something to work with either way.

Do you need any help with a particular formula? It's been a couple of years since I took those note, but the book I took them from is very explanatory and I probably can re-extract them without much trouble.

 

Nice work Geo. I'm somewhat busy right now but I will look it over.

Bill, Wikipedia has a nice article:

http://en.wikipedia.org/wiki/Ripple_(electrical)

 

Don't give me too much credit, guys. I just copied my sheet of notes, after reading the 3rd Chapter of Sedra/Smith Microelectronic Circuits 3rd edition.
The formulas were either proven in the book or a result of comprehension exercises.

 

Good to see this topic on the burner again, Bill.

One item that hasn't been dealt with at all so far, but is quite important is the one mentioned in post #22 by Norfindel. When a particular transformer is used to make a power supply with a capacitor input filter, since the current drawn from the transformer is in shorter pulses than half sine waves, the transformer heats up more. The manufacturer's ratings are almost always for a resistive load drawing AC current from the transformer. With a rectifier circuit, you can't get rated power. At the very least, a rule of thumb derating factor should be applied. I've seen a factor of 1.6 mentioned, but it depends on circuit details.

Also, an exact ripple formula would be, as you say, complicated. It would depend on properties of the transformer, diodes and filter capacitor that may be beyond the capability of a hobby builder to measure.

I have mentioned in another thread the effect of the transformer impedance on ripple. This effect can make big difference when comparing the performance of a transformer wound concentrically versus split bobbin construction. The split bobbin construction give much greater leakage inductance, and this construction method is becoming more common due to its better safety properties.

 

OK, I'm working on the Switching regulator section. I believe this was written by Tony. My problem is I'm not sure I buy into the description at all.

This reminds me of the PWM article I wrote, I had to educate myself as I wrote it.

************************************************************

OK, I'm going to do a slash and burn on the Switching Power Supply, so I'm keeping a record of the chapter here...

Switching

A switching regulated power supply ("switcher") is an effort to realize the advantages of both brute force and linear regulated designs (small, efficient, and cheap, but also "clean," stable output voltage). Switching power supplies work on the principle of rectifying the incoming AC power line voltage into DC, re-converting it into high-frequency square-wave AC through transistors operated as on/off switches, stepping that AC voltage up or down by using a lightweight transformer, then rectifying the transformer's AC output into DC and filtering for final output. Voltage regulation is achieved by altering the "duty cycle" of the DC-to-AC inversion on the transformer's primary side. In addition to lighter weight because of a smaller transformer core, switchers have another tremendous advantage over the prior two designs: this type of power supply can be made so totally independent of the input voltage that it can work on any electric power system in the world; these are called "universal" power supplies.

The downside of switchers is that they are more complex, and due to their operation they tend to generate a lot of high-frequency AC "noise" on the power line. Most switchers also have significant ripple voltage on their outputs. With the cheaper types, this noise and ripple can be as bad as for an unregulated power supply; such low-end switchers aren't worthless, because they still provide a stable average output voltage, and there's the "universal" input capability.

Expensive switchers are ripple-free and have noise nearly as low as for some a linear types; these switchers tend to be as expensive as linear supplies. The reason to use an expensive switcher instead of a good linear is if you need universal power system compatibility or high efficiency. High efficiency, light weight, and small size are the reasons switching power supplies are almost universally used for powering digital computer circuitry.

 

I would clarify that the transformer can be made much smaller because it's much more efficient at the higher frequencies that the SMPS works, and that the design is highly efficient because the transistors always work either in saturation or cut-off, so either voltage or current is always near zero, an so power dissipation is minimal on them.

Also, there are several kinds of switching power supplies. You could have a step-up supply powered by an AA battery for a LED lantern. They don't need to be specifically powered by direct AC rectification (also, it's safer for the hobbist to start with battery-powered, low-power stuff).

 

Will you be covering the same power supply internally grounded. That seems to be a big subject with audio circuits only, as well as choice of enclosure as I research help with ripple and will follow this topic