LM7805, TO220 package

SD-50A-5

  SDM-30

PMA8811SF

UT70A

Various type of voltage regulator design

a) Zener diode voltage regulator.

Suitable only for very low power application.

b) 3 rectifier diodes as voltage regulator.

Suitable only for very low power application.

c) Linear voltage regulator.

Suitable for application that requires low noise.

d) Switching voltage regulator.

Suitable for application that requires high power.

Circuit diagram taken from,

A dc-dc regulator/converter or another name known as buck regulator or switching regulator, provides stable regulated output voltage to supply electronic circuits. Schematic, PCB layout and component list are available on this page.

LM2576 circuits perform same function as the commonly known voltage regulator LM7805 from National Semiconductor. The 7805 voltage regulator dissipates a lot heat. The higher input voltage, the more heat is generated. The extra input energy is converted to heat, keeping the output voltage regulated at 5V.

LM78XX series is available to regulate 5, 6, 8, 9, 12, 15, 18, 24V. If you want the output voltage adjustable, there is also a adj model. For -negative voltage supply, you can use LM79xx series. These regulator is able to support up to a maximum of 1A current rating.

LM7805 IC requires input voltage to be higher than output in order to regulate the output voltage. Input voltage needs to be at least 7V (up to a maximum of 20V) in order for LM7805 to regulate at an output of 5V. It is advisable to supply a voltage input range from 7.5V to 10V. Any higher input voltage is consider inefficiency, generating a lot of  heat.

A switching mode power supply such as LM2576 dc-dc converter, uses switching control to reduce the input dc voltage on average. This is equivalent to a lower input voltage resulting in minimum heat dissipated. The control results in better regulated output, less energy wasted through heat and the use for high current application. Nowadays dc-dc converter are getting smaller and comes in the TO-220 package too. You can simply change your LM7805 to dc-dc converter without any change in your design.

The first commerical module I tried is the SD-50A-5 from Meanwell rated at 5V 10A. It is very good and easy to use. However it is very big and bulky. If size is a constraint, you might consider the model SDM-30. It is able to handle up to 5V 5A and is a lot smaller than SD-50A-5. However it generates a lot of heat through its metal casing.

The best dc-dc I have tried before is PMA8811SF from Ericsson. It is by far the most compact (smaller than SDM-30) and most efficient dc-dc. Heat is also dissipated through it ceramic package, however it does not scalded your finger as much as SDM-30 do.  The IC package is surface mount however soldering is relatively easy because the IC leads are quite broad. It is rated at output 5Vdc 16A and generate far less heat. Each pieces cost about S$60, a lot more than the other converter model.

Through some research, I get to learn about commercial standard dc-dc IC that perform with only a few external components. The following article discuss on LM2576 IC with rating up to 5V 3A. LM2576 is one of the dc-dc IC product range from National Semiconductor. There are also various brand of dc-dc regulator IC available.

The interfacing of most dc-dc IC requires the use of inductor. This is the case for LM2576 too. Try sourcing your local electronics shop for one if possible. I am not stopping you to make your own inductor. Just that making your own inductor takes up time and it is very likely to cost you more than what a shop might be selling.

If you are interested in making your own coil, you might interested in this website, http://www.skylab.org/~chugga/mpegbox/coil/. The aurthor Jeff Mucha had demonstrated a simple and creative way to make inductive. One Long screw, 2 board flat washer, 2 nut, 1 ring spacer, glue, and XXX is all the tools that is require to make your own air core inductor. It is really interesting.

More article: home brew your own inductors

Jens Moller has contributed a program which generate a table of information for building air core inductor. Simply input the inductance value you need, the program will display a table containing the wire coil height radius and number of turns required. You need not have to understand formula to make your own inductor. Take a look at the following website, http://www.colomar.com/Shavano/inductor_info.html

A greenhorn when I first attempt to use inductor. It is a tough job building circuits using inductor. I do not have proper equipment to measure the inductor on hand. Never able to find out the inductance value I have. Fortunately, there is this inductance measurement product selling at an affordable price. UT70A from Uni-Trend Technology. It also function as a multi-meter, and can be used to measure voltage, current, etc... . Even with an inductance meter, it is not a easy task to measure inductance accurately.

Other reference:

The practical basic of building a power supply.

- The Power Supply.pdf

http://www.talkingelectronics.com/

projects/ThePowerSupply/Page79PowerSupplyP1.html

Photos of DC-DC circuit built

This is the 1st successful DC-DC circuit I built.

There a variety of capacitors out there in the market. Capacitance, voltage rating, dielectric material, etc... . Choose a suitable voltage rating across the capacitor. The circuits deals with high current, therefore it will be better to choose a low ESR (equivalent series resistance) Aluminum electrolytic capacitor. As a general guide, a higher voltage rating has lower ESR rating.

The inductor coil use should be able to handle the current passing through the inductor coil. If the wire is too thin, the coil may be burn or just fail. My previous circuit uses small wattage inductor (package like a big resistor). The circuit couldn't work and was later found to be IC problem. I have not yet do a test to check on the possibility of the inductor contributing to the failure.

Using a inductor meter to measure the inductance will be easier. Inductance value can be observe immediately for any modification to the coil of wire. The inductance value can also be calculated, depend on the coil size, number of turns, wire size used, dielectric of the core etc... .

The 1N5822 is a high current, high speed, schottky diode and is suitable for this digital switching circuit. Schottky diode (Schottky Barrier Rectifier), means that the forward voltage drop is low. For this application, a low forward voltage diode is necessary.

Schematics

PCB Bottom Layer (PCB trace)

Component Layout (Silkscreen)

Bill of Material (BOM) for LM2576 circuit

Failure, my first prototype circuit to test out the performance of LM2575, LM2576.

Some of the various sizes of inductor tested and seems to be working with LM2576.

Initially I thought that I had use the wrong type of inductor, resulting in the circuit malfunction. Initially I had used a smaller type of inductor (looks like a resistor). Realizing that this circuit drive high current load, I should use a thicker inductor coil. That's why I modified the circuit with an inductor (enamelled wire, wound around the ferrite core).

Still it doesn't work. I guess that both IC LM2575, LM2576 must have been damage by my previous attempt. The capacitor used is suspected because the datasheet call for low ESR capacitor. It is very difficult to find these in the local shops, therefore I use a normal capacitor instead.

One day, I visited a shop selling ready made inductors and brought LM2576 at the same time. The circuit was rebuild and it finally works. My deduction at that time was either the inductor or the capacitor is giving me the problem. After further testing, I find out that ordinary capacitor works as well. There is hardly any difference in performance. Various type of inductor were tested (except the resistor like type). All inductor works too, big or small. Quite weird actually, and I couldn't figure it out the actual problem I had in my previous attempt.

The mystery is resolve finally. One fine day I went back to the shop where I first purchase my LM2576 and brought 2 additional LM2576 for more testing. A new circuit was build and the familiar failure was observed. The output voltage of 5V cannot be sustain and eventually drop when more than 1A of current is draw by the load. The lab power supply display a current loading limit warning. IC becomes very hot. The datasheet specify that LM2576 should be able to supply 3A without any  problem. Both brand new IC are tested to have the same problem.

This is weird, as the same inductor and capacitor previously tested do not result in this same old problem. However the circuit shows the same failure symptom. The next thing that comes to mind, is the IC. The IC LM2576 from the previous working circuit is then transfer over the new circuit board for testing. Everything works fine. It is then clear that the problem comes from the IC itself.

Checking up on the previous IC, I notice that they are from the same manufacturing batch number and believe that they are already damage in some way.

Sample of the 5Ω 50W aluminum house resistor used for testing 1A current performance.

General tester for LM series dc-dc IC chip.

Some of the newly fabricated board built to support other prototype projects. It has been tested to support a RF transceiver operating at 5V without any issue observed.

Home fabricated circuit board

It has been some time since I learn to use LM2576. The circuitry is able to handle a higher current at 3A 5V output. This translate to a higher cost and circuit size, since all component must be able to handle that high power capacity. These component include the LM2576, inductor and the diode. Since most electronics kit requires less than 1A power supply, it is wise learning how to apply a low power dc-dc regulator like LM2575. Cost can be reduce by 50%.

There is one day that I happen to come across this IC LM2575 while searching high and low for LM2576. LM2576 is actually quite difficult to find. There is only 2 shop I know of, but I have rule out one shop because they are selling a faulty batch of LM2576 IC. LM2575 seems very common from shops around and I decided to find out more about this chip. Indeed it is what I have been looking for, a low power regulator. So I purchase the IC and its component to try it out. When I started writing this acticle, only did I realize that I have actually tried it about 6 months ago. The experiment was forgotten after a series of failure.

The following experiment is done during the 1st test on LM2575 circuit. The experiment compare between the performance of using different inductor. One using a wire coil inductor, and the other smaller one inductor that looks like a resistor with it's color bands..

The photos on the left column shows the LM2575 circuit using the correct inductance value at 330uH but the inductor is low power rated. It is small and looks like a color coded resistor.

A few second after the left circuit is powered up, the small inductor turns very hot. The waveform observed at the output of the dc-dc regulator, contains a high amount of noise/ripple energy.

The photo on the right column shows the same circuit using a slightly higher inductance at 480uH but the coil is thicker and bigger in size.

The circuit using a high power rating inductor on the right shows a cleaner DC supply, although the inductance value is different from the design. There is still ripple at it's output but I guess it will be minimum using an inductance value of 330uH with higher power rating. Too bad, I do not have the right inductor to experiment further. It is either coil one myself or buy one from shop.

12 June 2006, Lim Siong Boon

I have found this article regarding about the property of inductor Isat (current saturation) and Irms (continuous current). They are usually one of the important specification to take note while selecting inductor from datasheet.

Current saturation means the amount of current required that flow through the inductor, in order to reduce the inductance of the component.

Continuous current means the amount of current required to heat up the inductor to a certain temperature. If the amount current continue to flow through the inductor, the inductor is basically becoming a heater. The temperature depends on the amount of current flowing through it.

The following contains information that I learn from.

Isat_Irms explain.pdf

02 Dec 2008, Lim Siong Boon

LM2575 Schematic taken from National Semiconductor LM2575 datasheet

Bill of Material (BOM) for LM2575 circuit

Switching Power IC

LM2575, LM2576, LM2596, LM2678

The following table provides a quick reference for power supply circuit. The circuit schematic and component list are selected from the manufacturer's datasheet.

For exact component value design, you need to the datasheet. The following component value is design for typical input voltage of 12Vdc or 24Vdc drawing power at 75% of the current rating.

DC to DC step down voltage regulator.

Wide input voltage 8Vdc to 40Vdc.

Part number:

- LM2575-3.3 (3.3Vdc output)

- LM2575-5.0 (5Vdc output)

- LM2575-12 (12Vdc output)

- LM2575-15 (15Vdc output)

- LM2575-ADJ (1.23Vdc to 37Vdc output)

LM2575 datasheet

Click for LM2575-adj circuit

Component list

- 100uF 50V electrolytic capacitor

- 1N5819 high speed schottky diode (1A)

- 330uF 100V electrolytic capacitor (higher voltage, lower ESR)

- wire coil inductor,

          330uH, 1A (for LM2575-3.3, LM2575-5.0)

          680uH, 1A  (for LM2575-12, LM2575-15)

- "for LM2575-adj IC" 5kΩ multi-turn variable resistor, set to ratio to R1=1.25kΩ, R2=3.75kΩ for voltage output of 5Vdc before soldering.

please refer to the table for resistors in parallel for more resistance design options.

    Vout, R1 & R2 design selection calculator

    Design calculator might not work on some web browser.

DC to DC step down voltage regulator.

DC to DC step down voltage regulator.

Wide input voltage 8Vdc to 40Vdc.

Part number:

- LM2576-3.3 (3.3Vdc output)

- LM2576-5.0 (5Vdc output)

- LM2576-12 (12Vdc output)

- LM2576-15 (15Vdc output)

- LM2576-ADJ (1.23Vdc to 37Vdc output)

tested working on 2007-06-26

tested working on 2007-06-26

LM2576 datasheet

Click for LM2576-5.0 layout

Click for LM2576-adj circuit

Click for LM2576-adj layout

Component list

- 100uF 50V electrolytic capacitor

- 1N5822 high speed schottky diode (3A)

- 1000uF 100V electrolytic capacitor (higher voltage, lower ESR)

- wire coil inductor,

          100uH, 3A  (for LM2576-3.3, LM2576-5.0)

          220uH, 3A  (for LM2576-12, LM2576-15)

- "for LM2576-adj IC" 5kΩ multi-turn variable resistor, set to ratio to R1=1.25kΩ, R2=3.75kΩ for voltage output of 5Vdc before soldering.

    please refer to the table for resistors in parallel for more resistance design options.

    please refer to above for design calculator for resistance value selective

DC to DC step up voltage regulator.

Wide input voltage 3.5Vdc to 40Vdc.

Part number:

- LM2577-12 (12Vdc output)

- LM2577-15 (15Vdc output)

- LM2577-ADJ (1.23Vdc to 37Vdc output)

tested working on 2006

tested working on 2007-06-21

tested working on 2007-06-21

LM2577 datasheet

Click for LM2577-adj circuit

Click for LM2577-adj layout

Component list

- 2.2kΩ 1/4W resistor

- 0.1uF capacitor

- 0.33uF capacitor

- 680uF 50V electrolytic capacitor

- 1N5822 high speed schottky diode (3A)

- wire coil inductor, 100uH (3A)

- "for LM2577-adj IC" 20kΩ multi-turn variable resistor, set to ratio to R2=2kΩ, R1=18kΩ for voltage output of 12Vdc before soldering.

DC to DC step down voltage regulator.

Wide input voltage 8Vdc to 40Vdc.

Part number:

- LM2596-3.3 (3.3Vdc output)

- LM2596-5.0 (5Vdc output)

- LM2596-12 (12Vdc output)

- LM2596-ADJ (1.23Vdc to 37Vdc output)

LM2596 datasheet

Component list

- 680uF 50V electrolytic capacitor

- 1N5824 high speed schottky diode (5A)

- Output 100V electrolytic capacitor (higher voltage, lower ESR)

          330uF (for LM2596-3.3, LM2596-5.0)

          180uF (for LM2596-12)

- wire coil inductor,

          33uH, 3A  (for LM2596-3.3, LM2596-5.0)

          68uH, 3A  (for LM2596-12)

DC to DC step down voltage regulator.

Wide input voltage 8Vdc to 40Vdc.

Part number:

- LM2678-3.3 (3.3Vdc output)

- LM2678-5.0 (5Vdc output)

- LM2678-12 (12Vdc output)

- LM2678-ADJ (1.2Vdc to 37Vdc output)

LM2678 datasheet

Component list

- 45uF 50V electrolytic capacitor

- 0.47uF

- 6TQ045S high speed schottky diode

- Output 100V electrolytic capacitor (higher voltage, lower ESR)

          360uF (for LM2678-3.3, LM2678-5.0)

          220uF (for LM2678-12)

- 0.01uF

- wire coil inductor,

          15uH, 5A  (for LM2678-3.3, LM2678-5.0)

          22uH, 5A  (for LM2678-12)

Schottky diode (1A)

1N5817, 1N5818, 1N5819, MBR120P, MBR130P, MBR140P, MBR150, MBR160, SR102, SR103, SR104, SR105, SR106, 11DQ03, 11DQ04, 11DQ05, 11DQ06

(smd alternative to 1N5819) MBRS140T3G, SS12, SS13, SS14, SK12, SK13, SK14

Schottky diode (3A)

1N5820, 1N5821, 1N5822, MBR320, MBR330, MBR340, MBR350, MBR360, SR302, SR303, SR304, SR305, SR306, 31DQ03, 31DQ04, 31DQ05, 31DQ06

(smd alternative to 1N5820, 1N5821, 1N5822) MBRS320T3, MBRS330T3, MBRS340T3, SS32, SS33, SS34, SK32, SK33, SK34

Schottky diode (4A-6A)

1N5823, 1N5824, 1N5825, 50WQ03, 50WQ04, 50WQ05, 50WR06, 50SQ060, MBR340

Resistor Colour Codes

Images taken Farnell.

W series- Vitreous enamelled wirewound resistors offering high power, high stability and reliability. Suit for use in harsh environment.

WH series- Aluminium clad resistors for applications where high power dissipation in a small space is required.

MFR series- High stability metal film resistors offering higher performance than carbon film with very low noise levels and high reliablility.

RC series- Very high stability metal film resistors offering very high reliability and tight tolerances.

WCR series- Surface mount resistors suitable for automatic placement. Features include nickel barriers, wide ohmic range and high reliability.

The DC-DC converter design for the adjustable IC version, you may need the following resistor standard EIA decade resistor values for references. Long time ago, when technology is not so advance, resistor manufacturing is not unable to produce precise resistor value, as in today. Due to its large variation in tolerance, the resolution of the range of standard resistor value is limited. Example is E3 series having tolerance of 50%, which have only resistors in decade of 100, 220, 470. There is not much point to define or differential between 100Ω and 101Ω, having a tolerance of 50%. With such high tolerance, there is hardly any difference between 100Ω and 101Ω. They should both belongs to the same class of 100Ω

The standard EIA decade resistor value is group into different series. Each is grouped according to their tolerance level. The higher the tolerance, the higher the resistor value resolution will be. The common resistor value range would be the E24 (tolerance 5%) and E96 (tolerance 1%) series.

To find the range of resistor value that is available in the industrial, multiply the normalise standard found in the table in terms of 100, 1000

- Example: E24 series referring to normalise value 1.0

   It means that under E24 series, you should be able to find these Ω range 100Ω, 1000Ω, 1kΩ, 10kΩ, 100kΩ, 1MΩ, 10MΩ, 100MΩ. Other resistor value under E24 can be determine from the rest of the normalised value in the table below. Lower Ω are not available in the series as they should be in resistor package for higher wattage

Standard EIA Decade Resistor Values

E24 (preferred standard resistor values with 5% tolerance)

E96 (preferred standard resistor values with 1% tolerance)

Tolerance Codes

B=0.1%, C=0.25%, D=0.5%, F=1%, G=2%, J=5%, K=10%, M=20%

website references:

- http://sound.westhost.com/miscc.htm

- http://www.logwell.com/tech/components/resistor_values.html

Table for resistor in parallel

This resistor table is interesting. While dealing with circuits prototype, I often need to use resistor value that may not be common. To keep sufficient stock for all resistor range is a bit too much to manage. A larger and better storage system will be needed. I find it difficult to manage the wide range of resistor. This brings me the idea of forming the required resistance from two commonly stocked resistor connecting in parallel. This means that I can keep fewer resistance range and easily stock larger quantity for each value.

On the following table, the 1st row and column represents the common resistor value that I normally keep stock. The rest of the cells present the various possible resistance I can obtain by having the resistance in parallel from the respective row and column. The computation is done in the microsoft excel sheet. formula: "=($A2*B$1)/($A2+B$1)". Those value highlighted in yellow are quite useful when designing my adjustable DC-DC circuit when I do not have the stock for the resistor that is not commonly in use.

High efficient ac-dc conversion IC

- isolated (smaller transformer component)

- non isolated (transformerless), LNK306DN

- isolated (smaller transformer component), VIPer12A

Ultra-Miniature High Voltage Power Supplies

Q Series

Q01-5 (5Vdc to 100Vdc)