Fixed Voltage Power Supply clic image quality
Wednesday, August 26, 2009
Clic image quality
clic image quality
clic image quality
clic image qualityvoltage from 0 to 15-volts,0 to 1-amperes.
I have done some work with DC PSUs over the past few months but some of the designs were either complex or "basic performers". I therefore decided that it was about right to produce a bench PSU that was simple enough to be made into another kit and have a reasonable performance too. The bench PSU specification had to:
- have variable voltage from 0 to 15-volts.
- have variable current limit from 0 to 1-amperes.
- have excellent voltage regulation with varying loads.
- be simple to construct and easy to understand.
- be expandable for higher output currents.
- employ commonly available components.
The design I came up with was almost a discrete component operation amplifier, with special attention to the zener diode voltage reference stability. This was achieved by feeding them with a constant current source. Here is the complete bench PSU circuit diagram: 
TR1 provides a constant current to a bank of three zener diodes, thus maintaining a constant voltage independant of supply voltage variations. The resistor Rx is used to sense the PSU output current and to switches off the current to the zener diodes should the output current become excessive. The 1M0 (logarithmic taper) pot causes the current sensor (TR2) to "kick-in early", so providing a variable current threshold limit. The value of RX should be (0.7/Amperes) ohms, where Amperes is the maximum output of the bench PSU. The prototype was set to 1-ampere, so Rx should be 0.5 / 1.000 = 0.5 ohms. The resistor should be rated at 1-watt, or 2x 1R0 500mW resistors in parallel for 1-ampere maximum output. I have used 4x 1R0 ohms in parallel (0.25 ohms) and have achieved 2-amperes (with a bigger heatsink).
The reference diodes provide a voltage reference of 10.7-volts with tappings at 4.7-volts and 7.7-volts. The 7.7-volt tapping is used as the reference voltage. In this way we have a 6-volt swing selected by the 50K voltage pot, above and below the 7.7-volt reference. This voltage swing is amplified by the operational amplifer formed by TR3, TR4, TR5 and TR6. The DC output of the Op-Amp is arranged to give a voltage gain of a little over 2, so the 4.7-volts to 10.7-volts reference will become 0 to 15.4-volts at the output. I had thought about using PCB mounting pots, but the finished PCB is quite large and the heatsink must be quite substantial (see below). Here is the finished PSU PCB. I must emphasise that this is a prototype and not the final PCB which is much better in the layout. 
A full-wave rectifier provides the 18-volts input from a 15-0-15 volt mains input transformer. The diodes 1N5401 are rated at 3-amperes. At first sight, it would appear that 1-Ampere diodes, such as 1N4001, would suffice, but this is not the case. Althought the averge current through these diodes is alittle over 500mA, the 2700uf input capacitor will be charged with short pulses, each considerably higher than 500mA. If the 2700uf capacitor were to be increased in value, then the rectifier diodes may need to be rated even higher. I used two parallel 47000uf in the prototype PSU, mainly because I wanted more output current so an additional regulator transistor is to be added. By the way, the supply smooting capacitors are not mounted on the board.Note that the TIP31 output transistor MUST be fitted with a heatsink.With minimum output voltage (say, 0.1vDC) and an 18.1v pre-regulator voltage, the TR6 TIP31 output transistor will be dissipating 18v x 1A = 18 watts. If we allow the device temperature to rise TO 120°C then we can allow an increase of 90°C over the ambient room temperature. 90°C / 18watts = 5°C per watt. A heatsink of 100 square cm is therefore required. If you wish to pull more current from this PSU then you will need a larger heatsink and change the value of Rx accordingly. Up to about 3-amperes is quite possible. TR6 may also be replaced with a darlington pair power transistor (such as TIP31 + 2N3055) if you want more than 3-Amperes, more attention will be required to other parts of the circuitry, such as the rectifier and reservoir capacitor.
The heatsink shown above is three pieces of thin aluminium sheet, 11cm x 3cm drilled with a 3mm hole dead centre. These are bolted between the PCB and the TIP31 transistor before any other components are mounted on the board. The fins can now be bent up to form six fins, each 5cm x 3cm or a total surface area of 180cm square. This is a little bigger than the calculation, but to over-engineer is not a fault (at least, not in my opinion). Here is the completed project.
The resistor Rx can also be used as a meter shunt for the output ammeter. I have used a 1mA meter with a series resistor selected to give a true current reading. Incidentally, take off the front scale of the meters, scan them and change the scale and lettering to suit your PSU. Then you can print the scale to paper and glue it over the existing meter scale. This gives a very professional finish. If you have an unsuitable scale ammeter, then open it up and cut the shunt wire between the two poles and then you have a small milli-ameter or micro-ameter that can be re-calibrated in this project.
I am presently updating the PCB foil pattern for this project, which I will include as soon as it is complete. I may perhaps offer this project in kit form, complete with heatsink, pots and all board-mounted components, but we will see what happens.
Power Supply Unit (PSU) for the workbench which can deliver 0-15 volts and up to 0-3 amperes
Here is a rather novel Power Supply Unit (PSU) for the workbench which can deliver 0-15 volts and up to 0-3 amperes. I have not shown any mains fuses as this I will leave to your own devices. You should as a minimum have a fuse in the primary and secondary of T1
clic image qulity
The circuit is a little cramped, but this is because I will also be posting the circuit to QRP@WW on the packet radio system. There are two resistors in the circuit marked "*see text" and these set the maximum output current limiter to a suitable value. Omit the 0R2 resistor and the PSU will deliver 100mA maximum. In this event the output TR8 can be almost any medium power transistor such as the 2N3053, BFY51 etc. It must be mounted on a heatsink no matter what output you need.
clic image qulityThe circuit is a little cramped, but this is because I will also be posting the circuit to QRP@WW on the packet radio system. There are two resistors in the circuit marked "*see text" and these set the maximum output current limiter to a suitable value. Omit the 0R2 resistor and the PSU will deliver 100mA maximum. In this event the output TR8 can be almost any medium power transistor such as the 2N3053, BFY51 etc. It must be mounted on a heatsink no matter what output you need.
The 6R8 resistor may be an ordinary 1/4 watt resistor and is used on all versions of the PSU. The additional 0R2 (0.2 ohms) sets the maximum current limiting to about 3 amperes. If you only want a 0-1 ampere PSU then use 0.68 ohms. T1 is a simple transformer suitably rated for the output power you want and having a 15 volt secondary. This should give over 21v of DC when rectified (measured at D1 cathode - ground). The anode of D7 should also have -21 volts with respect to ground.
If you have problems getting a 15 volt transformer then you can use a 12-volt transformer and add a few extra turns to the secondary. Most transformers have enough room for you to thread another 20 turns of wire through the former to put in series with the secondary winding. By sure to use the same thickness of wire as used for the factory made secondary.
D1, D2, D3, D4, D5, D6, D7 and D8 are all 1N4001 or better for up to 500mA but 1N5401 should be used if you want up to 3 amperes. D9 and D10 may be almost any small silicon diode such as 1N4148 etc. The 1K0 potentiometer in parallel with D9 is the current set pot. The 47K pot is the output voltage pot. With the values shown you should be able to get a range of 0v to a little over 15v.
HOW IT WORKS
TR4 and TR5 compare the ground (0v) with the wiper of the 47K pot which should also be 0v. If the voltages are not equal then TR5 controls the output darlington (TR6 and TR7) until the voltages become equal. If the 47K pot is set to maximum (top) then the output voltage of the PSU must be zero in order to get a balance. If the output voltage pot is set to the bottom then the voltage balance will only occur if there is +15 volts on the top of the 47K pot; the other end of the 15K being -5v.
In the event TR7 draws too much current then the voltage drop accross the 6R8 resistor will turn on TR1 which in turn turns on TR2. TR2 sinks the 0v reference to -20 volts. The TIP41 is incapable of delivering a negative output voltage so the output remains at 0v. The base of TR1 is "lifted" by up to 0.7v due to the current flowing through D10. This overcomes the TR1 base-emitter voltage drop needed to turn it on. You can adjust the 22K resistor a little so that the current control goes all the way down to 0mA and no further. 5K6 is about the absolute minimum resistance you should use.
The power supply itself is very simple and is nothing more than a full-wave bridge rectifier. The AC from the transformer is also capacitively coupled to yet another rectifier to provide the negative voltage required to allow the PSU to be controlled down to 0 volts. TR3 regulates this to stabilise the negative reference voltage.
POSSIBLE PROBLEMS
In the event of self-oscillation of the PSU then under no circumstances put any capacitance accross any signal in the control path. This will just make the problem worse. The cure for this would be to add capacitance between the base/emitter of the controlling transistors TR4 and TR5.
SUGGESTED MODIFICATION
One modification I have thought about is to replace the 22uf in the base of TR4 with a smaller value, say 10nf, and couple audio into the base of TR4 with a large electrolytic (1uf should do it). This will allow the output voltage of the PSU to be modulated with an audio signal. This could be a very interesting modification if you are "into" AM (MW?) broadcasting.
THE SIMPLEST PSU
One of the most useful circuits for the home constructor is the Power Supply Unit. A transformer with tapped secondaries can give a range of output voltages for testing and operating home-made equipment. 
This is a PSU in its simplest state. It is an ideal project for the beginner to homebrew. This circuit will deliver up to 3 amperes at 15 volts. The voltage may be reduced to 11.8 (ish) by selecting the 9v tapping. The voltage will of course fall a little when a heavy load is placed upon it. NO OUTPUT CURRENT LIMITING or voltage regulation has been provided. Shorting the output can become quite nasty so take care. The four diodes D1 - D4 may be replaced with a single bridge rectifier component.
The transformer tappings may be selected for any 3 volt increments from 3 volts (4.4v DC out approx) to 24 volts (34v DC out approx). Connect the bridge rectifier to the 9 and 12 volt tranformer taps for 3vDC output. 9 - 15 taps will give about 6.5 volts. 0 - 9 = 11.8 etc. The output voltage is equal to (Vt x 1.4) minus 1.4 volts, where Vt = the transformer tappings used. 
Here the circuit has been modified to include a very simple voltage regulator. This will give an output voltage between 0 volts and 15 volts. The power output transistor must be fitted to a heatsink. Again, NO OUTPUT CURRENT LIMITING is provided so care is still needed. All the component values not marked are the same as in the previous circuit.
The size of the heatsink can be quite complex but here is a guide.
The addition of a voltmeter accross the output terminals will be a distinct advantage. The DRAIN of TR1 is the correct point to insert an ammeter. The power FET chosen normally has the terminals marked (d, s + g). 
Here the circuit has been modified to provide current limiting to reduce the output voltage if too much current is demanded. This will reduct the possibility of damaging equipment by drawing excessive current. All component values not marked are the same as in the previous circuits. R1 = 0.7 / Amperes.
- R1 = 10 ohms for 60 mA
- R1 = 7 ohms for 100 mA
- R1 = 2.5 ohms for 300 mA
- R1 = 1 ohm for 700 mA
- R1 = 0.7 ohms for 1 A
- R1 = 0.25 ohms for 3 A
Have fun with this circuit. Best regards Harry.
Tuesday, August 18, 2009
RANGKAIAN REGULATOR POWER SUPPLY 20A
Quality Image 20A Regulated Power Supply Scheme Diagram
A heavy duty 13.8V power supply is a fine thing to have in the shack, but unless you acquire one secondhand, is an expensive little beastie to buy. This means building one should be considered, not only for the cost savings, but also because you can brag about it on air to your mates. Of course, careful consideration must be given to the properties of the completed supply, and after talking to a few of my friends who have built their own and fallen into all the traps, here are the printable ones : RF proof, easy to make, commonly available parts used, but above all CHEAP.
Well, last things first. Breaking down the construction costs of a heavy duty regulated supply, they are in order:
Friday, August 14, 2009
Subscribe to:
Posts (Atom)