Do-it-yourself blocking generator for an LED on one transistor: a self-powered circuit. Low-voltage voltage converters for LEDs LED flashlight on one transistor circuit

A word of caution: White LEDs are comparatively expensive, so I suggest including a small resistor (1 to 10 ohms) in series with the LED cathode to limit and measure the peak current. While testing the circuit, you can measure the voltage drop across this resistor using either an oscilloscope or a peak detector to see if the peak current is greater than the value recommended by the LED manufacturer. Based on these recommendations, for greater reliability, we will try to obtain a peak current no higher than half of the maximum.

Review

A compact switching converter that can provide enough voltage to power white LEDs consists of a minimum number of parts. The lamp we will receive is much more efficient in terms of lumen hours per pound of battery weight than an incandescent lamp. In addition, the color of the glow is determined by the emission of the LED phosphor, so the color of the glow practically does not change, even when the battery is completely discharged. As a result, the battery lasts a long time. This one is cheap and suitable for use in flashlights, emergency lighting and other applications that require white LEDs to be powered from one or two primary batteries.

Scheme

There couldn't be a simpler scheme than this. The blocking oscillator consists of a transistor, a 1 kOhm resistor and an inductor. When the power button is pressed, the transistor is turned on by current flowing through the 1 kΩ resistor. The voltage that appears across the inductance section from the midpoint to the collector of the transistor induces a voltage across the 1 kΩ resistor, which can be even higher than the battery voltage, thereby providing positive feedback. If there is voltage between the coil tap and the collector of the transistor, the collector current constantly increases. Due to positive feedback, the transistor remains in saturation until something happens to its base current.

At some point, the voltage drop across the inductance section from its midpoint to the transistor's collector approaches the value of the battery voltage (in fact, the battery voltage minus the transistor's collector-emitter saturation voltage). From this point on, the voltage is no longer induced in the coil from the tap to the 1 kΩ resistor, and the voltage at the base begins to decrease and becomes negative, thus accelerating the turn-off of the transistor. Although the transistor is now turned off, the inductor remains a source of current and the collector voltage rises.

The collector voltage quickly becomes high enough to generate current in the LED, and it flows until the inductance is discharged. The collector voltage then begins to “ring”, swinging from ground to power, turning on the transistor and starting another cycle.

Inductance

If you are designing this circuit for a non-commercial application, you have a wide range of inductor design options. The size of the core, its permeability and saturation characteristic (physical dimensions, µ and Bs) determine how many ampere-turns it can provide before saturation. If the core saturates faster than the voltage drop across the inductance section from the tap to the collector of the transistor reaches the battery voltage, the circuit will switch immediately anyway, because the saturation of the core makes the coil like a resistor and there is an inductive coupling between the collector and base (the side with the 1 kΩ resistor) the halves of the coil fall very strongly. This has the same effect as bringing the voltage drop across the coil closer to the battery voltage. The wire gauge determines how many amps the circuit produces before switching due to increasing voltage drop. The parameters of the inductor core (mainly physical dimensions and magnetic permeability) determine how many microseconds the coil is charged by the collector current, which will increase until the transistor turns off. These parameters also determine how long current will flow through the LED while the transistor is turned off. Almost all characteristics of the inductor affect the operation of this circuit.

I made this circuit using ferrite rings a few millimeters in diameter and toroidal cores with a cross-section up to a few centimeters (note the inductance on a rusty nail described below).

Here is, in general, the relationship between core sizes and inductor characteristics:

• Large core: easy to wind, low switching frequency, increased power.
• Small core: difficult to wind, higher switching frequency, lower power.

How to start. Take a coil core, preferably ferrite, and wind 20 turns on it. Make a tap in the form of a short loop of wire, then continue winding another 20 turns. An increase in the number of turns leads to a decrease in the operating frequency, a decrease leads to an increase in frequency. I wound only 10 turns with a tap from the middle (5+5) and this coil operated at a frequency of 200 kHz. Look at the circuit described below, assembled in the base of a light bulb, operating at a frequency of about 200 kHz.

Improved circuit

This scheme is attractive because it contains a minimum number of elements. The LED is powered by pulsed current. The pulse begins when the voltage across the LED reaches its direct operating voltage, which is higher than the battery voltage, which does not affect the switching of the transistor. The disadvantage is that the ratio of peak current to average LED current is quite high, it can be 3:1 or 5:1, depending on the circuit parameters (mainly coil inductance and battery voltage). If you want the LED to be brighter for a given peak current, you can add the diode and capacitor shown in the diagram below.

One critic suggested a good idea: if there is space available, add a decoupling capacitor between the negative terminal of the battery and the midpoint of the inductor. Some batteries have high output impedance, and this capacitor can increase the circuit's output current. A 10 uF capacitor should be sufficient, but if you are using a very high inductance inductor, it is better to increase the capacitance.

Where will you place the power source?

Since this circuit contains few elements, I made all of them, including an inductor, a 1K resistor, a 2N4401 transistor (in a TO-92 package, by the way), a rectifier diode, a chip capacitor and a Nichia NSPW315BS LED along with a small drop of glue place at the base of the pen lamp.

Using an LED instead of a light bulb allows you to develop a compact flashlight. It provides enough light to walk down the street on a moonless night. I estimated the operating time of the flashlight, which consumes a current of about 35 mA from a 1.5 V battery. It turned out that it would work continuously for at least 30 hours. It's quite long. Specifications for several Duracell alkaline batteries can be found.

The color of the glow remains consistently bluish-white, even when the battery voltage drops. If such a device is treated well, it will last a very long time. I had one of these flashlights, assembled according to the last diagram shown, for 18 months, and I used it every night. I only replaced the battery twice. If the contacts on the battery hadn't deteriorated due to corrosion, I wouldn't have known it was time to replace it, because the flashlight worked great.

Night light of a rusty nail

These blocking oscillator circuits work better with ferrite cores, but they can sometimes be hard to find. Some readers have expressed concern about the manufacture of inductors, which is understandable since inductors have an aura of mystery to many.

I undertake to prove that there is nothing complicated about inductors, and that they are very important. One day, waiting for a tow truck due to a car breakdown, I noticed a rusty nail near the road. It was 6.5 cm long and I decided to use it for the inductor core.

I pulled a twisted pair of ø0.5mm solid copper wire from a long CAT-5 (Ethernet) cable. This wire is similar to that used to install telephone lines inside buildings. I wound 60 turns of twisted pair in about three layers on a nail, then connected the beginning of one conductor to the end of another conductor, making a 120-turn inductor tapped from the middle.

I connected a 2N2222 transistor, a 1 kOhm resistor, a 1.5 V AA battery and a white LED to it. Nothing happened. Then I applied a 0.0027 uF capacitor to a 1 kOhm resistor (it was on the desktop) and the LED came to life. You may need a capacitor of about 0.001uF. The LED glows beautifully and the circuit draws 20mA of current from the AA battery. The signal on the oscilloscope screen looks terrible, but the main thing is that the circuit excited even on this rusty nail, and increased the initial 1.5 V of the AA element to more than 3 V, sufficient to glow the LED.

Those familiar with some aspects of coil core selection will immediately notice that eddy currents will be enormous, since iron has low resistance compared to ferrite, or air for example, and that there will probably be other losses. And the point is not that you should run out and buy nails to make an LED lamp, but that this circuit turned out to be very workable. If a rusty nail and some telephone wire is enough to light up a white LED, then the inductor is not a problem. So, take a break, go and buy a ferrite core and start working on the project.

Where to get ferrite cores

Wolfgang Driehaus from Germany wrote that ferrite cores are used in compact fluorescent lamps, and that he has successfully used them in LED power circuits. The next day I looked up and saw that some of the lights needed replacing.

Some of the CFLs in my house have burned out. After buying new lamps and replacing the burned out ones, I went to the garage to disassemble one of the lamps. The first problem was getting to the electronics in the lamp base. In a subsequent letter, Wolfgang told me that the lamp bulb can be opened and the circuit board removed without damaging the glass. Be careful not to break the glass tubes of the lamp, as they contain toxic mercury.

I wanted to make sure that these cores would be useful to me and removed the windings from the dumbbell and toroidal coil. During the process of disassembling the coil on the EE core, the ferrite cracked in several places, so I was not able to test it in my circuit.

I wound 50 turns of ø0.2 mm enameled wire onto the dumbbell core, made a central tap, and then wound another 50 turns. I assembled a device from this coil, a 2N4401 transistor, a 330 Ohm resistor connected to the base of the transistor, and a white LED in accordance with the diagram given at the beginning of the article. When I connected the 1.5V power supply, the LED flashed brightly. This confirmed that a coil with such a core can be used in this circuit.

I wound 10 turns of ø0.4 mm wire onto the toroidal core, made a tap and wound another 10 turns. Having connected the coil to the same circuit (2N4401, 330 Ohm, white LED) with a 1.5-volt power supply, I saw that the LED was lit, although not as brightly as with the previous coil, but after all, only 20 turns were wound on the toroid.

So now we know where to get ferrite cores. Compact fluorescent light bulbs are very affordable and will eventually break down and require replacement.

Another reader noted that another source of ferrite cores is computer peripheral cables. Monitor cables, keyboard cables, and some USB cables have plastic thickenings that actually contain ferrite cores. If you're going to throw your old keyboard in the trash, why not cut off the ferrite first?

Read the ending

A blocking generator is a signal generator with deep transformer feedback that generates short-term (usually about 1 μs) electrical pulses repeated at relatively large intervals. They are used in radio engineering and in pulse technology devices. Performed using one transistor or one lamp. (wikipedia)

I decided to build an LED flashlight that would shine for a very long time and be economical. The blocking generator allows power from low voltage. LED, for example, 5 mm LED with a current of 20-50 mA.
The plans were to use germanium low-power transistors of the MP37 brand, an LED strip, AAA pinky batteries, and a miniature case.
As a body, I took a paint marker, it was planned to build in batteries, a blocking generator, stick an LED strip and stuff it all into a headphone package - a plastic flask.

First, I cleaned the paint marker from paint with solvent and wiped it with a napkin. Then I cut out a compartment for 3 AAA batteries in the bottom, cut out contacts from tin and secured them from the bottom, inside the marker with hot-melt adhesive so that they were insulated from the metal of the marker. For the upper contacts, I cut a washer from thin PCB and glued the contacts onto it with 2-sided tape. The batteries are connected in series.

The aluminum flask was torn, so I had to seal it with F64 flux.

P.S. I have some another flashlights and if you would like, i can show to you my work.

CONVERTER FOR LED

Incandescent lamps have been replaced by LEDs, which in many cases successfully replace them. But due to the nonlinear current-voltage characteristic, various voltage converters are used to power lighting LEDs from a battery. As you know, an LED is powered by a voltage of at least 2 V, and depending on the type, up to 3.5 V. In addition, at least a simple current stabilizer is needed, because as the battery capacity decreases, the brightness of the LED also decreases. Therefore, a simple power resistor, from a battery with an increased voltage, will work worse than a converter. Below are diagrams of simple converters that can be assembled even by beginners.

The circuit is powered by one AA battery and is a blocking generator. Increased voltage pulses appear on the collector, are rectified by a Schottky diode and charge the capacitor. Transformer T1 is wound manually on a ring core. To do this, take a ferrite ring K10x6x4 and wind two windings of 20 turns with PEL 0.3 wire. In general, the number of turns can be 6:10, 10:10, or 10:15. For the best efficiency and brightness, they must be selected experimentally. Everything that is available is used for the frame.

The circuit uses a low-dropout transistor to achieve maximum efficiency. The output current can be adjusted by resistor R1.

Next we see a somewhat complicated scheme with more stable generation. Current consumption 15 mA. The voltage converter is also made according to the circuit of a single-cycle generator with inductive feedback on a transistor and transformer. The winding data is the same.

The next modernization of this converter was a circuit from a Chinese LED flashlight:

Here and in other circuits, a Schottky diode with a low voltage drop is used as a diode (after all, every half a volt counts). The diodes used are IN5817, 1GWJ43, 1SS319, or, as a last resort, the Soviet D311. These diodes can be taken from the power controller board of a non-working lithium-ion battery from a mobile phone.The following converter circuits are made on two transistors and are characterized by an increased output current - up to 25 mA. A correctly assembled converter does not need adjustment unless the transformer windings are reversed; otherwise, swap them.

The transformer used is similar, but the number of turns in the windings is 40. The transistors cost C2458 and C3279. Thanks to the feedback on the C2458 transistor, simple stabilization of the current and, accordingly, the brightness of the LED is obtained.

Another version of the converter with two transistors:

There is no need to wind the transformer here, since a ready-made inductor of 300 - 1000 μH is used.

The last converter circuit was also copied from a Chinese LED lamp and works great when assembled.

The first turn on of a correctly assembled device must be carried out in test mode, in which power from the battery is supplied through a 10 Ohm resistor, so that the transistors do not burn out if the transformer terminals are connected incorrectly. If the LED does not light, it is necessary to swap the terminals of the primary or secondary winding of the transformer. If this does not help, check the serviceability of all elements and installation.

From personal experience I can note that in all the above circuits, domestic transistors KT315 - KT3102 are often successfully launched. The number of transformer windings should be selected for maximum brightness and efficiency. Ready-made “everything that came to hand” from various equipment was used as chokes. It is not recommended to install the cheapest (0.1 W) 5 mm LEDs. It’s better to pay extra and buy a 10 mm LED for 0.5 euros. The brightness will increase significantly. Even better results will be achieved after installing special

Lyrical introduction

This article will discuss the modernization of a flashlight using the example of a device from the well-known Philips company. So, what disadvantages might it have? Like all pocket flashlights, this device was observed to have a significant decrease in the brightness of the incandescent lamp when the batteries were drained. And naturally, low efficiency and service life. Nevertheless, there is a solution to these eternal problems.

LEDs! But will it be enough to replace only the light source? No. Most flashlights use the now classic circuit, in which two 1.5-volt batteries are connected in series. But a voltage of 3 volts is not enough for the LED to glow brightly, therefore, it is worth including a converter in the circuit. The converter has a more stable output current when the input can be 0.5 V or less. What happens to a flashlight if its batteries are discharged to such a limit? That's right, it doesn't work. Therefore, the converter is the most successful move in solving this problem.

A new problem arises: where to place it? After all, there is often no space in the flashlight body. If you have open-frame components, you can mark them directly in the lamp base, but what if not? My article will help you figure this out.

Circuit design

As I said, there is a solution. Quite an original solution, I think.

Consider the converter circuit:

The diagram shows a blocking generator. Excitation is achieved by transformer coupling on transformer T1. The voltage pulses arising in the right (according to the circuit) winding are added to the voltage of the power source and are supplied to the LED VD1. Of course, it would be possible to eliminate the capacitor and resistor in the base circuit of the transistor, but then failure of VT1 and VD1 is possible when using branded batteries with low internal resistance. The resistor sets the operating mode of the transistor, and the capacitor passes the RF component.

The circuit used a KT315 transistor (as the cheapest) and a super-bright LED (as the brightest). Let's talk about the transformer separately. To make it, you will need a ferrite ring (approximate size 10x6x3 and permeability of about 1000 HH). Wire diameter is about 0.2 mm. Two coils of 20 turns each are wound on the ring. If you don’t have a ring, you can use a cylinder of similar volume and material. You just have to wind 60-100 turns for each of the coils. An important point: you need to wind the coils in different directions. At worst, you can use a nail, but a large nail, and about 150 turns are required for one coil. In addition, the efficiency of a nail is much lower than that of ferrite.

Let's move on to practice now.

Practice

Consider a photograph of a flashlight. This is necessary to understand the meaning of my research. There is nothing futuristic here, I will only note that the switch is located in the “fountain pen” button, and the gray cylinder is metal and conducts current.

So, step one. We create the “body” of the device.

We make a cylinder according to the standard size of the battery. For example, the size of the batteries in my flashlight is AAA. It can be made from paper (like I did), or you can use a piece of any rigid tube. For gluing we use “rubber” glue, as it is a good dielectric.

We make holes along the edges of the cylinder, wrap it with tinned conductor, and pass the ends of the wire into the holes. We fix both ends, but leave a piece of conductor at one end so that we can connect the converter to the spiral. (The nut shown in the figure is not needed yet)

Now let's start assembling the converter itself. I didn’t have a ferrite ring (and it wouldn’t fit into the flashlight), so I used a cylinder made of a similar material.

The cylinder was removed from an inductor from an old TV. The first coil is carefully wound onto it. The coils are held together with glue. I got about 60 turns. Then the second one swings in the opposite direction. I got 60 or so again; I definitely didn’t count it - I couldn’t wind it neatly. Secure the edges with glue. Let's dry it. The coil can be slightly warmed up during the drying process. I placed it on a piece of paper on the shade of the table lamp. Let it dry. And we move on.

We assemble the converter according to the diagram:

Everything is located as in the figure: transistor, capacitor, resistor, etc. Passive and active elements have been assembled, we solder the spiral on the cylinder, the coil. The current in the coil windings must go in different directions! That is, if you wound all the windings in one direction, then swap the leads of one of them, otherwise generation will not occur.

We are happy because we got the following:

We insert everything inside, and use nuts as side plugs and contacts.

We solder the coil leads to one of the nuts, and the VT1 emitter to the other. Glue it. We mark the conclusions: where we have the output from the coils we put “-”, where the output from the transistor with the coil we put “+” (so that everything is like in a battery).

All. You get something similar to what is shown in the previous figure.

Now you need to make a “lampodiode”. We take a regular base from a used light bulb, and...

One point: there must be a minus LED on the base. Otherwise nothing will work.

There was another solution to the problem. Of course, you can directly create a converter module with an LED in one package. In this case, as you have probably already noticed, you only need two contacts. You can do it this way. But in this solution, the LEDs cannot be easily changed. Why change? It’s very simple, because you can use an ultraviolet LED to check the authenticity of banknotes and much more. In addition, I believe that my way of solving the problem is more ergonomic and interesting.

Assembly technique

As is clear from the figure, the converter is a “substitute” for the second battery. But unlike it, it has three points of contact: with the plus of the battery, with the plus of the LED, and the common body (through the spiral). However, its location in the battery compartment is specific: it must be in contact with the positive of the LED. To put it simply, the assembly sequence in the picture cannot be changed. Otherwise, as you may have guessed, the device will not work.

Upgraded flashlight in action:

This flashlight is more economical, ergonomic and, due to the absence of a second battery, lightweight. And the main advantage! All parts can be found in the trash!

List of radioelements