Pure Sine Wave Inverter Circuit Using IC 4047

A very effective pure sine wave inverter circuit can be made using the IC 4047 and a couple IC 555 together with a few other passive components. I have explained the details below.

The Circuit Concept

In the previous post I explained the main specifications and datasheet of the IC 4047 where I have explained how the IC could be configured into a simple inverter circuit without involving any external oscillator circuit.

In this article we carry on the design a little ahead and learn how it can be enhanced into a pure sine wave inverter circuit using a couple of additional ICs 555 along with the existing IC 4047.

The IC 4047 section remains basically the same and is configured in its normal free running multivibrator mode with its output extended with the mosfet/transformer stage for the required 12V to the AC mains conversion.

How the IC 4047 Functions

The IC 4047 generates the usual square waves to the connected mosfets creating a mains output at the secondary of the transformer which is also in the form of square wave AC.

The integration of the two 555 IC to the above stage completely transforms the output into a pure sine wave AC. The following explanation reveals the secret behind the IC555 functioning for the above.

Referring to the below shown IC 4047 pure sine wave inverer circuit (designed by me), we can see two identical IC 555 stages, wherein the left section functions as a current controlled sawtooth generator while the right hand side section as a current controlled PWM generator.

The triggering of both the 555 ICs are derived from the oscillator output readily available across pin#13 of IC 4047. This frequency would be 100Hz if the inverter is intended for 50Hz operations, and 120Hz for 60Hz applications.

Using IC 555 for the PWM Generation

The left 555 section generates a constant sawtooth wave across its capacitor which is fed to the modulating input of the IC2 555 where this sawtooth signal is compared with the high frequency signal from pin3 of IC1 555 creating the required pure sine wave equivalent PWM at pin#3 of  555 IC2.

The above PWM is directly applied to the gates of the mosfets. so that the square pulses here generated through pin10/11 of IC4047 gets chopped and "carved" as per the applied PWMs.

The resulting output to the transformer also causes a pure sine wave to be stepped up at the mains AC secondary output of the transformer.

The formula for calculating R1, C1 is given in this article which also tells us about the pinout details of the IC 4047

For the NE555 stage C may be selected near 1uF and R as 1K.

Assumed output waveform

More info on how to use IC 555 for generating PWM

An RMS adjustment could be added to the above design by introducing a pot voltage divider network across pin5 and the triangle source input, as shown below, the design also includes buffer transistors for improving mosfet behavior

The above pure sine wave inverter design was successfully tested by Mr. Arun Dev, who is one of the avid readers of this blog and an intense electronic hobbyist. The following images sent by him prove his efforts for the same.

More Feedback

Inspiring response received from Mr. Arun regarding the above IC 4047 inverter results:

After completing this circuit, the result was amazing. I got full wattage by the 100 W bulb. Couldn't believe my eyes.

The only difference i had made in this design was replacing the 180 K in the second 555 with a 220 K pot to adjust the frequencies accurately.

This time the result was fruitful in all respects... On adjusting the pot, i could get a non disturbing non flickering full wattage glow in the bulb, also the 230/15 V transformer connected as the load gave a frequency in between 50 and 60 ( say 52 Hz ).

The pot was adjusted gently to get a high frequency ( say 2 Khz ) output from pin#3 of second ic 555. The CD4047 section better calibrated to get 52 Hz at the two output terminals....

Also I am facing a simple problem. I have used IRF3205 mosfets at the output stage. I forgot to connect the safety diodes across the drain terminals of each mosfets...

So when I had tried connecting an another load ( say table fan ) in parallel to the given load ( 100 W bulb ), the glow of the bulb also the speed of the fan was reduced a little and one of the MOSFET was blown due to the absence of the diode.

The above 4047 sine wave inverter circuit was also tried successfully by Mr. Daniel Adusie (biannz), who is a regular visitor of this blog, and a hardworking electronic enthusiast. Here are the images sent by him verifying the results:

Sawtooth Waveform Oscilloscope Output

Illuminating a 100 Watt Test Bulb

The following images show the modified waveforms at the output of the transformer as captured by Mr. Daniel Adusie after connecting a 0.22uF/400V capacitor and a suitable load.

The waveforms are somewhat trapezoidal and are far better than a square wave which clearly shows the impressive effects of the PWM processing created by the IC555 stages.

The waveforms could be probably even further smoothened by adding an inductor along with the capacitor.

Showing an near Sinewave Oscilloscope Trace  after PWM Filtration

Interesting feedback received from Mr. Johnson Isaac who is one of the dedicated readers of this blog:

Good day
In your post, Pure Sine Wave Inverter using 4047, in the second I.c stage (ic.1) you used 100 ohms resistor in between pin 7 and 6.,
Is that correct? I use to think an astable multivibrator using 555 pin configuration should have the 100 ohms between pin 7 and 6. Also, the 180k variable between pin 8(+) and pin 7. Pls check the pin connection and correct me pls. Because it oscillate sometimes and it doesn't sometimes also. Thanks,
Isaac Johnson

Solving the Circuit Issue:

In my opinion, for a better response you can try connecting an additional 1k resistor across the 100 ohm outer end and pin6/2 of IC1

Johnson:

Thank you very much for your response. I actually constructed the inverter you gave in your blog and it worked.

Though I don't have an oscilloscope to observe the output waveform BUT I bet readers its a good one cos it operated a fluorescent tube lamp in which any modified or pwm inverter can't power on.

See the picture sir. But my challenge now is when I add load, the output flickers sometimes. But am happy its a sine wave.

Video Test Waveform

Simplifying the above Concept

The above explained IC 4047 pure sine wave inverter can be actually simplified greatly by eliminating two IC 555 configurations from the design.

The basic idea is to produce a synchronized sine PWM output for chopping the MOSFET switching so that this SPWM waveform is replicated at the output of the transformer and is transformed into a pure sine waveform. The complete circuit diagram can be seen in the following diagram:

In this simplified design, we yet again exploit the pin#13 square wave output from the IC 4047 and convert it into triangle waves. This triangle wave is then used to modulate the PWM output of a IC 555 astable to create the required synchronized SPWM for chopping the MOSFET gates.

The capacitors Cx and Cy will need to be experimented a bit to create a proper triangle waves from the RC network output.

Remember, you will need an oscilloscope to check and confirm the various waveform discussed in the above diagram.

Calculating the Cx and Cy Values for the Square wave to Sine wave Converter stage using RC Integrator

A 100 Hz frequency from pin#13 of the IC can only generate a crude sine wave using the shown double passive RC integrator circuit. We can make things easier by choosing the cutoff frequency fc to be pretty close to the fundamental frequency of the square wave which is fo and that happens to be 100 Hz. By doing this, we can reduce the higher harmonics a bit while still keeping that fundamental frequency intact. However if you want to get a decent sine wave, then remember we need to make sure that the input frequency is at least 300 Hz.

Formula:

The RC time constant relationship is:

fc = 1 / (2 * π * R * C)

R * C = 1 / (2 * π * fc)

For fc = 100 Hz:

R * C = 1 / (2 * π * 100) = 1592 * 10^-6 seconds

Choosing Component Values:

First Stage (R1, Cx):

Choose R1 = 10 kOhms:

Cx = 1592 * 10^-6 / (10 * 10³) = 159.2 nF

Second Stage (R2, Cy):

Similarly, for the second stage:

R2 * Cy = 1592 * 10^-6 seconds

Choose R2 = 10 k Ohms:

Cy = 1592 * 10^-6 / (10 * 10³) = 159.2 nF

Final Values for a Crude Sine Wave:

• R1 = 10 kOhms
• Cx = 159 nF (rounded to the nearest standard value)
• R2 = 10 kOhms
• Cy = 159 nF

Expected Output:

With these values our circuit will do a moderate job of cutting down those harmonics, and what we will end up with is a rough sine wave that has quite a bit of distortion. If you want to get things just right, you can tweak the values of the resistors or capacitors a little bit to fine-tune that waveform.

Another Simpler Looking Options

The following concept discuses a rather simpler method of modifying an ordinary square wave inverter using IC 4047 into a sine wave inverter through PWM technology. The idea was requested by Mr. Philip

Technical Specifications

I hope that i am not going to be a bother, but I need some advice with a PWM-controlled modified sine wave inverter I am designing so I want to seek your expert opinion.

This simple design is tentative, I haven't implemented it yet but I would like you to take a look at it and tell me what you think.

Also I want you to help answer some questions which I have not been able to find answers to.

I have taken the liberty of attaching an image of a quasi-block diagram of my tentative design for your consideration.

Please help me out. In the diagram, the IC CD4047 in the inverter is responsible for generating square wave pulses at 50Hz which will be used to alternately switch on MOSFETS Q1 and Q2.

The PWM circuit will be based on IC NE555 and its output will be applied to the gate of Q3 so that Q3 will provide the PWM. Besides this, I have two questions.

First, can I use square waves for the PWM pulses? Second, what is the relationship between PWM frequency and supply frequency? What PWM frequency should I use for a 50Hz inverter output?

I hope that this design is feasible, I think it is feasible, but I want your expert opinion before I commit scarce resources to implement the design.

Looking forward to hearing from you sir!

Sincerely, Philip

Solving the Circuit Request

The configuration shown in the second figure above would work only if the center tap mosfet drain/source pins are swapped and N-channel mosfets are used for the 4047 inverter oscillator output, as shown below:

The PWM section should be built as I have explained in this article:

The PWM transforms the flat square waves into a modified square wave by chopping them into smaller calculated sections such that the overall RMS of the waveform becomes as close as possible to an actual sine counterpart, yet maintaining the peak level equal to the actual square wave input. The concept may be learned in details here:

However the above transformation does not help to eliminate the harmonics.

The PWM frequency will be always in the form of chopped square waves.

The PWM frequency is immaterial and may be of any high value, preferably in kHz.

How to Add Feedback Voltage Control to a 4047 IC Inverter

In this section we will discuss how to add an automatic feedback voltage control to a 4047 IC based inverter circuit.

If you don't want to read the whole explanation, you can watch the following video instead:

This type of feedback network is intended to ensure that the AC output voltage from the inverter never exceeds a predetermined safe value. For example, using this automatic feedback control, a 220 Volt inverter circuit output could be restricted to maximum 230V AC level, which may be safe for most 220 Volts appliances.

For a 120 Volt inverter system, this feedback network could be adjusted to restrict the output AC to a maximum of 115 Volts AC, which may be safe for most 110 Volt appliances.

How the Feedback Circuit Works

Now let's understand how the simple feedback network is designed to work, by referring to the following circuit diagram:

The mains AC voltage from the inverter transformer secondary is first converted to a DC voltage using a bridge rectifier.

This bridge rectifier is built using 4 numbers of 1N4148 diodes.

The DC 220 Volts output from the bridge rectifier is applied to an adjustable voltage divider network built using a 220 k resistor and a 100 k preset or trimpot.

The 100 k preset is carefully adjusted such that its central wiper terminal generates around 1.2 volts in response to an input AC voltage which is considered high, and needs to be restricted.

Let's say, the input AC needs to be restricted at around 270 Volts, then the 100 k preset may be adjusted so that its wiper terminal generates a proportional DC voltage level of 1.2 Volts.

We select a 1.2 Volt output because this is the equivalent DC level required to just switch ON the attached BJT device.

That means, whenever the inverter output AC voltage tries to exceed the 270 Volts mark, it causes the BJT to turn ON and become conductive.

When the BJT conducts, it starts grounding the CT pin of the IC 4047.

Since the RC network associated with the CT and the RT pinouts of the IC determines the output frequency and PWM, the grounding of the CT pin instantly causes the output PWM and frequency to nullify, which in turn causes the output AC voltage to dip.

When this happens, the feedback voltage to the BJT also drops proportionately until the BJT is turned off. In this situation the IC output PWM is restored back to its previous form, and if the output voltage continues to rise, it yet again activates the BJT via the feedback network, and this feedback on off cycle keeps repeating rapidly, ensuring that the output AC is not able to rise above the set threshold, which is 270 volts in our example.

This concludes our feedback control circuit explanation, now let's try to understand how this feedback circuit can be setup prior to integrating it with the inverter circuit.

How to Setup

The next diagram below shows the isolated feedback circuit, which can be set up by implementing the following steps, before interfacing it with your inverter circuit:

Initially keep the 100k preset's wiper arm fully towards the ground side of the DC supply.

Now, switch ON the 12 DC supply and then switch ON the AC supply to the bridge rectifier whose voltage level is around the desired high cut-off limit.

Next, slowly adjust the preset using a well insulated screw driver, until the LED just begins to illuminate.

To confirm the results, try varying the input AC voltage around the cut-off threshold and watch the LED turn on and off at the thresholds. Although the cut-off may not be too sharp, yet it will do the job of the intended output voltage control of the inverter quite well.

That's it! This simple procedure should be enough to complete the setting up of the feedback circuit.

After this, make sure to seal the preset adjustment with some super glue.

So, this concludes our explanation on a feedback circuit for the 4047 inverter.

Pseudo-sinusoidal IC 4047 Inverter Circuit

From a continuous source, battery, or vehicle alternator, it is possible to produce a 50 Hz voltage/pseudo-sinusoidal waveform, using this simple IC 4047 inverter circuit. The design is capable of operating many low-power devices that require a voltage of 220 V AC, typically supplied by the AC mains grid distribution network. The power of this module depends largely on the chosen transformer and, above all, on the continuous source's ability to supply the circuit without failure.

Working Principle of the circuit

Being able to power a device on the 220 V AC network when no mains AC power is available can be very convenient. We particularly think of campers, caravanners, fishermen, and motorists.

Although this simple setup does not provide a 100% clean sinusoidal wave, it can still satisfy many undemanding devices such as incandescent or fluorescent lighting, soldering irons, small motors, etc.

The power can reach up to 250 VA if the continuous source is strong enough and an adequate toroidal power transformer is chosen.

The power transistors are rated for 30 A, provided that a substantial heatsink is used.

We even suggest adding a small fan directly connected to the 220V output, which will prevent it from operating without a load.

Analysis of the Circuit Diagram

To obtain an alternating voltage from a DC source, the simplest solution is to use an inverted transformer: the low-voltage secondary winding has a center tap.

It is then sufficient to alternately supply power to the two windings to produce a symmetrical, but non-sinusoidal, variable voltage on the primary side if the low-voltage secondary winding is driven by a rectangular signal with steep edges.

Two complementary and opposite signals are required, and the CMOS integrated circuit with the reference 4047 is used to generate them, capable of either building an astable or monostable flip-flop.

The proposed diagram is shown in the figure below.

Of course, we will operate our circuit as an astable generator, ensuring a precise frequency of 50 Hz. This task is entrusted to components C1 and P1. Some pins of the IC1 circuit are connected to the ground or the positive supply, depending on our needs.

For more information, you will find an explanation in the appendix about the multiple possibilities of this practical little integrated circuit.

The symmetrical square voltages available at pins Q and Qdash (corresponding to terminals 10 and 11 of the circuit) are routed through resistors R1 and R2 to an adjustable element. Its role is to balance the amplitude of the two signals and facilitate the adjustment of the control module.

Note that the duty cycle will be exactly 50% by design. The sliders of elements P2 and P3 control the base of a 2N1711 transistor, which is responsible for driving the downstream power semiconductors.

Components R7, R9, C4, as well as R8, R16, and C5, form a coarse filter capable of smoothing out the signal angles to some extent.

This will allow us to create a "pseudo-sinusoidal" signal that will control the intermediate transistors T3 and T4 mounted on the small control board.

The BDX 77 models are rated for an emitter-collector voltage of 80V and a maximum peak current of 8A. They are delivered in a TO-220 package and will benefit from a heatsink.

The secondary windings of the transformer do have a common point connected to the positive pole of the power supply.

The other end of the coils will receive the ground through the power transistors T5 and T6, a model not much larger than the famous 2N3055 but with a maximum peak current of around 30A in this case.

It is the NPN bipolar model with the reference BUX 39.

Diodes D1 and D2 absorb the unavoidable breakdown current produced at each interruption of the inductive element, which is the transformer.

The primary of the transformer, a toroidal model in this case, delivers a 50 Hz AC voltage.

The component labeled Gemov (or varistor) is placed between the output pins to clamp excessive voltage surges that may occur.

Two wires can also be tapped from the output to operate a fan directly powered by 220V, which will provide airflow to cool the transistor-heatsink assembly.

It is important to ensure that the voltage of the continuous source does not drop too much or be completely depleted in the case of a lead-acid battery.

It goes without saying that the input DC current can reach considerable values if, for example, a power of 200VA is desired. It would be prudent to include a robust series fuse, similar to those found in a household installation.

Construction

The creation of two printed circuit boards is necessary: the first one, shown in the figure below, is designed for the components described in the control module.

The layout is not very dense, but it is still recommended to use pre-sensitized copper-clad boards for reproduction.

After etching and rinsing, various drillings are carried out. Care must be taken to carefully mount transistors T3 and T4 on a small heatsink, which is directly attached to the epoxy board.

From this board, six wires will go to the significantly larger power module, and the track layout at a 1:1 scale can be found in the figure below.

It should be noted that certain tracks on this simple IC 4047 inverter circuit board will carry high currents and require special treatment. We recommend soldering bare copper lengths, with a section of approximately 1.5 mm², directly onto these tracks.

For example, electrician's copper wire with the insulation removed can be used. Care should also be taken to ensure all connections between the two boards, and in particular, to avoid reversing the wires.

The transformer has six wires, with the two thinnest wires corresponding to the primary, and therefore our 220V output.

It is important to properly identify the other windings before soldering them permanently. This assembly can be housed in a sturdy, adequately ventilated metal enclosure.

How to Setup

We leave it to you to complete this final step for this simple IC 4047 inverter circuit. After a thorough check, the frequency can be adjusted to 50 Hz under no-load conditions using the adjustable preset P1, if you have an oscilloscope or frequency meter.

Otherwise, simply set this element to the midpoint position. The other two adjustable components are set to obtain a symmetrical signal in amplitude.

Initial tests can be performed using a simple socket where different power lamps can be connected. For resistive or slightly inductive loads, there should be no serious issues.

Proceed with caution when testing devices that typically require a true sinusoidal voltage.